Click Here for Presentation

1,247 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,247
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
19
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • We will discuss some highlights of DB2 V9.1 for z/OS (V9). The beta announcement was May 2, 2006. The beta program began in June 2006. It will become generally available when we know that the release is ready for customers, and we expect that to be soon. In these highlights, we’ll talk briefly about the new function for pureXML and SQL. We’ll note the improvements for regulatory compliance, in performance and in availability. XML work across the DB2 family is a much larger step than ever before. While V7 and V8 removed many differences from DB2 for Linux, UNIX & Windows, V9 SQL takes the next big step to improved productivity and consistency. Performance improves for utilities, optimization and LOBs. Data definition on demand extends the theme of online schema evolution from V8. Additional Unicode and text enhancements continue the work from V7 and V8. Utility enhancements help with new function, more LOB and XML support, better performance and improved availability, removing the BUILD2 step from online reorg. V9 enhances DB2's ability to handle new and enterprise applications. V9 improves with XML, large objects, and many SQL and security improvements. V9 builds upon and extends DB2 traditional strengths and the ground-breaking V8 in many areas: online schema evolution, Unicode, XML, DB2 family SQL, utilities, security and 64-bit virtual storage.
  • The primary messages for our customers today are about migration. V7 has been generally available for six years now, and will be out of service as of June 2008, so V7 and prior customers need to migrate now to stay in service. The formal end of service announcement has been made. V7 customers need to move to V8 as soon as possible. V8 has been generally available for three years. Tools are ready from all vendors for V8. IBM tools were ready for New Function Mode at General Availability for V8 and will be ready for V9. As customers move to V8, they can start some of the V9 planning. V9 migration is only from V8 NFM, and V9 will be generally available very soon. IBM Information Management tools have been running with DB2 for the past three years. Some changes have been made to DB2 V8 in the service stream, but they are fairly small in number. The small number has helped to keep V8 quality and stability high. There is a lot of information in redbooks, in the library and on the web.
  • Cross loader with LOBs V7 V8: PQ90263 Built in functions ASCII, TIMESTAMPDIFF V8: PQ95795 DSN1COPY with 1000 OBIDs V7 V8: PK05758 QMF with multirow fetch PQ99482 Online Check Index V8: PQ92749 & PQ96956 z/OS 1.7 up to 7257 extents V7 V8 PK07590 PK10594 LOAD, UNLOAD with LOBs V7 V8: PK22910, PK27125, II13767 (LOB service) zIIP PK18454, PK19920, PK20487, PK19921, PK26901, PK27578, PK30087, PK30474, PK30548 ; Info APARs II14219 for zIIP, II12836 for parallelism ; DB2PM/DB2PE PK27712 ; Omegamon PK25395, OA15900, OA15899, OA15898 http://www.ibm.com/systems/z/ziip/ http://www.ibm.com/servers/eserver/zseries/zos/downloads/#ziip DB2 for z/OS Performance Topics SG24-6465 http://www.redbooks.ibm.com/redbooks/pdfs/sg246465.pdf updated June 2006 DB2 for z/OS Data Integrity SG24-7111 also redpaper on disk and DB2 http://www.redbooks.ibm.com/redbooks/pdfs/sg246465.pdf new June 2006 Design Guidelines for High Performance and Availability, SG24-7134 http://www.redbooks.ibm.com/redbooks/pdfs/sg247134.pdf update Feb 2006 DB2 for z/OS V8 publications update on web new in late February 2006 http://www.ibm.com/software/data/db2/zos/v8books.html Stored Procedure http://www.redbooks.ibm.com/redbooks/pdfs/sg247083.pdf updated Feb 2006 DB2 for z/OS and WebSphere: The Perfect Couple, SG24-6319 http://www.redbooks.ibm.com/redbooks/pdfs/sg246319.pdf Planning for Multilevel Security & Common Criteria (GA22-7509) http://publibz.boulder.ibm.com/epubs/pdf/e0z2e122.pdf Multilevel Security & DB2 Row-Level Security Revealed, SG24-6480 http://www.redbooks.ibm.com/redpieces/pdfs/sg246480.pdf Ready for Java http://www.redbooks.ibm.com/redbooks/pdfs/sg246435.pdf Large Objects http://www.redbooks.ibm.com/redbooks/pdfs/sg246571.pdf Cross-Platform SQL Reference V2 Sept 2004 http://ibm.com/developerworks/db2/library/techarticle/0206sqlref/0206sqlref.html
  • Part of the improvement in overall quality measures is due to more rigorous testing. Some of the improvement is due to the reduced amount of new function added in the service stream. Still these numbers represent the averages, and we know that our customers are not average. These results are highly skewed, so they might not reflect your experience. The average, median and mode customer submits 0 new APARs that have never been encountered before. We anticipate that the average customer will submit fewer than 10 problems in a year, with some questions and some problems that are diagnosed with an existing fix.
  • While DB2 for z/OS V8 provides many enhancements for security, there are still many more needs and much more work to do. Roles are used to provide a more flexible technique than groups or users in assigning and controlling authorization, while improving consistency with the industry. A network trusted context provides a technique to work with other environments more easily, improving flexibility. The instead of trigger is an SQL technique that allows a trigger to be used in place of a view, consistent with DB2 for Linux, UNIX and Windows. Improved audit selectivity is needed for being able to see that security is functioning. Secure Socket Layer or SSL implementation provides encryption of data on the wire. Some additional techniques for data encryption will help protect data at rest and in backups.
  • Trusted security context: Today, you have the option to set a system parameter which indicates to DB2 that all connections are to be trusted. It is unlikely that all connection types, such as DRDA, RRS, TSO, and batch, from all sources will fit into this category. It is likely that only a subset of connection requests for any type and source may be trusted or that you want to restrict trusted connections to a specific server. More granular flexibility will allow for the definition of trusted connection objects . Once defined, connections from specific users via defined attachments and source servers will allow trusted connections to DB2. The users defined in this context can also be defined to obtain a database role .
  • Database role: A database role is a virtual authorization ID that is assigned to the user via the context mentioned next. DB2 privileges are assigned to the defined role. The role exists as an object independent of its creator, so creation of the role does not produce a dependency on its creator. This capability can allow a DBA to have privileges to create objects and manage them for a time, even though ownership is to be another id. The role can be assigned and removed from individuals via the trusted authorization context as needed. This allows a DBA to perform object maintenance during a change control window on a Saturday night, for example. But when Monday arrives, they do not have the authority to do this same work. Auditing trails of the work completed during the maintenance window are available for verification by a security administrator or auditor.
  • Improved trace filtering makes the jobs of auditing and of performance management easier. Many more options can be used to minimize the amount of data collected, so the overhead is reduced and the extraneous data does not need to be processed. Being able to include or exclude, based upon more qualifications reduces performance concerns and the amount of data to store and process.
  • ROLEs and Trusted Context can be used to provide added security for your network-attached application servers. These new capabilities allow the DBA to make GRANT statements conditional, so that they can only be used from a specified list of IP addresses. If someone steals the application server’s userid/password, they won’t be able to access the database unless they are also able to execute the SQL statement on one of the approved application servers.
  • ROLEs and Trusted Context also enable customers to improve DB2 system auditing. Today, many customers use a “system” userid to access DB2 so that they don’t have to grant dynamic SQL privileges to their end users. A second benefit to the system userid is connection pooling in the application server. With V9, customers will be able to grant dynamic SQL table privileges to a ROLE, and specify that the end user can only use that ROLE when the end user is running on an approved application server. This has several benefits: A ROLE can be used as a single database authid that can be used to simplify administration of dynamic SQL privileges. The end user’s authid can be used to run database transactions, so that the DB2 audit is able to identify the end users individually (important capability for meeting some regulatory compliance requirements). The Trusted Context retains many of the performance benefits of connection pooling, while eliminating the restriction that a single authid (the system authid) must be used for all the uses of the connections in the pool.
  • Statement of Direction: To address customers' growing concern with data security, IBM is announcing a statement of direction for the development, enhancement and support of encryption capabilities within storage environments such that the capability does not require the use of host server resources. This includes the intent to offer, among other things, capabilities for products within the IBM TotalStorage ® portfolio to support outboard encryption and to leverage the centralized key management functions planned for z/OS ICSF. The first change comes in the TS1120 tape drive, but the rest is beyond currently announced products, including DB2 V9.
  • The second category of innovation is COST SAVINGS THROUGH OPTIMIZATION o Increased security and regulatory compliance through implementation of roles, network-trusted contexts, and enhanced auditing o Performance-boosting innovations such as load and reorg CPU reductions, improved performance for varying length data, and improved logging and insert performance o Synergy with IBM System z and z/OS in areas that include XML parsing, zIIP, MIDAW channel improvements, encryption, IPv6 and Secure Socket Layer (SSL) o Query management enhancements to make accessing your data even faster and more accurate with indexing improvements that include index on expression, randomization, and larger index page sizes and optimization improvements that provide better data for the optimizer, improved optimization techniques, and better management with optimization services
  • This text just shows the relationship of DB2 for Linux, Unix & Windows with DB2 for z/OS and OS/390 Version 7, comparing a March 2001 z/OS version with an October 2004 LUW version. V7 has almost no unique function, there is a small set of common function, and a larger set of SQL unique to LUW. The next step in the process is DB2 for z/OS Version 8. There are three sets of SQL noted above, with none that is unique to DB2 for z/OS in the first group, SQL that is common across DB2 for Linux, Unix, Windows and z/OS in the large group in the middle, then SQL that is unique to DB2 for Linux, Unix and Windows in the bottom group. Sheryl Larsen provided the base for this information, but the mistakes are probably mine.
  • This text just shows the relationship of DB2 for Linux, Unix & Windows with DB2 for z/OS, comparing the z/OS Version 8 from March 2004 with the LUW version from October 2004. There are three sets of SQL noted above, with some that is unique to DB2 for z/OS in the first group, SQL that is common across DB2 for Linux, Unix, Windows and z/OS in the large group in the middle, then SQL that is unique to DB2 for Linux, Unix and Windows in the bottom group. Sheryl Larsen provided the base for this information, but the mistakes are probably mine. If you want to improve DB2 family consistency, then DB2 for z/OS Version 8 is a big step, changing the game from one of catch up to one of leapfrog.
  • This text just shows the relationship of DB2 for Linux, Unix & Windows with DB2 for z/OS. This step in the process is DB2 V9 for z/OS, (V9). V9 moves about half of the LUW unique items into the common set and adds a little more that is unique to the z platform. At about this time we’ll also have a new release of DB2 V9.1 for LUW, code named Viper. We are able to move more from the z list to the common list with Viper. There are three sets of SQL noted above, with some that is unique to DB2 for z/OS in the first group, SQL that is common across DB2 for Linux, Unix, Windows and z/OS in the large group in the middle, then SQL that is unique to DB2 for Linux, Unix and Windows in the bottom group. Sheryl Larsen provided the base for this information, but the mistakes are probably mine.
  • As in Version 8, there are many improvements for SQL and for XML in V9. Improvements in the SQL have made migrating from other platforms, such as Unix and Windows much easier. V9 continues the progress in SQL, with many new functions, statements and clauses. The biggest changes are in XML on the prior slide. There are new SQL data manipulation statements in MERGE and TRUNCATE. There are new data types with DECIMAL FLOAT, BIGINT, BINARY and VARBINARY types. Improvements in LOBs provides more consistent handling and improved performance. Intersect and Except set operations make some SQL operations simpler to specify. Security is improved with ROLEs and network trusted context. Data definition consistency and usability are improved. V9 is another big step in DB2 family consistency and in the ability to port applications to DB2 for z/OS.
  • Native SQL stored procedures Stored procedures written in SQL procedure language enhance portability and ease of use when using DB2 for z/OS as your enterprise information source. This language is an ANSI standard language. It is similar to the proprietary stored procedure languages of several competitive databases, which assists in migrating and porting to DB2 for z/OS. This is very similar to the changes in DB2 for LUW V8.2. SQL stored procedures are supported by the DB2 Development Center tooling, providing an environment to code, test, and debug modules from your connected workstation. This language is currently converted to C when the CREATE PROCEDURE statement is executed. The C program is then automatically prepared, compiled, linked, and bound. The developer does not need to work with the C code. SQL stored procedures code will be natively integrated into the DB2 engine, eliminating the conversion to C. Additionally, extensions to the bind command will allow for the promotion of the program and access paths between environments without needing to recreate the stored procedure. Native SQL stored procedures are eligible to have a portion run on zIIP processors when they are invoked from a remote client.
  • This statement provides a fast way to delete rows with SQL, with better application portability. Truncate Table provides for the rapid removal of rows from a table. You can use this function to delete the contents of a table before applying data via LOAD or INSERT or MERGE.
