Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Using XA for Batch – Bad idea? (article)


Published on

Oracle Scene - Autumn 13, Issue 51 - October 7, 2013

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Using XA for Batch – Bad idea? (article)

  1. 1. OracleSceneServing the Oracle Community Autumn 13 Issue 51 Creating Statistics For a Database? Histograms in 12c Advanced Analytics in Oracle 12c Understand the Oracle Data Miner tool and new in-database features The Tools of Fusion: Oracle JDeveloper and Oracle ADF Keep your skills up to date An independent publication not affiliated with Oracle Corporation This edition’s sponsors: Welcome to the future of 12c
  2. 2. At VIP Apps Consulting we appreciate the need to maximise your current investment in your business systems. Using our AMOBI methodology, we can help you find innovative ways to improve and optimise your existing systems and processes in order to realise additional value from your investment. Visit to explore how we can help your business work up a sweat. SWEAT YOUR SYSTEM • How hard are your Business Systems working for you? • Are your people and business processes in harmony with your IT solutions? value.innovation.process Are you really working your Oracle Investment?
  3. 3. Inside this issue Oracle Scene Editorial Team Editor Geoff Swaffer Email: Deputy Editor Brendan Tierney Email: UKOUG Contact Brigit Wells Email: Sales Isabella Grantham Email: Project Manager Lavinia Mildwater Email: UKOUG Governance A full listing of Board members, along with details of how the user group is governed, can be found at: UKOUG Office UK Oracle User Group, User Group House, 591-593 Kingston Road, Wimbledon London, SW20 8SA Tel: +44 (0)20 8545 9670 Email: Web: Produced and Designed by Why Creative Tel: +44 (0)7900 246400 Web: Future Oracle Scene issues Issue 52 Content deadline: 17th January 2014 Publish month: March 2014 Photography Pages 05 & 06: Liesbeth Verdegaal Page 11: Steve Walke 08 TECH13 - THE NEXT BIG THING 15 32STREAMLINING EPM DEPLOYMENT by Jennifer Toomey 38LEARNING THE FUSION TECHNOLOGY STACK by Grant Ronald OracleScene© UK Oracle User Group Ltd The views stated in Oracle Scene are the views of the author and not those of the UK Oracle User Group Ltd. We do not make any warranty for the accuracy of any published information and the UK Oracle User Group will assume no responsibility or liability regarding the use of such information. All articles are published on the understanding that copyright remains with the individual authors. The UK Oracle User Group reserves the right, however, to reproduce an article, in whole or in part, in any other user group publication. The reproduction of this publication by any third party, in whole or in part, is strictly prohibited without the express written consent of the UK Oracle User Group. Oracle is a registered trademark of Oracle Corporation and /or its affiliates, used under license. This publication is an independent publication, not affiliated or otherwise associated with Oracle Corporation. The opinions, statements, positions and views stated herein are those of the author(s) or publisher and are not intended to be the opinions, statements, positions, or views of Oracle Corporation. TECHNOLOGY Fine Tuning your Data Stack for Success by Dermot O’Kelly 12 Row Limiting, PL/SQL WITH SQL, and Temporal Validity: Three in 12c by Melanie Caffrey 18 Drive Value From Your ERP with a Post Implementation System Assessment (PISA) by Dale Kaplan 24 New Features of Oracle Advanced Analytics in Oracle 12c by Brendan Tierney 28 APPLICATIONS TRANSFORMATION Achieving Procurement Excellence Through Oracle Fusion by Jaskirat Singh & Anurag Kalita 44 EVENTS Tech13: New Conference, New Location: Manchester 10 ADVERTORIALS Cognizant: Redefining Enterprise Transformations 23 VIP Apps: Maximise your Return on (Oracle) Investment (ROI) 40 REGULAR FEATURES News and Reviews 06 UKOUG Supplier Guide 45 03 AUTUMN 13 Welcome to Oracle Scene HISTOGRAMS IN 12c by Jonathan Lewis Follow us @UKOUG OracleScene D I G I T A L View additional content and get more from your Oracle Scene at: Usability – Ignored by Developers and Undervalued by Managers 48 Getting the Most Out of My Oracle Support 52 What Does It Take to be Part of The IT Gold Rush? 55 Event Driven Architecture using Fusion Middleware 59 Three Real Reasons to Run Database Applications on Flash Storage 63 Book Review of Getting Started with Oracle Event Processing 11g 65 Using XA for Batch – Bad Idea? 67 Storage Performance: It’s All About Balance… 70 Top Ten Secrets of Successful Oracle ADF Projects 73 How the Oracle DML Logging Feature can be used within Oracle Data Integrator Interfaces and Knowledge Modules 77