  • New data type support is provided. Support will be added for: Decimal floating point numbers, similar to calculator mathematics and supporting the IEEE standard. These numbers can have more precision than current floating point. BIGINT support of double word (8 byte) integer values VARBINARY, providing better comparison support for binary strings
  • Merge Online transaction processing (OLTP) workloads that need to place data for several rows into a table will benefit from merge SQL. This capability will allow for updating of existing rows and creation of new rows through a single SQL statement. Multi-row insert-type capability is extended, via the MERGE statement, to take an array of column values from a program and perform insert and update operations against a table. DB2 will use the matching criteria in the merge SQL to update existing rows and perform inserts for rows that do not exist, through a single SQL statement.
  • Select from DELETE, UPDATE, and MERGE: The object-relational capabilities of DB2 allow for the incorporation of business logic into the database. This extends the power of SQL. Sometimes the application needs to know the results of this logic, when applied to the SQL issued. A subsequent SELECT for the data adds complexity and execution time to the application. The insert within select feature of DB2 for z/OS Version 8 has been expanded to include the retrieval of columns from rows that are modified via DELETE, UPDATE, and MERGE SQL. One SQL call to DB2 modifies the table contents and returns the resultant changes to the application program. When used with DELETE, the application now has the option to code a destructive read from a table. This is particularly useful when a table is used as a data queue, as with many vendor packages. Subquery improvements: Correlated and non-correlated subqueries will benefit from improved optimization. They will provide added flexibility with the support of ORDER BY and FETCH FIRST clauses.
  • The text improvements in DB2 9 include many improvements and new functions which can provide fast search for text, including indexing, using the new index on expression with text functions like UPPER that is sensitive to locale, COLLATION_KEY, SOUNDEX and DIFFERENCE functions. Improvements for large objects and XML also help with many text applications. Specialized text search with a separate server is noted on the next pages.
  • The text improvements in DB2 9 include many improvements and new functions which can provide fast search for text, including indexing, using the new index on expression with text functions like UPPER that is sensitive to locale, COLLATION_KEY, SOUNDEX and DIFFERENCE functions. Specialized text search is noted on the next page.
  • A specialized text search engine on a separate server is expected to come in V9, but will deliver after general availability. While the server is separate, the text indexes are saved into DB2 tables to provide improved backup and recovery.
  • GIS = Geographic Information System. Aka Geospatial. E.g. Google Maps. OGC = Open Geospatial Consortium You can use IBM Spatial Support for DB2 for z/OS to generate and analyze spatial information about geographic features, and to store and manage the data on which this information is based. A geographic feature is anything in the real world that has an identifiable location, or anything that could be imagined as existing at an identifiable location. A feature can be: An object (that is, a concrete entity of any sort); for example, a river, forest, or range of mountains. A space; for example, a safety zone around a hazardous site, or the marketing area serviced by a particular business. An event that occurs at a definable location; for example, an auto accident that occurred at a particular intersection, or a sales transaction at a specific store. IBM Spatial Support for DB2 for z/OS makes available facts and figures about the locations of geographic features. You can use functions to provide spatial data. IBM Spatial Support for DB2 for z/OS provides services to import spatial data in Shapefile and SDE Export formats.
  • Additional family compatibility Additional compatibility will be added for: Default databases and table spaces: When porting from other DBMS that do not have the same use for database and table space, providing them automatically will reduce the effort for delivering good performance. Automatic unique indexes are created to support defined primary keys. The default for a table space will change to segmented, improving performance and manageability for many customers. Existing simple table spaces are supported, but new table spaces will be segmented or partitioned.
  • Application programming is using a wider range of tools, environments and languages. The Eclipse framework is growing strongly. We need to connect the new languages and environments to the scale and value of the existing infrastructure. The Rational and WebSphere product lines provide part of the connection, with products like Rational Data Architect and WebSphere Information Integration. DB2 clients provide more support for new environments and new languages.
  • Improving data warehousing and reporting: Today’s complex applications include both transactions and reporting, so performing both well is imperative. The key improvements for reporting are optimization enhancements to improve query and reporting performance and ease of use. Improved data is provided for the optimizer, with improved algorithms and a rewritten approach to handling performance exceptions. More queries can be expressed in SQL with new SQL enhancements. The set operators INTERSECT and EXCEPT clauses make SQL easier to write. OLAP extensions for RANK, DENSE_RANK and ROW_NUMBER add new capabilities. Other SQL statements improve consistency with the DBMS industry. V9 continues the progress in SQL, with many new functions, statements and clauses. The biggest changes are in XML on a prior slide. New SQL data manipulation statements are MERGE and TRUNCATE. New data types with DECIMAL FLOAT, BIGINT, BINARY and VARBINARY. Improvements in LOBs provide new function, more consistent handling and improved performance. Data definition consistency and usability are improved. V9 is another big step in DB2 family consistency and in the ability to port applications to DB2 for z/OS. Histogram statistics enable DB2 to improve access path selection by estimating predicate selectivity from value-distribution statistics that are collected |over the entire range of values in a data set. RUNSTATS cannot collect histogram statistics on randomized key columns. DB2® chooses the best access path for a query based on predicate selectivity estimation, which in turn relies heavily on data distribution statistics. Histogram statistics summarize data distribution on an interval scale by dividing |the entire range of possible values within a data set into a number of intervals. DB2 creates equal-depth histogram statistics, meaning that it divides the whole range of values into intervals that each contain about the same percentage of the total number rows.
  • Example: Fact Table (“Transaction”) and 4 dimension tables (Prod, Cust, Stores, Time). All local predicates on the Dim tables – they are the only ones with meaningful data. Fact table just has id’s. No local predicates on fact table. SELECT Pid, Pname, SUM(sales) FROM tr, c, p, s, t WHERE tr.cid=c.id + other joins p.type IN (‘a’,’b’,’c’) s.region IN(‘west’,’east’) t.year IN (‘2001’,’2002’) GROUP BY pid,pname ORDER BY pid A good access path starts with 1 or more dim tables joined to fact table to exploit the indexes in Fact. Problem is that the dim tables don’t join directly with one another – they must all join to the Fact. Pre V9 relies on multi-column indexes. V8 performs ok if the right indexes are defined, but this is hard to figure out (Terry Purcell skills). In V9 we implemented dynamic index ANDing to solve this. Do pair-wise join for each Dim table to Fact in parallel. Get the RIDs, then intersection. (LUW does this, but they do not have the ‘dynamic’ part of this that follows). Problem is that the predicate selectivity estimates aren’t reliable. If wrong, then could order the dim tables in the wrong way. Dynamic pairwise index ANDing solves this. Kick off all the dim tables in parallel and child tasks return RIDs to parent on intervals. Parent can dynamically evaluate selectivity on each dimension and adjust access path on the fly according to runtime observations. Can abort one or more legs if not looking good (e.g. they could use up RID pool space). Now you can create single column indexes on the Fact and not have to worry about which multi-col indexes are needed. Much better out of the box solution with SAP BW.
  • In V9, 71 out of 100 used PWJ Key messages: V9 performs better than V8 for BW workloads with well-tuned index design (see V8 BW workload for details) V9 outperforms V8 for BW workloads with inadequate multi-column index support (see new BW workload for details) More room for zIIP off-load in V9 for BW workload V9 is a great out-of-box solution which removes the burden of index design from users Overall, a significant TCO reduction in V9 for BW customers
  • OSC to be available for V8 by end ’07 OE has 1) index advisor, 2) query advisor, 3) access path advisor.
  • We have had a UNION and a UNION ALL for a long time, but the only way to express the intersection and set difference was to code them in the WHERE predicates, rather than as a set operation. Now we have a full set of set operations.
  • The second category of innovation is COST SAVINGS THROUGH OPTIMIZATION o Increased security and regulatory compliance through implementation of roles, network-trusted contexts, and enhanced auditing o Performance-boosting innovations such as load and reorg CPU reductions, improved performance for varying length data, and improved logging and insert performance o Synergy with IBM System z and z/OS in areas that include XML parsing, zIIP, MIDAW channel improvements, encryption, IPv6 and Secure Socket Layer (SSL) o Query management enhancements to make accessing your data even faster and more accurate with indexing improvements that include index on expression, randomization, and larger index page sizes and optimization improvements that provide better data for the optimizer, improved optimization techniques, and better management with optimization services
  • The key performance improvements in V9 are reduced cpu time in the utilities, improved LOB performance and scalability, improved optimization for SQL, the zIIP processing for remote native SQL procedures, reduced cpu time for data with varying length and better sequential access. I’ll discuss the optimization improvements on the query slide. V8 SQL procedures were not eligible to run on the zIIP, but changing to use the native SQL Procedure Language on V9 will make the work eligible for zIIP processing. Varying length data can improve substantially if there are large numbers of varying length columns. Several improvements in disk access can reduce the time for sequential disk access.
  • The latest System z9 processor improvements for DB2 are the zIIP and the new Business Class and Enterprise Class processors. DB2 V9 remote native SQL procedures are enabled for zIIP processing. V9 adds IPv6, SSL and decimal float and BIGINT data types, with enhancements for Parallel Sysplex, backup and restore, added security and encryption, more Unicode with collation, and uses the WLM in new ways. Channel enhancements (MIDAW) and improved DS8000 performance were included with the System z9 announcements. DB2 uses the latest improvements in hardware and operating system to provide better performance, improved value, more resilience and better function. DB2 benefits from large real memory, faster processors, specialty engines, and better hardware compression. DB2 uses Parallel Access Volume and Multiple Allegiance features of the IBM DS8000 and Enterprise Storage Server™. FlashCopy® can be used for DB2 backup and restore. DB2 makes unique use of the z/Architecture™ instruction set, and recent instructions provide improvements in reliability, performance and availability. DB2 continues to deliver synergy with hardware data compression, FICON™ (fiber connector) channels, disk storage, advanced networking function, and Workload Manager (WLM).
  • MIDAW requires z9 (2094) and z/OS 1.6 with APARs OA10984 8/05, PK13324/13384 9/05 or later. Refer to earlier chart for performance figures with throughput and response times. Then add Faster channels (4 Gb) Faster disk DS8000 Turbo Sequential improvements in V9 – read and insert Logging improvements in V9
  • DB2 9 preliminary figures for utility cpu use are very encouraging. There is substantial variation in the figures, depending upon the exact workload, but the improvements are large and across most of the key utilities: copy, reorg, load, rebuild, recover, runstats, and check index.
  • Many other improvements help with performance, with scalability and with availability. Index on an expression can be combined with caseless comparisons to improve text search. Improved insert rates can result from improved log performance, larger index page sizes, and improved index page splits. See the next slide for index compression. Data sharing logging rates should increase significantly. Archive logs can use striping, providing a performance boost when needed. A randomized index key will optimize for inserting, like the APPEND option for data. Significant reductions in cpu usage are provided with the new utilities. REBUILD INDEX has an online option for the first time, while REORG becomes much more online by removing the BUILD2 phase. The not logged table and index space can help with scaling where the log is a bottleneck, but won’t increase performance significantly. The XML index is very different from any other index.
  • Index compression relies upon page level compression instead of row-level compression (technique used for tables). Indexes with 32K page sizes can save up to 8x on disk space with the compression feature turned on. The solution has to guarantee that a given page will compress to 4K on disk, so DB2 will sometimes force pages in memory to be “partial” so that the compression will indeed result in no more than a 4K byte page on disk.
  • DSN1COMP utility to simulate compression ratio without real index compression Notes No compression or decompression in each Insert or Fetch; instead at I/O time Load or Reorg not required for compression Based on a very limited survey so far Higher for relatively unique indexes with long keys CPU time impact under study
  • Sequential insert performance is improved by avoiding page splits with larger index page sizes and the ability to split a page more effectively. Other changes improve logging rates. Larger page sizes: up to 8x fewer splits Asymmetric index split: up to 2x fewer splits, cpu/elapsed time savings. Up to -20% cl2 cpu, -31% elapsed
  • Table Append option The Table Append option offers increased performance for inserting data into the end of a table. It reduces the instructions used to target the locations for new rows. Index maintenance techniques are also made more efficient when DB2 detects that entries are added to the beginning or ending of the index.
  • Impact on log record: log size can be bigger or smaller. This is not an option. This is only for data, not for indexes.
  • Large object improvements: Large objects (LOBs) were introduced in DB2 Version 6. Usage has increased substantially in the past few years, and major enhancements have been made in DB2 Version 8. APARs on Version 8 deliver the ability to use utilities for loading and unloading large LOB data. File reference variables are used to let the large objects be accessed from data sets instead of from storage. The abilities to reorganize and to recover space are provided. Future changes will help with improved function and usability, DB2 family compatibility, cost of ownership, performance, and scalability.
  • Large object improvements: Large objects (LOBs) were introduced in DB2 Version 6. Usage has increased substantially in the past few years, and major enhancements have been made in DB2 Version 8. APARs on Version 8 deliver the ability to use utilities for loading and unloading large LOB data. File reference variables are used to let the large objects be accessed from data sets instead of from storage. The abilities to reorganize and to recover space are provided. Future changes will help with improved function and usability, DB2 family compatibility, cost of ownership, performance, and scalability.
  • Distributed connections to DB2 for z/OS will benefit from z/OS V1R7 changes that DB2 will exploit. DB2’s distributed communication processes (the distributed address space ) will access data directly from the database manager address space, instead of moving the data. The distributed address space will also exploit 64-bit addressing, as the database manager and lock manager address spaces do today with Version 8. This internal change will benefit new and existing workloads, where distributed communications are configured with another logical partition (LPAR) or to an application running on the zSeries platform.
  • Cost of Complexity As clients build out their infrastructures….the complexity becomes enormous. They need to optimize IT assets NOW to fuel growth, improve ROI, increase staff productivity and improve quality of service. Often, the cost of acquiring systems is now dwarfed by the cost of managing them. IDC recently came out with a report where they stated that 15 years ago, 75% of IT budgets were spent on new hardware and software; today 75% is spent on just running the infrastructure. System vendors selling simplicity likely will win in the future, according to Steve Milunovich of Merrill Lynch. Server units are growing at a 20% rate, causing management headaches. Global server support costs could top $140 billion in 2008, nearly 2.5X the estimated hardware sales of $60 billion. Milunovich said he thinks IBM is well positioned with its on demand initiative and $1 billion of R&D spent to automate services. HP has the vision though it lacks some of the pieces and must show it can execute. Sun faces an uphill mindshare battle though its systems understanding shouldn’t be underestimated. Dell doesn’t have the know-how to succeed but may not need it as it leverages industry R&D and simplifies through low-cost, standard products. Microsoft and Dell do recognize that industry standard servers can be difficult to administer and have joined forces on systems management. Our Virtualization Engine addresses this problem, enabling clients to lower the cost of their existing infrastructure, reduce the complexity of adding to that infrastructure, and build heterogeneous infrastructure across multiple datacenters that are more responsive to their business needs. So clearly simplifying your environment is critical to your future success.
  • Temporary storage architecture and use Many recent improvements have enhanced the use of temporary storage when it is needed by the DB2 engine, for example, for interim materialized result sets. The different temporary pools are converged to a single source. Use of temporary storage is eliminated when processing small amounts of data.
  • V8 provided a new BACKUP utility using FlashCopy technology to take very fast backups of the entire subsystem without any disruption. RECOVER is only for the whole subsystem. With Vnext we expect to be able to recover an object, rather than the whole subsystem. This makes the job of backup and recovery simpler and easier.
  • DB2 periodically reports the buffer pool size and the hit ratio for random reads to WLM via a "data collection" exit. WLM plots the size and hit ratio over time to allow it to project the effects of changing the size, then adjusts the buffer pools, if WLM goals will improve.
  • Version 9 builds upon the structure from V8, with the same modes, but clarifying the situation if you drop back after moving to ENFM or NFM.
  • Skeletons above the bar: For customers that use heavy package and plan activity such as banks, this is the most significant DBM1 below the bar storage relief in V9. For customers that use very few or small packages, such as SAP environments, the savings is smaller. LI 702 – move spaceblk (SPA) above the bar. S PA to be split into 2, 1 above, 1 below. Only a few, non-complex RTs are being considered for V9. Simple insert, delete Expected results will vary by SQL mix. (-5 to 30%)?
  • RLF is extended to support the Set Client Information APIs that are exposed in CLI and JDBC. This allows the customer to control CPU cost of dynamic queries for packaged applications and user-written applications that exploit these APIs. CREATE | ALTER STOGROUP DATACLAS dcname MGMTCLAS mcname STORCLAS scname The existing VOLUME clause is now optional: it can be omitted if any of the DFSMS classes is specified. If explicitly specified to DB2, the new attributes are recorded in SYSIBM.SYSSTOGROUP
  • This change will help with migration of applications which use application consistency techniques. Application programmers will not need to be responsible for updating the timestamps. A simple timestamp predicate can be used to make sure that other updates are not lost.
  • Now let’s discuss the wide range of new function being provided inside the DB2 for z/OS engine. This work is being done in parallel with similar changes in DB2 for Linux, Unix and Windows. DB2 for z/OS extensibility has been implemented via extenders . Extenders for text, image, audio, video, and XML are delivered with DB2 and are optional installation components. They provide the tooling, user defined data types (UDT), user defined functions (UDF), and stored procedures to managed non-relational data. The XML extender provides for the storage and indexing of an XML document as a character large object (CLOB), or for the shredding of the XML document into relational columns for storage and query. DB2 Version 8 expanded on XML support by implementing several XML publishing operations as built-in DB2 functions. This allows you to perform XML document composition from relational data with improved performance without the XML Extender. There will be expanded support of XML in DB2 by integrating more features into the engine. This includes an XML data type, native storage of XML documents, integration of the XPath language, and catalog extensions to support definitions of XML schemas. Utilities will support creation and maintenance of XML data.
  • Support for pureXML is provided in DB2 by integrating more features into the engine. This includes an XML data type, native storage of XML documents, integration of the XPath language, and extensions to support definitions of XML schemas. Utilities support creation and maintenance of XML data. The SQL/XML publishing functions provided in DB2 V8 are enhanced, and many new functions and predicates are added.
  • The utilities are extended to support XML. You can LOAD, UNLOAD, CHECK DATA or INDEX, COPY, REBUILD, RECOVER and REORG the XML data. The XML schema support handles a schema repository and uses tables to store XML schema information. Stored procedures are provided to register XML schemas. IBM Systems Journal issue on XML: http://www.research.ibm.com/journal/sj45-2.html
  • DRDA supports internally encoded or externally encoded XML type (in serialized string format).
  • One of the key initiatives of V8 was online schema evolution, and that theme is expanding and changing to be data definition on demand. One of the important changes is to be able to replace one table quickly with another. Another is to be able to rename a column or an index. Index page size can be altered. A new type of table space combines the attributes of segmented and partitioned, without a partitioning key. Rebuild index can be run with much less disruption. Online table space reorganization for a few partitions is improved a lot, removing the BUILD2 phase for all types of secondary indexes. Table space and index logging can be altered. A new ability to change the DB2 early code does not require an IPL.
  • Fast replacement of a table with another This function will allow you to generate a copy of a current table with the same attributes and same data, in the same table space. It appears to an application or user as a very fast replacement of data within a table. Web-based applications, striving for maximum availability will benefit from the option of implementing a pair of tables that are clones of each others’ structure. Copies for application testing and auditing can be easily created. These clone tables will have the unique ability to change names quickly. Applications can quickly and almost transparently switch between dual mirror tables.
  • The tables can be partitioned or non-partitioned, and are created with the CREATE TABLE syntax. The primary table’s structure, including indexes, large objects (LOBs) and before triggers, will be copied. Information can then be inserted or loaded into the copy table, and the copy can have its own image copies. When the data in the copy table needs to become active to an application, an ALTER statement will switch the name, providing fast replacement of the original data
  • Partitioned by growth Partitioned tables have required key ranges to determine the target partition for row placement. When a table is partitioned, you gain the benefits of scaling objects to hold more data. You also benefit from more granular locking and parallel operations by spreading the data over more data sets. The option to partition by growth will allow segmented tables to be partitioned as they grow, without needing key ranges. These segmented tables will gain increased table space limits and the SQL and utility parallelism, afforded for partitioned tables. CREATE TABLESPACE … (explicit specification) MAXPARTITIONS integer CREATE TABLE … (implicit specification) PARTITIONED BY SIZE EVERY integer G PBR can also be UTS via CREATE TABLESPACE … SEGSIZE integer NUMPARTS integer
  • More online utilities are coming! Online rebuild build an index in place after setting RBDP. Dynamic SQL will avoid the index until available. This works well for non-unique indexes. Cloned tables are somewhat like Online Load Replace. There is the live table, and then there is the shadow or clone. One can load the clone, and then when ready for the clone to become live, a “switch” can be trigger with DDL. Parallel log apply for SHRLEVEL REFERENCE and CHANGE Multiple concurrent jobs no longer needed/supported – DSNU180I NPSIs are also shadowed (higher disk requirement compared to when just logical part was reorganized), thus resulting in the NPIs also being reorganized -> therefore, don’t follow with REORG of NPSIs. Higher availability results in increased resource consumption REORG SHRLEVEL REFERENCE now has a log phase since entire NPSI is shadowed and reorg’d Online reorg solves the outage during the BUILD2 phase when reorging a partition at a time. Online Check Data and Check Lob are the same as existing utilities, only with SHRLEVEL CHANGE. Volume-based utilities allow recovery of individual objects after a volume based backup. It also extends system based backups and restores to and from tape. Broken page recovery is a serviceability improvements when data has become broken because log apply fails. Template switching allows using tape vs dasd if a size or limit on the ds is exceeded. Modify Recovery is extended to allow n number of backups vs an age. Check || is the SHRLEVEL CHANGE technique extended to SHRLEVEL REFERENCE CPU reduction is to improve where we don’t match up well against other vendors on a cost basis. Histogram stats uses quantiles to yet give the optimizer more stats.
  • More online utilities are coming! Online rebuild build an index in place after setting RBDP. Dynamic SQL will avoid the index until available. This works well for non-unique indexes. Cloned tables are somewhat like Online Load Replace. There is the live table, and then there is the shadow or clone. One can load the clone, and then when ready for the clone to become live, a “switch” can be trigger with DDL. Online reorg solves the outage during the BUILD2 phase when reorging a partition at a time. Online Check Data and Check Lob are the same as existing utilities, only with SHRLEVEL CHANGE. Volume-based utilities allow recovery of individual objects after a volume based backup. It also extends system based backups and restores to and from tape. Broken page recovery is a serviceability improvements when data has become broken because log apply fails. Template switching allows using tape vs dasd if a size or limit on the ds is exceeded. Modify Recovery is extended to allow n number of backups vs an age. Check || is the SHRLEVEL CHANGE technique extended to SHRLEVEL REFERENCE CPU reduction is to improve where we don’t match up well against other vendors on a cost basis. Histogram stats uses quantiles to yet give the optimizer more stats.
  • What is beyond DB2 9? Work for the future provides the most comprehensive database functionality (OLTP and BI), the best performance and scalability, the highest RAS characteristics and, most importantly, much reduced administration cost. Today’s database servers consist of independent components that require significant database and systems administration cost. Using tight integration of DBMS, operating system and hardware we can dramatically reduce the cost by: eliminating need for OS skills and System z-unique skills reducing DBA to absolute minimum use open, Web interfaces when interaction necessary What are targeted ecosystems? DB2 can be used for all applications, new and existing ones, home-grown or created by ISVs, transactions, queires, web serving or a mixture. One key target will be existing and new ISV installations such as SAP, Siebel, ... Why are these changes attractive to customers and ISVs? Customer and ISV objective is to drive down the customers total cost of ownership(TCO). Their game plan is to achieve this through non-ISV (such as IBM's) delivered hardware and software stack commoditization and people cost (systems administration) reduction. IBM's goal is also lowering the customers TCO. The people cost (systems administration) improvements through better usability, improved productivity and delivery of autonomic function complement hardware and software improvements in value.