  4. 4. 04 AUTUMN 13 Welcome to Oracle Scene OracleScene First word You may have noticed over the past few months that Oracle has released version 12c of their database. With this release we have a new architecture and lots of new or enhanced features. In this edition of Oracle Scene, we have a number of articles that look at some of these features. You can use these articles to help you get started with the new release. In addition to the 12c launch, many other Oracle products have been updated and released. It has been a busy few months for Oracle. In addition to the print publication, we have a bumper digital edition of Oracle Scene. This is available on the UKOUG website. Do check out all the excellent articles that are available in the digital edition covering application development, applications, management and more. Over the coming months, many of you will be upgrading to and using 12c or one of the many newly released products. You will be learning lots, discovering many hidden features and tricks, identifying work arounds, finding out how to get different products to work together and so on. Our 12c supplement is a great place to start. Check it out online at what-we-offer/oracle-scene. Oracle Scene is a great place for you to share your discoveries and experiences, by writing a short or long article about them. This is what being part of a user group is all about, sharing our knowledge. Content can be submitted all year round to Information about editorial deadlines, article guidelines etc. can be found on the UKOUG website. I would like to thank all of you who have contributed articles to Oracle Scene, our readers and the advertisers. ABOUT THE DEPUTY EDITOR Brendan Tierney Consultant, Brendan is an Oracle ACE Director, independent consultant and lectures on Data Mining and Advanced Databases in DIT in Ireland. Brendan has extensive experience working in the areas of Analytics, Data Mining, Data Warehousing, Data Architecture and Database Design for over 20 years. He has worked on projects in Ireland, UK, Belgium and USA. He started working with the Oracle 5 Database, Forms 2.3 and ReportWriter 1.1, and has worked with all versions since then. Brendan is the deputy editor of the UKOUG Oracle Scene magazine and is the deputy chair of the OUG Ireland BI SIG. Brendan is a regular presenter at conferences in Ireland, UK, Norway, Brazil and USA, including Oracle OpenWorld. Contact the deputy editor at:
  5. 5. JDE13 & OUG Ireland 2014 05 Join us at the Madejski Stadium on 12th–13th November for another fantastic year of JD Edwards content. JDE13 is all about Onwards, Upwards, Edwards with an agenda packed with presentations on upgrading, going mobile and implementation. Don’t miss out on Oracle keynotes John Schiff & Denise Grills, opening motivational speaker ex-referee Graham Poll and user stories from Tupperware Products, Sinclair International, Aggregate Industries and Calor Gas. UKOUG JD Edwards Conference 2013 View the agenda and register online now! Calling all papers…OUG Ireland 2014 The key event for the Oracle community in Ireland is back for its tenth year. Get involved to learn, share, network, and join the celebrations! We’ve now opened our call for papers inviting submissions covering Oracle E-Business Suite, Applications Transformation, BI and Oracle Database. We would especially like to receive papers in the following categories: • Customer stories • Real life business experiences • Innovative ideas • Updates • Latest information and developments • Hints and tips on issues that may be faced in the work environment OUG Ireland 2014 | 11th March | The Convention Centre, Dublin More information about submitting papers can be found here:
  6. 6. 06 AUTUMN 13 News and Reviews OracleScene for a smooth conference registration 1. Make sure you create an online account by going to and clicking the ‘Join UKOUG’button in the top right corner. 2. Attach yourself to your organisation’s UKOUG membership. You will need to obtain the membership number from your main contact. 3. If you have an online account but have forgotten your password, use the forgotten password function on the UKOUG website. Feature in the next We’re looking for compelling stories about your experiences with your Oracle products – from a technical and/or functional perspective. Whatever your story and whether it’s good news or bad, future plans, innovative use of your applications, integrations with other solutions, we want to know about it. Send your submissions to: OracleScene Serving the Oracle Community Autumn 12 Issue 48 Confused by UPK? Tim Poynter sheds some light on UPK Professional Cheshire Shared Services From Systems Integration to Employee Self-Service Jonathan LewisOn Star Transform ations in Standard Edition www.ukoug.orgAn indep endent publication not affiliated with Oracle Corporation The Road AheadFor BI Apps This edition’ssponsors: 68217 UKOU G Oracle Scene 48_AR TWOR K 48PP_ MM_12 1012 amend ed.indd 1 12/10/2 012 12:02 OracleScene Serving the Oracle Community Spring 13 Issue 49 Oracle Business Analytics 2013 trends product direction Oracle ADF Everything a server administrator needs to know Taking the Direct Path More from James Morle on efficient table scanning An independent publication not affiliated with Oracle Corporation Mobile: the Enterprise Gamechanger This edition’s sponsor: UKOUG _Oracle Scene[4 9]_ST3- PA v2.indd 