  • Main DB2 for z/OS web page: pointers to most of the following: http://www.ibm.com/software/data/db2/zos/index.html V9 beta announcement: http://www.ibm.com/common/ssi/fcgi-bin/ssialias?subtype=ca&infotype=an&appname=iSource&supplier=897&letternum=ENUS206-098 http://www.ibm.com/common/ssi/rep_ca/8/897/ENUS206-098/ENUS206-098.PDF DB2 9 for z/OS main page: http://www.ibm.com/software/data/db2/zos/db2zosv91.html Overview presentation, webcast and foils with notes: http://www.ibm.com/software/os/zseries/webcast/18may/ ftp://ftp.software.ibm.com/software/data/db2zos/DB2V9zOS.pdf Redbooks: SOA book includes overview of V9 XML http://www.redbooks.ibm.com/abstracts/SG247259.html?Open LOBs book includes V9 http://www.redbooks.ibm.com/abstracts/SG247270.html?Open Coming: security, V9 overview, V9 Performance Topics Detailed presentations: Start on the Events page http://www.ibm.com/software/data/db2/zos/events.html Click on Presentations from previous conferences. Sort results by date - newest first. Some require registration. http://www.ibm.com/support/docview.wss?rs=64&context=SSEPEK&dc=DA400&q1=presentation&uid=swg27008769&loc=en_US&cs=utf-8&lang=en http://www.ibm.com/support/docview.wss?rs=64&context=SSEPEK&dc=DA400&q1=presentation&uid=swg27008767&loc=en_US&cs=utf-8&lang=en http://www.ibm.com/support/docview.wss?rs=64&context=SSEPEK&dc=DA400&q1=presentation&uid=swg27008766&loc=en_US&cs=utf-8&lang=en http://www.ibm.com/support/docview.wss?rs=64&context=SSEPEK&dc=DA400&q1=presentation&uid=swg27008743&loc=en_US&cs=utf-8&lang=en Access the ftp site directly: ftp://ftp.software.ibm.com/software/data/db2zos/ IOD2559DB29ClonesLyle.pdf V9 clone tables IOD1851DB2v9onlineUtilitiesHartmann.pdf V9 utilities IOD1855DB2v9designHartmann.pdf V9 applicationdesign IOD1869DB2zOSv9performPreShibamiya.pdf V9 performance preview IOD1819PurcellV9.pdf V9 optimization IOD1730DB2v9xmlZhang.pdf V9 XML IOD1729aDB2v9backuprecoveryTeng.pdf V9 backup and recovery IOD1641DB2V9autoQueryTuningFuh.pdf V9 Optimization Service Center part 1 IOD1642DB2v9autoQueryTuningFuh.pdf V9 Optimization Service Center part 2 IOD1450aDB2v9SAPHrle.pdf V9 for ERP and SAP IOD1438db2v9LOBsWeihrauch.pdf V9 LOBs IOD1345DB2zOSmigrationg.pdf V8 & V9 migration IOD1166DB2v9zOSbeyondCotner.pdf V9 overview DB2V9forzOS.pdf V9 overview DB2V9zOS.ppt V9 overview PowerPoint IOD conference. There were about 20 presentations there, and most are on the conference web site and on the recently shipped CD. Sessions from list above are 1166, 1345, 1438, 1450, 1641, 1642, 1729, 1730, 1819, 1851, 1855, 1869, and 2559. 1388 QMF V8 V9 1439 DB2 and SOA 1745 stored procedures 1747 Java stored procedures 2110 V9 not logged table spaces 2461 scalability V8 V9 When I am asked about GA, the first answer is "When it's ready." Our first job is making sure the product is solid at GA, so the date will move to achieve the first objective. We expect the date to be sometime in the first half of 2007, but I don't write that down. "When it's ready." is the real answer. When we know, we will have the announcement. V9 and V8 migration resources are on the web, and we do have webcasts for some. The primary pointer for resources is the DB2 Events page, http://www.ibm.com/software/data/db2/zos/events.html This page has some webcasts on it directly. V8 Are you done yet? V8 virtual storage. They are recorded, so listen and watch. Clicking on recent presentations shows 215 presentations today. Add V9 to the qualification and Sort by date - newest first. About 13 are listed. V9 Utilities Go Online V9 overview (beta launch)
  • IBM Information On Demand 2007 global conference is the premier information management event for business and IT executives, managers, professionals, DBA's and developers. More IBM technical and business solutions content will be in one place than ever before! Select from over 500 sessions: a 2-1/2 day business leadership track with 100 sessions and a 5 day technical track with 500 sessions. Why should you attend this conference? - Participate in the PREMIER discussion on the future of Information Management - Experience the entire Information Management software portfolio - Learn how to unlock business value and drive competitive advantage - Hear how your peers are realizing ROI - Understand the roadmap to long term strategic advantage - Learn best practices in your industry - Receive the best in technical education and free certification testing - Extensive opportunities for networking with both your peers and industry experts - Most of all, by combining individual conferences into one global event, you gain access and exposure to the full breadth of IBM Information Management technologies and experts The agenda offers the flexibility to easily manage your participation based on your needs. In our huge product Expo Solution Center you can talk to experts about the products, tools, and services you need from IBM and IBM Business Partners to manage your information infrastructure. Meet one-on-one with IBM executives and technical leaders, and attend multiple Birds of a Feather roundtables. Who should attend? IT Executives, Line of Business Executives, IT Decision Makers, Technical Staff, Developers, DBA's, IBM Business Partners, Consultants, Press and Analysts
  • For more information please consult these references.
  • Overview of the key information resources available. * Give this slide if you don’t have time to discuss resource individually* Information center Access the entire DB2 for z/OS library in a single location Search across all DB2 for z/OS, IMS, QMF, and Tools documentation Search messages and codes easily using the LookAt tool Official Introduction to DB2 for z/OS Get started with DB2 Version 8 Prepare for the DB2 Family Fundamentals certification exam Practice with sample questions Information roadmap Take advantage of an anchor site that provides links to a wide variety of useful information resources about DB2 for z/OS, Version Examples trading post Share examples of actual code Learn how customers are using the product Support Search APARs Find presentations, Redbooks, and white papers Read technotes for hints and tips
  • This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice. Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk. The licensed program described in this information and all licensed material available for it are provided by IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement, or any equivalent agreement between us. Any performance data contained herein was determined in a controlled environment. Therefore, the results obtained in the operating environments may vary significantly. Some measurements may have been made on development-level systems and there is no guarantee that these measurements will be the same on generally available systems. Furthermore, some measurements may have been estimated through extrapolation. Actual results may vary. Users of this document should verify the applicable data for their specific environment. Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. All statements regarding IBM's future direction or intent are subject to change or withdrawal without notice, and represent goals and objectives only. This information may contain examples of data and reports used in daily business operations. To illustrate them as completely as possible, the examples include the names of individuals, companies, brands, and products. All of these names are fictitious and any similarity to the names and addresses used by an actual business enterprise is entirely coincidental.
  • Click Here for Presentation

    1. 1. DB2 9 for z/OS Overview Jeff Josten DB2 for z/OS Development
    2. 2. <ul><li>Key Messages: </li></ul><ul><li>V7 end of service June 2008 </li></ul><ul><li>V9 general availability March 2007 </li></ul><ul><ul><li>Requires migration to V8 NFM </li></ul></ul><ul><li>DM Tools V8 & V9 Ready at GA </li></ul><ul><li>On V7? Migrate to V8 now </li></ul><ul><li>Plan for DB2 9 </li></ul>
    3. 3. DB2 for z/OS Version 8 News <ul><li>Cross loader with LOBs </li></ul><ul><li>Built in functions ASCII, TIMESTAMPDIFF </li></ul><ul><li>DSN1COPY with 1000 OBIDs </li></ul><ul><li>QMF with multirow fetch </li></ul><ul><li>Online Check Index </li></ul><ul><li>z/OS 1.7 up to 7257 extents </li></ul><ul><li>LOAD, UNLOAD with LOBs </li></ul><ul><li>IBM System z9 Integrated Information Processor (zIIP) </li></ul><ul><li>New and updated books: Library refresh Feb. 2007 </li></ul><ul><li>Messages, Codes separate books </li></ul><ul><li>Redbooks: SOA, Large Objects, Data Sharing in a Nutshell, Data Integrity, MIDAW performance, Disk & DB2, Design Guidelines for High Performance & Availability, Business Value, Performance Topics, WebSphere, Security & MLS, Disaster Recovery, others updated … </li></ul>
    4. 4. Worldwide V8 field experience totals V8 field experience is better than V7! <ul><li>V8 LOWER </li></ul>Multi-System Outages <ul><li>V8 LOWER </li></ul>PE Rate <ul><li>V8 QUICKER </li></ul>APAR Closure Time <ul><li>Comparable to V7 </li></ul>APAR/PMR Ratio <ul><li>V8 Severity 1’s LOWER </li></ul>Field APAR Severity <ul><li>V8 LOWER </li></ul>PMR Volume V8 / V7 Comparison Key Measurement
    5. 5. <ul><li>Improved IT Infrastructure for Compliance Efforts </li></ul><ul><ul><li>Trusted security context </li></ul></ul><ul><ul><li>Database roles </li></ul></ul><ul><ul><li>Auditing, encryption improved </li></ul></ul><ul><li>Simplify development and porting </li></ul><ul><ul><li>Many SQL improvements </li></ul></ul><ul><ul><li>Native SQL stored procedures </li></ul></ul><ul><ul><li>Default databases and table spaces </li></ul></ul><ul><li>Data Warehousing </li></ul><ul><ul><li>Dynamic index ANDing for star schema </li></ul></ul><ul><ul><li>EXCEPT and INTERSECT </li></ul></ul><ul><li>Decrease Complexity and Cost </li></ul><ul><ul><li>Partition by growth </li></ul></ul><ul><ul><li>Performance improvements </li></ul></ul><ul><ul><li>Volume-based COPY/RECOVER </li></ul></ul><ul><ul><li>Index compression </li></ul></ul><ul><ul><li>Optimization Service Center </li></ul></ul><ul><li>Evolve Your Environment & SOA </li></ul><ul><ul><li>Integrated pureXML ® </li></ul></ul><ul><ul><li>WebSphere ® integration </li></ul></ul><ul><li>Continuous Availability </li></ul><ul><ul><li>Schema evolution enhancements </li></ul></ul><ul><ul><li>Fast table replacement </li></ul></ul>DB2 9 for z/OS – Addressing corporate data goals
    6. 6. <ul><li>Key implementations </li></ul><ul><li>Network Trusted Contexts </li></ul><ul><li>Roles </li></ul><ul><li>Improved auditing </li></ul><ul><li>Secure Socket Layer </li></ul><ul><li>Instead of triggers </li></ul><ul><li>Data Encryption </li></ul>Regulatory Compliance in DB2 V9 for z/OS
    7. 7. DB2 Server CREATE CONTEXT WAS1 SYSTEM USERID WASPROD ADDRESS MY.WAS.SERVER ALLOW USER JOE WITHOUT AUTHENTICATION, SAM WITHOUT AUTHENTICATION; Trusted Security Context <ul><li>Identifies “trusted” DDF, RRS Attach, or DSN application servers </li></ul><ul><li>Allows selected DB2 authids on connections without passwords </li></ul><ul><ul><li>reduces complexity of password management </li></ul></ul><ul><ul><li>reduces need for an all-inclusive “system authid” in app servers with ability to switch users </li></ul></ul><ul><ul><li>more visibility/auditability of which user is current running </li></ul></ul><ul><ul><li>enables mixed security capabilities from a single app server </li></ul></ul>MY.WAS.SERVER SAM or JOE w/o password
    8. 8. Database ROLEs <ul><li>ROLE is a “virtual authid” </li></ul><ul><ul><li>Assigned via TRUSTED CONTEXT </li></ul></ul><ul><ul><li>Provides additional privileges only when in a trusted environment using existing primary AUTHID. </li></ul></ul><ul><ul><li>Can optionally be the OWNER of DB2 objects </li></ul></ul>CREATE ROLE PROD_DBA; GRANT DBADM … TO PROD_DBA; CREATE TRUSTED CONTEXT DBA1 … DEFAULT ROLE PROD_DBA OWNER(ROLE);
    9. 9. Auditing: DB2 Trace Filtering <ul><li>New filtering capabilities for –START TRACE that INCLUDE or EXCLUDE based on these keywords: </li></ul><ul><ul><li>USERID -- client userid </li></ul></ul><ul><ul><li>WRKSTN -- client workstation name </li></ul></ul><ul><ul><li>APPNAME -- client application name </li></ul></ul><ul><ul><li>PKGLOC -- package LOCATION name </li></ul></ul><ul><ul><li>PKGCOL -- package COLLECTION name </li></ul></ul><ul><ul><li>PKGPROG -- PACKAGE name </li></ul></ul><ul><ul><li>CONNID -- connection ID </li></ul></ul><ul><ul><li>CORRID -- correlation ID </li></ul></ul><ul><ul><li>ROLE – end user’s database ROLE </li></ul></ul>
    10. 10. DB2 Server CREATE CONTEXT SAP1 SYSTEM USERID SAPPROD ATTRIBUTES( ADDRESS ‘1.2.3.4’) o o o DEFAULT ROLE SAP_ROLE Example 1: ROLEs and Trusted Context used to Secure App Servers <ul><li>Most existing application servers connect to DB2 using userid/password pairs: </li></ul><ul><ul><li>Significant exposure if someone steals the userid/password!!! </li></ul></ul><ul><li>Trusted Context and ROLEs can be used to limit exposure: </li></ul><ul><ul><li>GRANTs to SAP_ROLE can be restricted so that they are only valid when used by a valid SAP app server IP address </li></ul></ul><ul><li>No change required to the code in the application server </li></ul>SAP App Server IP Address=1.2.3.4 SAPPROD connections
    11. 11. DB2 Server CREATE CONTEXT WAS1 SYSTEM USERID WASPROD … ALLOW USER JOE DEFAULT ROLE JDBC_ROLE, SAM DEFAULT ROLE JDBC_ROLE Example 2: ROLEs and Trusted Context for Dynamic SQL Auditing <ul><li>Better auditing controls: </li></ul><ul><ul><li>GRANT dynamic SQL privileges to a ROLE </li></ul></ul><ul><ul><li>End user identity can be delegated directly to DB2 without granting dynamic SQL privileges directly to the end user </li></ul></ul><ul><ul><li>End user passwords can be optional. </li></ul></ul><ul><ul><li>No added complexity for administration of GRANTs, while retaining the ability to audit the end user’s identity!!! </li></ul></ul>WebSphere JOE SAM WASPROD connections SAM
    12. 12. Future Directions – Extending Encryption to IBM TotalStorage <ul><li>Statement of Direction: To address customers' growing concern with data security, IBM is announcing a statement of direction for the development, enhancement and support of encryption capabilities within storage environments such that the capability does not require the use of host server resources. </li></ul><ul><li>This includes the intent to offer, among other things, capabilities for products within the IBM TotalStorage ® portfolio to support outboard encryption and to leverage the centralized key management functions planned for z/OS ICSF. </li></ul>Data Encryption keys Data Encryption keys Disk Statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only MFE_140 Tape TS1120 Enterprise-wide Key Management
    13. 13. DB2 9 for z/OS Innovation: SQL <ul><li>Numerous new SQL capabilities </li></ul><ul><li>Easier application porting </li></ul><ul><li>Simplified application development </li></ul>
    14. 14. DB2 SQL z z/OS V7 common luw Linux, Unix & Windows V8.2 Inner and Outer Joins, Table Expressions, Subqueries, GROUP BY, Complex Correlation, Global Temporary Tables, CASE, 100+ Built-in Functions, Limited Fetch, Insensitive Scroll Cursors, UNION Everywhere, MIN/MAX Single Index, Self Referencing Updates with Subqueries, Sort Avoidance for ORDER BY, and Row Expressions, Call from trigger, statement isolation Updateable UNION in Views, ORDER BY/FETCH FIRST in subselects & table expressions, GROUPING SETS, ROLLUP, CUBE, INSTEAD OF TRIGGER, EXCEPT, INTERSECT, 16 Built-in Functions, MERGE, Native SQL Procedure Language, SET CURRENT ISOLATION, BIGINT data type, file reference variables, SELECT FROM INSERT, UPDATE, or DELETE, multi-site join, 2M Statement Length, GROUP BY Expression, Sequences, Scalar Fullselect, Materialized Query Tables, Common Table Expressions, Recursive SQL, CURRENT PACKAGE PATH, VOLATILE Tables, Star Join Sparse Index, Qualified Column names, Multiple DISTINCT clauses, ON COMMIT DROP, Transparent ROWID Column, FOR READ ONLY KEEP UPDATE LOCKS, SET CURRENT SCHEMA, Client special registers, long SQL object names, SELECT from INSERT, MDC z l u w c o m m o n Range partitioning
    15. 15. DB2 SQL z z/OS V8 common luw Linux, Unix & Windows V8.2 Multi-row INSERT, FETCH & multi-row cursor UPDATE, Dynamic Scrollable Cursors, GET DIAGNOSTICS, Enhanced UNICODE for SQL, join across encoding schemes, IS NOT DISTINCT FROM, Session variables, range partitioning Inner and Outer Joins, Table Expressions, Subqueries, GROUP BY, Complex Correlation, Global Temporary Tables, CASE, 100+ Built-in Functions including SQL/XML, Limited Fetch, Insensitive Scroll Cursors, UNION Everywhere, MIN/MAX Single Index, Self Referencing Updates with Subqueries, Sort Avoidance for ORDER BY, and Row Expressions, 2M Statement Length, GROUP BY Expression, Sequences, Scalar Fullselect, Materialized Query Tables, Common Table Expressions, Recursive SQL, CURRENT PACKAGE PATH, VOLATILE Tables, Star Join Sparse Index, Qualified Column names, Multiple DISTINCT clauses, ON COMMIT DROP, Transparent ROWID Column, Call from trigger, statement isolation, FOR READ ONLY KEEP UPDATE LOCKS, SET CURRENT SCHEMA, Client special registers, long SQL object names, SELECT from INSERT Updateable UNION in Views, ORDER BY/FETCH FIRST in subselects & table expressions, GROUPING SETS, ROLLUP, CUBE, INSTEAD OF TRIGGER, EXCEPT, INTERSECT, 16 Built-in Functions, MERGE, Native SQL Procedure Language, SET CURRENT ISOLATION, BIGINT data type, file reference variables, SELECT FROM UPDATE or DELETE, multi-site join, MDC z l u w c o m m o n
    16. 16. DB2 SQL z z/OS V9 common luw Linux, Unix & Windows V9 Multi-row INSERT, FETCH & multi-row cursor UPDATE, Dynamic Scrollable Cursors, GET DIAGNOSTICS, Enhanced UNICODE for SQL, join across encoding schemes, IS NOT DISTINCT FROM, Session variables, TRUNCATE, DECIMAL FLOAT, VARBINARY, optimistic locking, FETCH CONTINUE, ROLE, MERGE, SELECT from MERGE Inner and Outer Joins, Table Expressions, Subqueries, GROUP BY, Complex Correlation, Global Temporary Tables, CASE, 100+ Built-in Functions including SQL/XML, Limited Fetch, Insensitive Scroll Cursors, UNION Everywhere, MIN/MAX Single Index, Self Referencing Updates with Subqueries, Sort Avoidance for ORDER BY, and Row Expressions, 2M Statement Length, GROUP BY Expression, Sequences, Scalar Fullselect, Materialized Query Tables, Common Table Expressions, Recursive SQL, CURRENT PACKAGE PATH, VOLATILE Tables, Star Join Sparse Index, Qualified Column names, Multiple DISTINCT clauses, ON COMMIT DROP, Transparent ROWID Column, Call from trigger, statement isolation, FOR READ ONLY KEEP UPDATE LOCKS, SET CURRENT SCHEMA, Client special registers, long SQL object names, SELECT from INSERT, UPDATE or DELETE, INSTEAD OF TRIGGER, Native SQL Procedure Language, BIGINT, file reference variables, XML, FETCH FIRST & ORDER BY in subselect and fullselect, caseless comparisons, INTERSECT, EXCEPT, not logged tables , range partitioning, compression Updateable UNION in Views, GROUPING SETS, ROLLUP, CUBE, 16 Built-in Functions, SET CURRENT ISOLATION, multi-site join, MERGE, MDC, XQuery z l u w c o m m o n
    17. 17. SQL: Productivity, DB2 family & porting <ul><li>XML </li></ul><ul><li>MERGE & TRUNCATE </li></ul><ul><li>SELECT FROM UPDATE, DELETE, MERGE </li></ul><ul><li>INSTEAD OF TRIGGER </li></ul><ul><li>BIGINT, VARBINARY, BINARY, DECIMAL FLOAT </li></ul><ul><li>Native SQL Procedure Language </li></ul><ul><li>Nested compound </li></ul><ul><li>Optimistic locking </li></ul><ul><li>LOB File reference variable & FETCH CONTINUE </li></ul><ul><li>FETCH FIRST & ORDER BY in subselect and fullselect </li></ul><ul><li>INTERSECT & EXCEPT </li></ul><ul><li>ROLE & trusted context </li></ul><ul><li>Many new built-in functions, caseless comparisons </li></ul><ul><li>Index on expression </li></ul><ul><li>Improved DDL consistency </li></ul><ul><li>CURRENT SCHEMA </li></ul>
    18. 18. Native SQL Procedural Language <ul><li>Eliminates generated C code and compilation </li></ul><ul><li>Fully integrated into the DB2 engine </li></ul><ul><li>Extensive support for versioning: </li></ul><ul><ul><li>VERSION keyword on CREATE PROCEDURE </li></ul></ul><ul><ul><li>CURRENT ROUTINE VERSION special register </li></ul></ul><ul><ul><li>ALTER ADD VERSION </li></ul></ul><ul><ul><li>ALTER REPLACE VERSION </li></ul></ul><ul><ul><li>ALTER ACTIVATE VERSION </li></ul></ul><ul><li>BIND PACKAGE with new DEPLOY keyword </li></ul>zIIP enabled
    19. 19. TRUNCATE Statement <ul><li>Allows fast delete of all rows in a given table ( segmented, partitioned or simple) </li></ul><ul><li>Very useful for nightly refresh of summary tables, warehouses, etc. </li></ul>TRUNCATE TABLE TABLE-NAME < DROP STORAGE | REUSE STORAGE> < RESTRICT WHEN DELETE TRIGGERS | IGNORE DELETE TRIGGERS> < IMMEDIATE>
    20. 20. Decimal Floating Point <ul><li>New datatype DECFLOAT </li></ul><ul><ul><li>Well suited to typical customer financial calculations </li></ul></ul><ul><ul><li>Similar to “calculator” mathematics </li></ul></ul><ul><ul><ul><li>Eliminates rounding errors by using base 10 math </li></ul></ul></ul><ul><ul><ul><li>Has up to 34 digits of precision </li></ul></ul></ul><ul><ul><ul><li>Floating point convenience with fixed point precision!!! </li></ul></ul></ul><ul><ul><li>Hardware support will be provided in the next System z processor generation (new IEEE standard) </li></ul></ul><ul><ul><ul><li>Software emulation provided for other models </li></ul></ul></ul>
    21. 21. MERGE <ul><li>Multi-row MERGE operation, using arrays </li></ul><ul><li>Targets OLTP applications like SAP </li></ul>MERGE INTO account AS T USING VALUES (:hv_id, :hv_amt) FOR 5 ROWS AS S(id,amt) ON T.id = S.id WHEN MATCHED THEN UPDATE SET balance = T.balance + S.amt WHEN NOT MATCHED THEN INSERT (id, balance) VALUES (S.id, S.amt) NOT ATOMIC CONTINUE ON SQLEXCEPTION
    22. 22. SQL Improvements – Family Compatibility <ul><li>INSTEAD OF triggers </li></ul><ul><li>SELECT FROM UPDATE </li></ul><ul><li>SELECT FROM DELETE </li></ul><ul><li>SELECT FROM MERGE </li></ul><ul><li>BIGINT, BINARY and VARBINARY data types </li></ul><ul><li>ORDER BY and FETCH FIRST in subselect </li></ul>
    23. 23. Text improvements in DB2 9 <ul><li>30 new & improved character functions </li></ul><ul><li>Index on expression: e.g. UPPER, COLLATION_KEY </li></ul><ul><li>LOB improvements </li></ul><ul><li>pureXML </li></ul><ul><li>Text search server </li></ul>
    24. 24. Text function improvements in DB2 9 <ul><li>New built-in character functions: COLLATION_KEY, </li></ul><ul><ul><li>ASCII_CHR, ASCII_STR, COLLATION_KEY </li></ul></ul><ul><ul><li>DIFFERENCE, EBCDIC_CHR, EBCDIC_STR </li></ul></ul><ul><ul><li>IS_IDENTICAL_TO, LEFT, LOCATE_IN_STRING </li></ul></ul><ul><ul><li>LPAD, NORMALIZE_STRING, OVERLAY </li></ul></ul><ul><ul><li>RIGHT, RPAD, SOUNDEX, UNICODE </li></ul></ul><ul><ul><li>UNICODE_STR, VARCHAR_FORMAT </li></ul></ul><ul><ul><li>XMLATTRIBUTES, XMLCOMMENT, </li></ul></ul><ul><ul><li>XMLDOCUMENT, XMLPARSE, XMLPI </li></ul></ul><ul><ul><li>XMLQUERY, XMLSERIALIZE, XMLTEXT </li></ul></ul><ul><li>Index on expression: e.g. UPPER, LOWER </li></ul>
    25. 25. Text Search Server (not included at general availability) <ul><li>Text search for CHAR, VARCHAR, CLOB & XML columns </li></ul><ul><li>Provide a text index server </li></ul><ul><li>Efficient communication interaction with DB2 for z/OS </li></ul><ul><li>Text indexes are persisted into DB2 tables for </li></ul><ul><li>backup & recovery purposes </li></ul>DB2 DB2 DB2 DB2 Parallel Sysplex Text Server Text Server TCP/IP
    26. 26. DB2 9 Spatial Support <ul><li>Enabling Open Geospatial Consortium (OGC) compliant geospatial applications </li></ul><ul><ul><li>Spatial data types </li></ul></ul><ul><ul><li>Spatial functions and predicates </li></ul></ul><ul><ul><li>Spatial indexes </li></ul></ul><ul><ul><li>Spatial search </li></ul></ul><ul><ul><li>OGC-compliant spatial catalog </li></ul></ul>Map GPS Address Street State …
    27. 27. DDL Porting Improvements <ul><li>Automatic selection of DATABASE and TABLESPACE when DDL omits these keywords </li></ul><ul><li>Automatic CREATE of UNIQUE index for PRIMARY KEY </li></ul><ul><li>Deprecated simple table space, default to segmented structure, partition by growth </li></ul>
    28. 28. Leverage Application Development Skills <ul><li>Use new converged SQL </li></ul><ul><li>Key Database Technologies </li></ul><ul><ul><li>SQL, SQL Procedures </li></ul></ul><ul><ul><li>XML </li></ul></ul><ul><ul><li>SOA, Web Services </li></ul></ul><ul><li>Developer communities </li></ul><ul><ul><li>COBOL, PL/I, REXX, C, C++, assembler, Fortran </li></ul></ul><ul><ul><li>Java (JDBC / SQLJ) </li></ul></ul><ul><ul><li>.NET (C#, VB .NET) </li></ul></ul><ul><ul><li>Open Source </li></ul></ul><ul><ul><ul><li>PHP </li></ul></ul></ul><ul><ul><ul><li>Perl </li></ul></ul></ul><ul><ul><ul><li>Python </li></ul></ul></ul><ul><ul><ul><li>Ruby on Rails </li></ul></ul></ul><ul><ul><ul><li>TOAD for DB2 </li></ul></ul></ul>
    29. 29. DB2 9 for z/OS Innovation: Data Warehousing <ul><li>Dynamic index ANDing for star schema </li></ul><ul><li>INTERSECT, EXCEPT </li></ul><ul><li>Query optimization improvements </li></ul><ul><li>Improved query performance </li></ul><ul><li>Index compression </li></ul><ul><li>Plan stability </li></ul><ul><li>Optimization Service Center </li></ul>
    30. 30. Query Enhancements <ul><li>SQL enhancements: INTERSECT, EXCEPT, cultural sort, caseless comparisons, FETCH FIRST in fullselect, OLAP specifications: RANK, DENSE_RANK, ROW_NUMBER … </li></ul><ul><li>pureXML integration and text improvements </li></ul><ul><li>Index improvements: Index on expression, Index compression, … </li></ul><ul><li>Improved Optimization statistics: Histogram </li></ul><ul><li>Optimization techniques </li></ul><ul><ul><li>Cross query block optimization and REOPT(AUTO) </li></ul></ul><ul><ul><li>Generalize sparse index & in-memory data cache method </li></ul></ul><ul><ul><li>Dynamic Index ANDing for Star Schema </li></ul></ul><ul><li>Analysis: instrumentation & Optimization Service Center </li></ul>
    31. 31. Modified SAP/BW Workload <ul><li>Data </li></ul><ul><ul><li>Populated with SAP benchmark BW 3.5 toolkits </li></ul></ul><ul><ul><li>Fact table size: 58.4M rows , 8 indexes </li></ul></ul><ul><ul><li>Dimension tables: 8 (2 ~ 99326 rows) </li></ul></ul><ul><ul><li>Snowflakes: 6 (added to increase query complexity) </li></ul></ul><ul><li>Queries (100) </li></ul><ul><ul><li>Developed by DB2 development and performance </li></ul></ul><ul><ul><li>Based on V8 BW workload </li></ul></ul><ul><ul><li>New queries added to better reflect the customer scenarios learned from the V8 service stream </li></ul></ul><ul><li>Represent customer workloads without adequate (multi-column) index support (this is the norm) </li></ul>
    32. 32. Performance Comparison -1.5 % 7514 7400 Total CPU time (seconds) 6775 (90%) 2924 (39.5%) CPU time eligible for zIIP off-load 88% 8544 71660 Total Elapse Time (seconds) Improvement DB2 9 DB2 V8
    33. 33. <ul><li>Optimization Service Center </li></ul><ul><li>Identify Problem Query </li></ul><ul><li>Tune Problem Query </li></ul><ul><li>Monitor & Capture Query Workload </li></ul><ul><li>Tune Query Workload </li></ul>
    34. 34. Intersect/Except/Union semantics INTERSECT EXCEPT (Difference) UNION R1 R1 R2 R2 * There are some variations and restrictions UNION ALL R1 R2 R1 R2
    35. 35. DB2 9 for z/OS Innovation: Cost <ul><li>Cost Savings through Optimization </li></ul><ul><ul><li>Security and Regulatory Compliance </li></ul></ul><ul><ul><li>Performance improvements </li></ul></ul><ul><ul><li>Synergy with System z </li></ul></ul><ul><ul><li>Query enhancements </li></ul></ul><ul><ul><li>Index Compression </li></ul></ul><ul><ul><li>LOB improvements </li></ul></ul><ul><ul><li>DDF improvements </li></ul></ul><ul><li>Reduced Complexity </li></ul>
    36. 36. DB2 9 for z/OS Performance Improvements <ul><li>Synergy with new hardware: zIIP, MIDAW, DS8000 </li></ul><ul><li>Significant CPU time reduction in most utilities </li></ul><ul><li>Performance/Scalability Enhancements </li></ul><ul><ul><li>Especially Insert / Update / Delete </li></ul></ul><ul><li>Query/Access Path Performance Enhancements Other Performance Enhancements </li></ul><ul><ul><li>Native SQL procedure, index compression, LOB, Varchar, … </li></ul></ul><ul><li>Improved virtual storage usage below bar </li></ul>
    37. 37. System z Synergy & DB2 9 <ul><li>System z9 Integrated Information Processor (zIIP) Enterprise Class & Business Class </li></ul><ul><li>Enhanced Cryptography </li></ul><ul><li>Channels (4 Gb & MIDAW) </li></ul><ul><li>Faster Processors </li></ul><ul><li>Up to 54 Processors EC </li></ul><ul><li>More memory, better value; 64 bit virtual storage </li></ul><ul><li>z/Architecture new instructions </li></ul><ul><li>Parallel Sysplex </li></ul><ul><li>IPv6 </li></ul><ul><li>SSL </li></ul><ul><li>Java </li></ul><ul><li>Decimal float </li></ul><ul><li>Backup & restore </li></ul><ul><li>Security </li></ul><ul><li>Unicode collation </li></ul><ul><li>Compression </li></ul><ul><li>System z Application Assist Processor (zAAP) </li></ul><ul><li>WLM enhanced … </li></ul>
    38. 38. Synergy with new I/O hardware <ul><li>DS8000 with Ficon Express and MIDAW </li></ul><ul><ul><li>(Modified Indirect Data Address Word) </li></ul></ul><ul><ul><li>MIDAW requires z9 (2094) and z/OS 1.6 or later </li></ul></ul><ul><ul><li>Sequential read throughput </li></ul></ul><ul><ul><ul><li>40MB/sec on ESS 800 69MB/sec with DS8000 </li></ul></ul></ul><ul><ul><ul><li>109MB/sec with DS8000 and MIDAW </li></ul></ul></ul><ul><ul><ul><li>138MB/sec with 2 stripes </li></ul></ul></ul><ul><ul><li>Bigger read, write, preformat quantity </li></ul></ul><ul><ul><ul><li>183MB/sec in sequential read with 2 stripes </li></ul></ul></ul><ul><ul><li>Similar for write </li></ul></ul><ul><ul><li>Performance gap between EF (Extended Format) and non EF datasets or 4K and bigger page practically gone </li></ul></ul>
    39. 39. Utility CPU time reduction up to – * primarily from index processing <ul><li>10% to 20% in Copy, Recover table space / index* </li></ul><ul><li>5% to 30% in Load*, Reorg*, Rebuild Index* </li></ul><ul><li>20% to 60% in Check Index* </li></ul><ul><li>35% in Load Partition* </li></ul><ul><li>30% to 40% in Runstats Index* </li></ul><ul><li>40% to 50% in Reorg Index* </li></ul><ul><li>70% in Load Replace Partition with dummy input* </li></ul>
    40. 40. Index Changes <ul><li>INDEX on expression </li></ul><ul><li>Page sizes 8K, 16K, 32K </li></ul><ul><li>Improved page split </li></ul><ul><li>Index compression </li></ul><ul><li>Online REBUILD INDEX </li></ul><ul><li>REORG without BUILD2 – not just for DPSI </li></ul><ul><li>Randomized index key </li></ul><ul><li>Not logged index space </li></ul><ul><li>XML index </li></ul>
    41. 41. Index Compression <ul><li>Compression of indexes for BI workloads </li></ul><ul><ul><li>Indexes are often larger than tables in BI </li></ul></ul><ul><li>Solution provides page-level compression </li></ul><ul><ul><li>Data is compressed to 4K pages on disk </li></ul></ul><ul><ul><li>8K, 16K or 32K pages results in 2x, 4X or 8x disk savings </li></ul></ul><ul><ul><li>No compression dictionaries – compression on the fly </li></ul></ul>
    42. 42. Index Compression : Differences between data and index compression 25% - 75% (3) 10% - 90% Average Comp Ratio No (2) Yes Comp Dictionary No Yes Comp in Log No Yes Comp in Buffer Pool Yes Yes Comp on disk Page (1) Row Level Index Data
    43. 43. Relief for Sequential Key INSERT <ul><li>New page sizes: 8K, 16K, 32K for INDEX pages </li></ul><ul><ul><li>Fewer page splits for long keys </li></ul></ul><ul><ul><li>More key values per page </li></ul></ul><ul><li>INSERT at the end of the key range used to result in 50% free space in each index page </li></ul><ul><ul><li>Enhanced support dynamically adapts page split boundary to minimize wasted space in index pages </li></ul></ul><ul><li>Index key randomization </li></ul>
    44. 44. CREATE TABLE … APPEND(YES) <ul><li>New APPEND option: </li></ul><ul><ul><li>Maximizes performance for “INSERT at end” </li></ul></ul><ul><ul><li>Avoids overhead of attempting to preserve clustering sequence </li></ul></ul><ul><ul><li>CREATE or ALTER table </li></ul></ul>
    45. 45. Varchar Performance Improvement <ul><li>Old tuning recommendation for rows with many columns with any varchar present </li></ul><ul><ul><li>V9 DB2 internally executes this recommendation and more </li></ul></ul><ul><li>2 times or more improvement observed when many rows with many varchars are scanned and/or fetched using many predicates </li></ul><ul><li>No difference if no varchar , Under 5% improvement for a typical online transaction </li></ul><ul><li>Reorg with rebuild compression dictionary if varchar columns when migrating to V9 </li></ul>F1 F2 V3 F4 F5 V6
    46. 46. LOB Function <ul><li>SQL </li></ul><ul><ul><li>File reference variables </li></ul></ul><ul><ul><li>FETCH CONTINUE </li></ul></ul><ul><ul><li>Automatic object creation </li></ul></ul><ul><li>Utilities </li></ul><ul><ul><li>REORG reclaim fragmented space and improve access performance </li></ul></ul><ul><ul><li>REORG share level reference (read only) </li></ul></ul><ul><ul><li>Online CHECK LOB & CHECK DATA </li></ul></ul><ul><ul><li>Sample unload DSNTIAUL </li></ul></ul>
    47. 47. LOB Performance/Scalability <ul><li>LOB lock avoidance – LRSN and page latching is used instead for consistency checks </li></ul><ul><li>New network flows for delivering LOBs </li></ul><ul><ul><li>JDBC, SQLJ, and CLI will let server determine whether to flow LOB values or LOCATORs based on size thresholds </li></ul></ul><ul><ul><li>Significant reduction in network traffic </li></ul></ul><ul><ul><li>Greatly reduces frequency of FREE LOCATOR statements </li></ul></ul>
    48. 48. DDF Improvements <ul><li>64-bit addressing by DDF </li></ul><ul><ul><li>Special “shared private” with xxxDBM1 to eliminate many data moves on SQL operations </li></ul></ul><ul><li>Support for IPv6 and SSL </li></ul><ul><li>VTAM definition is now optional </li></ul><ul><li>Prepare for elimination of PRIVATE protocol requester </li></ul><ul><ul><li>Includes tools for identifying which packages need to be bound at remote servers </li></ul></ul>
    49. 49. IDC : Since 2000, Labor Costs Have Exceeded the Cost of All Servers … and are Still Growing Source: IDC Cost of mgmt. & admin. 10% CAGR New server spending (USM$) 3% CAGR Spending (USB$) Installed Base (M Units) $ 0 $20 $40 $60 $80 $100 $120 $140 $160 $180 $200 1996 ’ 97 ’ 98 ’ 99 2000 ’ 01 ’ 02 ’ 03 ’ 04 ’ 05 ’ 06 ’ 07 ’ 08 5 10 15 20 25 30 35
    50. 50. Converged TEMP Space <ul><li>Single source for all temporary space in DB2, replacing: DSNDB07, temp databases, workfile database </li></ul><ul><li>Access is virtualized for small amounts of data, eliminating cost of work file creation (reduced CPU and I/O) </li></ul><ul><li>Supports 4K and 32K page sizes, with automatic selection of the appropriate page size </li></ul><ul><li>New Zparm for preventing workfile monopolization </li></ul>
    51. 51. Volume-based COPY/RECOVER <ul><li>FlashCopy technology used to capture entire content of disk volumes </li></ul><ul><li>RECOVER modified to enable object-level recovery from volume FlashCopy </li></ul><ul><ul><li>Restore assumes that the object has not moved volumes </li></ul></ul><ul><li>Eliminates labor associated with setting up COPY jobs for each database / table space </li></ul><ul><li>Full integration of tape into BACKUP/RESTORE SYSTEM utilities </li></ul>
    52. 52. WLM assisted buffer pool management BP0 DBM1 WLM Data Collection DB2 Periodic Report Buffer Pool Sizes Hit Ratio for Random Reads BP1 BP7 BP2 1 Plots size and hit ratio over time. 2 Projects effects of changing the size Bufferpool Adjustment + - 25% ALTER BUFFERPOOL AUTOSIZE option
    53. 53. V9 Modes – An Overview CM Compatibility Mode - This is the mode DB2 is in when V9 is started for the first time from V8. It will still be in CM when migration job DSNTIJTC has completed. No new function can be executed in CM. Data sharing systems can have V8 and V9 members in this mode. DB2 can only migrate to CM from V8 NFM. ENFM Enabling New Function Mode - This mode is entered when CATENFM START is executed (the first step of job DSNTIJEN). DB2 remains in this mode until all the enabling functions are completed. Data sharing systems can only have V9 members in this mode. NFM New Function Mode - This mode is entered when CATENFM COMPLETE is executed (the only step of job DSNTIJNF). This mode indicates that all catalog changes are complete and new function can be used. ENFM* This is the same as ENFM but the * indicates that at one time DB2 was at NFM. Objects that were created when the system was at NFM can still be accessed but no new objects can be created. When the system is in ENFM* it can not fallback to V8 or coexist with a V8 system. CM* This is the same as CM but the * indicates that at one time DB2 was at a higher level. Objects that were created at the higher level can still be accessed. When DB2 is in CM* it can not fallback to V8 or coexist with a V8 system.
    54. 54. DB2 9 Vstor Constraint Relief <ul><li>DDF address space runs in 64-bit addressing mode </li></ul><ul><ul><li>Shared 64-bit memory object avoids xmem moves between DBM1 and DDF and improves performance </li></ul></ul><ul><ul><li>Constraint relief </li></ul></ul><ul><li>DBM1, the following are moved above the bar in V9 </li></ul><ul><ul><li>Parse trees </li></ul></ul><ul><ul><li>EDM fixed pools </li></ul></ul><ul><ul><li>SKPTs / SKCTs (primarily static SQL). Also part of CTs/PTs </li></ul></ul><ul><ul><li>Pageset blocks, RTS blocks </li></ul></ul><ul><ul><li>Local SQL statement cache </li></ul></ul><ul><ul><li>Some thread-related storage </li></ul></ul><ul><li>For installations that are constrained on DBM1 vstor: </li></ul><ul><ul><li>200 to 300MB or more of savings expected </li></ul></ul><ul><ul><li>Mainly from EDM related storage (static SQL) and dynamic statement cache (dynamic SQL) </li></ul></ul>
    55. 55. Other cost of ownership improvements <ul><li>Resource Limit Facility enhanced to allow CPU cost to be controlled based on: </li></ul><ul><ul><ul><li>Client workstation / app name </li></ul></ul></ul><ul><ul><ul><li>Client userid, IP address </li></ul></ul></ul><ul><li>SMS integration </li></ul><ul><li>Utilities template switching </li></ul><ul><li>RENAME SCHEMA, VCAT </li></ul><ul><li>REOPT(AUTO) </li></ul><ul><li>Command line processor </li></ul><ul><li>Optimization Service Center </li></ul>
    56. 56. DB2 9 for z/OS Innovation: SOA and XML <ul><li>Integration with WebSphere </li></ul><ul><li>Native XML data type, hybrid data base server </li></ul>
    57. 57. Optimistic Locking Support <ul><li>Built-in timestamp for each row or page </li></ul><ul><ul><li>Automatically updated by DB2 </li></ul></ul><ul><ul><li>Allows simple timestamp predicate to validate that row has not changed since last access </li></ul></ul><ul><li>Eliminates need for complex predicates on WebSphere CMP updates, improves performance </li></ul>
    58. 58. B2B Web Services xml xml xml xml xml xml xml xml xml xml DB2 9 for z/OS … Powering SOA Solutions XML is the Key Link Life Insurance YTP Pensions ITP Pensions Investments Banking Mortgage Business Services Call Center Data Warehouse Process Mgmt. Customer Data Server Web Services Private Portal 3 rd Party portal Corporate Sales system Private Sales system Corporate Portal Clients Systems xml xml
    59. 59. Contrasting the Models XML and Relational <ul><li>XML </li></ul><ul><ul><li>Strength: Semi-structured, frequently changing data </li></ul></ul><ul><ul><ul><li>Self-describing, flexible schema </li></ul></ul></ul><ul><ul><ul><li>Easily modified format </li></ul></ul></ul><ul><ul><li>Strength: Retrieving sequences </li></ul></ul><ul><ul><ul><li>Documents, subdocuments, related documents </li></ul></ul></ul><ul><li>Relational </li></ul><ul><ul><li>Strength: Static data </li></ul></ul><ul><ul><ul><li>Strict schema ensures data integrity </li></ul></ul></ul><ul><ul><ul><li>High performance indexing on fixed data </li></ul></ul></ul><ul><ul><li>Strength: ‘Set-based’ data </li></ul></ul><ul><ul><ul><li>Multiple results returned </li></ul></ul></ul><ul><ul><ul><li>Retrieving rows </li></ul></ul></ul>Over $20B Annual Customer Technology Investment in RDB Alone… XML database investments growing twice as fast as total database investment…
    60. 60. XML Data Needs Relational Maturity Complementing XML Processing <ul><li>XML Data Needs Protection </li></ul><ul><ul><li>Backup and recovery features to ensure continuity </li></ul></ul><ul><ul><li>Data is protected using database security </li></ul></ul><ul><li>Simplified XML Data Access </li></ul><ul><ul><li>Centrally store and access difficult to retrieve data </li></ul></ul><ul><ul><li>SQL or XPath can be used to retrieve data </li></ul></ul><ul><ul><li>Join XML data with it’s related relational data </li></ul></ul><ul><li>Search Speed </li></ul><ul><ul><li>Search documents quickly and efficiently using proven search optimization engine of mature database </li></ul></ul><ul><li>Optimize Existing Investments </li></ul><ul><ul><li>Use existing technology infrastructure and skills to store and manage both relational and XML </li></ul></ul>
    61. 61. pureXML <ul><li>Support XML data type </li></ul><ul><li>Store the XML document natively </li></ul><ul><li>DDL -- </li></ul><ul><ul><li>CREATE/ALTER Table with XML type column </li></ul></ul><ul><ul><ul><li>Implicitly create XML Auxiliary objects (tablespace/table/index) - one per XML column </li></ul></ul></ul><ul><ul><li>Index support </li></ul></ul><ul><ul><ul><li>Created by users </li></ul></ul></ul><ul><ul><ul><li>uses XPath to determine which nodes in the XML document to index. </li></ul></ul></ul><ul><ul><ul><li>CREATE INDEX dependentName ON deptTable(deptDocs) </li></ul></ul></ul><ul><ul><ul><li>GENERATE KEY USING XMLPATTERN '/department/empl/dependent/name' ATOMIC AS SQL VARCHAR(20); </li></ul></ul></ul><ul><li>INSERT/UPDATE/DELETE </li></ul><ul><ul><li>INSERT with VALUES and SUBSELECT </li></ul></ul><ul><ul><li>No Subdocument update </li></ul></ul>
    62. 62. pureXML -- Query <ul><li>Enhanced V8 XML Constructors (XML Publishing Functions) </li></ul><ul><li>SQL/XML Functions and Predicates </li></ul><ul><ul><li>XMLParse - Convert a XML text to XML value </li></ul></ul><ul><ul><li>XMLSerialize - Converts XML to character type </li></ul></ul><ul><ul><li>XMLQuery - executes an XPath expression against an XML value. </li></ul></ul><ul><ul><ul><li>SELECT XMLQUERY ( '//item[USPrice = $price] ' PASSING PO.POrder, </li></ul></ul></ul><ul><ul><ul><li>T.price AS “price”) FROM PurchaseOrders PO, T; </li></ul></ul></ul><ul><ul><li>XMLCast - Cast XML to other types or other types to XML </li></ul></ul><ul><ul><li>XMLExists - a predicate, which returns TRUE if the XPath expression evaluates to a non-empty sequence </li></ul></ul><ul><li> SELECT PO.pid FROM PurchaseOrders PO, T </li></ul><ul><li> WHERE XMLEXISTS( '//item[USPrice = $price] ' PASSING PO.POrder, T.price AS “price”) </li></ul>
    63. 63. pureXML <ul><li>XPATH supported features from XPath 2.0: </li></ul><ul><li>Utility Support </li></ul><ul><ul><li>LOAD/UNLOAD, CHECK DATA/INDEX, COPY, REBUILD, RECOVER, REORG, etc. </li></ul></ul><ul><li>XML Schema Support </li></ul><ul><ul><li>XSR – XML Schema Repository </li></ul></ul><ul><ul><li>Tables to store XML schemas </li></ul></ul><ul><ul><li>Stored procedures to register XML schemas </li></ul></ul><ul><li>DSN_XMLVALIDATE() SQL/XML function </li></ul><ul><ul><li>Test XML values for validity against XML schema </li></ul></ul><ul><ul><li>Obtain default values and schema normalized values from XML schema </li></ul></ul><ul><li>XML decomposition using annotated XML schema </li></ul>
    64. 64. API Support <ul><li>XML type is supported in </li></ul><ul><ul><li>Java (JDBC, SQLJ), ODBC, </li></ul></ul><ul><ul><li>C/C++, COBOL, PL/I, Assembly </li></ul></ul><ul><ul><li>.NET </li></ul></ul><ul><li>Applications use: </li></ul><ul><ul><li>XML as CLOB(n) </li></ul></ul><ul><ul><li>XML as DBCLOB(n) </li></ul></ul><ul><ul><li>XML as BLOB(n) </li></ul></ul><ul><ul><li>All character or binary string types are supported </li></ul></ul><ul><li>XMLParse and XMLSerialize apply (implicitly or explicitly) </li></ul>
    65. 65. When To Use XML? <ul><li>Sparsely populated data </li></ul><ul><li>Frequent DDL changes </li></ul><ul><li>Short term complex data </li></ul><ul><li>Complex snapshot data </li></ul><ul><li>Relatively static data that is not frequently updated </li></ul><ul><li>Data which is not frequently referenced on WHERE predicates and not frequently updated </li></ul>Tedious normalization and frustrated changes of schema are an indicator for using native XML.
    66. 66. Example1: Auto Insurance Policy Variations <ul><li>Each vehicle has many different features, and insured may choose different policy variations </li></ul><ul><li>New features may come up each model year, and new policy variations can come up too. </li></ul><ul><li>It’s hard to design a set of columns to cover all possible features and variations </li></ul><ul><li>Some of the features and variations need to be searched upon </li></ul><ul><li>Solution: use XML column </li></ul>
    67. 67. Example2: Customer Statements <ul><li>Customer statements get generated in XML format </li></ul><ul><li>XML file is used to print/mail to customer </li></ul><ul><li>XML documents are tagged with keywords </li></ul><ul><li>XML document can be stored natively in DB2 </li></ul><ul><li>Keywords are searched to respond to customer inquiries </li></ul><ul><li>Able to easily recreate the original document sent to customers (no transformations needed) </li></ul><ul><li>Instead of side table and CLOB, use XML and indexing on the tagged keywords </li></ul><ul><li>Benefit: flexible, high performance </li></ul>
    68. 68. DB2 9 – Summary of pureXML Support <ul><li>XML as a native data type </li></ul><ul><li>Pure XML storage and indexing </li></ul><ul><li>SQL/XML and XPath support </li></ul><ul><li>Integration with traditional relational data </li></ul><ul><li>XML Schema Repository </li></ul><ul><li>Schema validation </li></ul><ul><li>Application Support (Java, C/C++, .NET, PHP, COBOL, PL/1 etc.) </li></ul><ul><li>Visual Tooling, Control Center Enhancements </li></ul><ul><li>DB2 Utilities: Load, Unload, Reorg, etc. </li></ul><ul><li>… and more </li></ul>Secure and Resilient Infrastructure for a New Breed of Agile Applications DB2 9
    69. 69. DB2 9 for z/OS Innovation: Continuous Availability <ul><li>Online schema evolution </li></ul><ul><li>More online utilities </li></ul><ul><li>Data sharing enhancements </li></ul>
    70. 70. Schema Evolution – Database Definition On Demand <ul><li>Fast replacement of one table with another </li></ul><ul><li>Rename column and index </li></ul><ul><li>Rename SCHEMA and VCAT </li></ul><ul><li>Table space that can add partitions, for growth </li></ul><ul><li>Improve ability to rebuild an index online </li></ul><ul><li>Online reorganization with no BUILD2 phase </li></ul><ul><li>Modify early code without requiring an IPL </li></ul><ul><li>Alter table space and index logging </li></ul><ul><li>Create & alter STOGROUP SMS constructs </li></ul>
    71. 71. CLONE Tables <ul><li>Allows fast replacing production data without renames and rebinds </li></ul><ul><ul><li>A capability to support online load replace </li></ul></ul><ul><li>ALTER TABLE to create a Clone Table </li></ul><ul><ul><li>All indexes are also cloned </li></ul></ul><ul><ul><li>Table and Index data are not copied </li></ul></ul><ul><ul><li>Base and Clone tables share the same table space and index names </li></ul></ul><ul><ul><li>Underlying data sets are differentiated by a data set instance number </li></ul></ul>
    72. 72. CLONE Tables… <ul><li>A clone table can only be created </li></ul><ul><ul><li>On a single table in a table space (partitioned or non-partitioned) </li></ul></ul><ul><ul><li>No RI or Trigger on the base table </li></ul></ul><ul><li>Use insert or load to populate clone tables </li></ul><ul><li>Utilities (except RUNSTATS) can operate on clone tables with a new CLONE keyword </li></ul>
    73. 73. Partition by Growth <ul><li>New partitioning scheme: </li></ul><ul><ul><li>Single table tablespace, where each partition contains a segmented pageset (allows segmented to increase from 64GB to 16TB or 128 TB with 32K pages) </li></ul></ul><ul><ul><li>Eliminates need to define partitioning key and assign key ranges </li></ul></ul><ul><ul><li>A new partition is created when a given partition reaches DSSIZE (defaults to 64G) </li></ul></ul><ul><ul><li>Retains benefits of Utilities and SQL parallelism optimizations for partitioned tables </li></ul></ul>
    74. 74. DB2 9 Utilities <ul><li>Support for all new functions in DB2 Version 9 for z/OS product (universal table spaces, XML, not logged, etc.) </li></ul><ul><li>More online utilities </li></ul><ul><ul><li>Rebuild Index SHRLEVEL CHANGE </li></ul></ul><ul><ul><ul><li>Great for building new non-unique indexes </li></ul></ul></ul><ul><ul><li>Reorg enhancements </li></ul></ul><ul><ul><ul><li>Reorg LOB now supports SHRLEVEL REFERENCE </li></ul></ul></ul><ul><ul><ul><li>LOB space reclamation </li></ul></ul></ul><ul><ul><ul><li>Partition-level capabilities (not available with REBALANCE) </li></ul></ul></ul><ul><ul><ul><ul><li>Partition parallelism (UNLOAD/RELOAD) in a single utility statement </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Elimination of the BUILD2 phase outage </li></ul></ul></ul></ul><ul><li>Recover to consistent PIT without need for a quiesce </li></ul>
    75. 75. DB2 9 Utilities <ul><li>More online utilities </li></ul><ul><ul><li>Check data, LOB and repair locate … SHRLEVEL CHANGE </li></ul></ul><ul><ul><li>Check index SHRLEVEL REFERENCE supports parallel for > 1 index </li></ul></ul><ul><ul><li>Load replace (shrlevel change) with CLONE TABLE function </li></ul></ul><ul><li>Always perform CHECKPAGE on the COPY utility </li></ul><ul><ul><li>Prior to V9, CHECKPAGE was optional, with about ~5% CPU overhead, and if a broken page was encountered (DSNU441I for space maps or DSNU518I for others, both RC8), then copy-pending was set </li></ul></ul><ul><ul><li>Now, COPY always performs these checks (with reduced overall CPU!) and no longer sets copy-pending, so…. Check those RCs! </li></ul></ul><ul><ul><li>A new SYSCOPY record type is written if a broken page is detected to force a full image next since dirty bits may have already been flipped off in the space map pages </li></ul></ul>
    76. 76. DB2 9 Data Sharing Enhancements <ul><li>Restart performance enhancements </li></ul><ul><ul><li>Reduced impact of retained locks </li></ul></ul><ul><ul><li>Open datasets ahead of log apply </li></ul></ul><ul><ul><li>Avoid locks that have potential of ‘hanging’ restart </li></ul></ul><ul><li>Auto-recover GRECP/LPL objects on group restart </li></ul><ul><ul><li>Useful in Disaster Recovery or GDPS scenarios </li></ul></ul><ul><li>Index performance improvements </li></ul><ul><ul><li>Option to randomize key </li></ul></ul><ul><ul><li>Sequential key insert performance improvement </li></ul></ul><ul><ul><li>Larger page size, index compression </li></ul></ul><ul><li>Use of DPSI to avoid data sharing overhead </li></ul><ul><ul><li>DPSI can be unique within partition </li></ul></ul><ul><li>Detect use of unused indexes so they can be dropped </li></ul><ul><ul><li>Reduces data sharing overhead </li></ul></ul>
    77. 77. DB2 9 Data Sharing Enhancements… <ul><li>Command to remove GBP-dependency at object level </li></ul><ul><ul><li>ACCESS DB MODE(NGBPDEP) </li></ul></ul><ul><ul><li>Typical usage would be before batch run </li></ul></ul><ul><ul><li>Issue on the member on which you plan to run batch </li></ul></ul><ul><li>Command to “prime” open dataset </li></ul><ul><ul><li>START DB MODE(OPEN) [PART] </li></ul></ul><ul><li>Log latch contention relief (avoid LRSN “spinning”) </li></ul><ul><li>Improved performance for GBP writes </li></ul><ul><ul><li>Avoid copying pages for batched writes </li></ul></ul><ul><li>DB2 overall health taken into account for WLM routing </li></ul><ul><li>Balance group attach connections across multiple members on same LPAR (V7, V8 usermod) </li></ul>
    78. 78. Beyond DB2 9 <ul><li>Total Cost of Ownership (improve performance and productivity) </li></ul><ul><li>Availability improvements (fewer planned outages) </li></ul><ul><li>Data warehousing improvements </li></ul><ul><li>Application development and application portability (SQL, XML) </li></ul><ul><li>Ease migration </li></ul>
    79. 79. Click to edit Master title style References
    80. 80. DB2 9 for z/OS References Main DB2 for z/OS web page: pointers to most of the following: http://www.ibm.com/software/data/db2/zos/index.html V9 beta announcement: http://www.ibm.com/common/ssi/rep_ca/8/897/ENUS206-098/ENUS206-098.PDF http://www.ibm.com/common/ssi/fcgi-bin/ssialias?subtype=ca&infotype=an&appname=iSource&supplier=897&letternum=ENUS206-098 DB2 9 for z/OS main page: http://www.ibm.com/software/data/db2/zos/db2zosv91.html Overview presentation, webcast and foils with notes: http://www.ibm.com/software/os/zseries/webcast/18may/ ftp://ftp.software.ibm.com/software/data/db2zos/DB2V9zOS.pdf Redbooks including V9: SOA book - overview of V9 XML http://www.redbooks.ibm.com/abstracts/SG247259.html?Open LOBs book http://www.redbooks.ibm.com/abstracts/SG247270.html?Open Security book coming soon SG24-6480 V9 Technical Overview coming SG24-7330 V9 Performance Topics coming Detailed presentations: Start on the Events page http://www.ibm.com/software/data/db2/zos/events.html Click on Presentations from previous conferences. Sort results by date - newest first. Access the ftp site directly: ftp://ftp.software.ibm.com/software/data/db2zos/ About 15 of more than 200 presentations that address V9. Presentations from IOD conference, IDUG, Share, … See notes below for more detail.
    81. 81. IBM Information On Demand 2007 ibm.com/events/InformationOnDemand <ul><li>The premier information management event for business and IT executives, managers, professionals, DBA's and developers. </li></ul><ul><li>Select from over 600 sessions: a 2-1/2 day business leadership track with 120 sessions and a 5 day technical track with 480 sessions. </li></ul><ul><li>Latest strategy and product announcements </li></ul><ul><li>Huge Expo Solution Center </li></ul><ul><li>Hands on labs </li></ul><ul><li>One-on-ones with executives and specialists </li></ul><ul><li>Birds of a Feather roundtables </li></ul>IBM Information On Demand 2007 October 14-19, 2007 Las Vegas, Nevada <ul><li>Participate in discussions on the future of Information Management </li></ul><ul><li>Experience the entire Information Management portfolio – Software, Hardware and Services </li></ul><ul><li>Learn how to unlock business value and drive competitive advantage </li></ul><ul><li>Hear how your peers are realizing ROI </li></ul><ul><li>Understand the roadmap to long term strategic advantage </li></ul><ul><li>Learn best practices in your industry </li></ul><ul><li>Receive the best in technical education and free certification testing </li></ul><ul><li>Extensive opportunities for networking with both your peers and industry experts </li></ul>Why attend:
    82. 82. References <ul><li>DB2 for z/OS home page </li></ul><ul><ul><li>http://www.ibm.com/software/data/db2/zos/index.html </li></ul></ul><ul><li>DB2 for z/OS and OS/390 Version 7 Performance Topics, SG24-6129 </li></ul><ul><li>DB2 for z/OS Version 8 Performance Topics, SG24-6465 </li></ul><ul><li>DB2 for z/OS and OS/390 Version 7: Using the Utilities Suite, SG24-6289 </li></ul><ul><li>DB2 Magazine Fall 1998 - DB2 OS/390 Online Reorganization </li></ul><ul><ul><li>http://www.db2mag.com/db_area/archives/1998/q3/98fextra.shtml </li></ul></ul><ul><li>DB2 Magazine Quarter 2, 2003 - Programmer's Only - Programs vs. Utilities </li></ul><ul><ul><li>http://www.db2mag.com/db_area/archives/2003/q2/programmers.shtml </li></ul></ul><ul><li>Implementing Online Reorg in a Production Environment </li></ul><ul><ul><li>http://www.ibm.com/software/data/db2/os390/pdf/oreorg.pdf </li></ul></ul><ul><li>Moving Data Across the DB2 Family, SG24-6905 </li></ul><ul><li>Recommendations for Tuning Large DFSORT Tasks </li></ul><ul><ul><li>http://www.ibm.com/servers/storage/support/software/sort/mvs/tuning/index.html </li></ul></ul><ul><li>DFSMSrmm SMS ACS Support </li></ul><ul><ul><li>http://www.redbooks.ibm.com/abstracts/TIPS0530.html?Open </li></ul></ul>
    83. 83. <ul><li>Information center </li></ul><ul><ul><li>http:// publib.boulder.ibm.com/infocenter/dzichelp/index.jsp </li></ul></ul><ul><li>Information roadmap </li></ul><ul><ul><li>http://ibm.com/software/db2zos/roadmap.html </li></ul></ul><ul><li>DB2 for z/OS library page </li></ul><ul><ul><li>http://ibm.com/software/db2zos/library.html </li></ul></ul><ul><li>Examples trading post </li></ul><ul><ul><li>http://ibm.com/software/db2zos/exHome.html </li></ul></ul><ul><li>DB2 for z/OS support </li></ul><ul><ul><li>http://ibm.com/software/db2zos/support.html </li></ul></ul><ul><li>Official Introduction to DB2 for z/OS </li></ul><ul><ul><li>http:// ibm.com /software/data/education/bookstore </li></ul></ul>DB2 for z/OS information resources Take advantage of the following information resources available for DB2 for z/OS:
    84. 84. 1 1 Information contained in this material has not been submitted to any formal IBM review and is distributed on &quot;as is&quot; basis without any warranty either expressed or implied. Measurements data have been obtained in laboratory environment. Information in this presentation about IBM's future plans reflect current thinking and is subject to change at IBM's business discretion. You should not rely on such information to make business plans. The use of this information is a customer responsibility. IBM MAY HAVE PATENTS OR PENDING PATENT APPLICATIONS COVERING SUBJECT MATTER IN THIS DOCUMENT. THE FURNISHING OF THIS DOCUMENT DOES NOT IMPLY GIVING LICENSE TO THESE PATENTS. TRADEMARKS: THE FOLLOWING TERMS ARE TRADEMARKS OR ® REGISTERED TRADEMARKS OF THE IBM CORPORATION IN THE UNITED STATES AND/OR OTHER COUNTRIES: AIX, AS/400, DATABASE 2, DB2, e-business logo, Enterprise Storage Server, ESCON, FICON, OS/390, OS/400, ES/9000, MVS/ESA, Netfinity, RISC, RISC SYSTEM/6000, iSeries, pSeries, xSeries, SYSTEM/390, IBM, Lotus, NOTES, WebSphere, z/Architecture, z/OS, zSeries, System z, pureXML The FOLLOWING TERMS ARE TRADEMARKS OR REGISTERED TRADEMARKS OF THE MICROSOFT CORPORATION IN THE UNITED STATES AND/OR OTHER COUNTRIES: MICROSOFT, WINDOWS, WINDOWS NT, ODBC, WINDOWS 95 For additional information see ibm.com/legal/copytrade.phtml Disclaimer and Trademarks

    ×