1 01/03/20 13 16:43 Article submissions deadline: 17th Jan 2014 Publish month: March 2014 All About High Availability The day started with an overview, followed by descriptions of the advanced clustering features built into WebLogic, the integration with the Oracle database (particularly RAC) and after lunch, moved into handling HA in the layered products like SOA, BPM, OAM and EM 12c. We also squeezed the usual Support and discussion sessions in between. As a couple of delegates pointed out on the post-event questionnaire, the scope was probably a little ambitious as any one of the subject areas could have filled a day itself. It did mean that the SIG had a bit of a whirlwind feel to it, but as someone else wrote, there was a “really good mix of very experienced presenters who were up for interactive discussions”. In an effort to build on the opportunities to interact with your peers and speakers, we are trying something different at the next SIG.The morning consisting of a couple of main presentations with some 10 minute “Quick Bites”and in the afternoon we will have a “PerformanceTuning Clinic”. In this session the speakers will answer performance questions suggested by the audience, giving advice with possible solutions. The next SIG is on 9th October in Reading and we hope you can join us then. Register at The last UKOUG Application Server and Middleware SIG, held on 19th June in London, was themed around High Availability (HA). Pictured left: Jon Reddy Pictured centre: Bernardo Mirones Pictured right: Rob Honeyman 4. Check with your main contact to see if a conference ticket is available for you to use this year. If not, contact us and we will work out the cheapest way for you to attend. If you have any questions, please contact our membership team on +44 (0)20 8545 9670 or and they will be able to help you. Top tips www.ukoug.orgAn Independent Publication not affiliated with Oracle Corporation IN THIS ISSUE focus OracleScene Lots of 12c topics forthe DBA, Architect, Developer and Data Scientist Supplementsponsor: Autumn 13 Oracle Database 12c Supplement Brought to you by Oracle Scene An in depth look atOracle Database 12c 12c Special! We had more 12c content than we could fit in one edition so we’ve created a 12c supplement. Dive in and discover more articles on new features and more via the UKOUG website:
  7. 7. News and Reviews 07 Get your hands on this hot offer! UKOUG SIGs are the perfect way to prepare for Tech13. Learn about and discuss some of the key conference topics, share experiences and build your network with fellow delegates, users, Oracle representatives and partner members in a relaxed, informal atmosphere. Our Technology SIGs include: • Solaris SIG Meeting • Database Server SIG Meeting • Application Server Middleware SIG Meeting • Availability Infrastructure Management SIG Meeting • Oracle Spatial SIG Meeting • Higher Education SIG Meeting • Application Express (APEX) SIG Meeting UKOUG Platinum membership provides you with: • 10 passes to any Special Interest Group (SIG) meetings* • Two full conference tickets to use at any UKOUG conference** including Tech13 And many other member benefits. Find out more about Platinum membership: The cold weather may be on its way, but we have many Special Interest Group (SIG) meetings this autumn to get you warmed up and ready for the UKOUG Technology Conference 2013 in December. For a spec-tech-ular offer quote “WarmUp” on your application before 30th November 2013 to save the admin fee. Contact our membership team today at *Up to four passes can be used at any one SIG event **All tickets are fully transferrable between colleagues social media followers – thank you! 10,000 The last few weeks have seen us celebrate the birth of the Royal Baby with an offer on Gold membership and this autumn we have hit 10,000 worldwide followers on Twitter, LinkedIn, Facebook, Google+ and YouTube combined. This makes the global Oracle community stronger than ever, better enabling members to make connections, share information and take part in topical discussions online. Of course, we’re not stopping there, so if you’re on social media make sure you’re not missing out. Connect with UKOUG today and help us get to the 20,000 milestone! UKOUG is encouraging all those interested to start using Twitter hashtags when they tweet about UKOUG and our events in order to interact with us and our presenters, attendees and other interested parties. View the full list of the hashtags at UKOUG’s 30th year is turning into a big one! SAVE UP TO 40% Remember all our members can claim discounts on all Oracle titles from our selected publishers. Why not purchase a book and immerse yourself as the evenings draw in to improve your knowledge on the Oracle products you work with? We have discounts from McGraw-Hill, Packt Publishing, O’Reilly, Pearson Education and Sideris. Log into the members area of the UKOUG website ( to access the discount codes or contact Karina on +44 (0)20 8545 9674. Don’t forget about your publisher discounts! Shhh!don’t tell anyone... 6 Secrets to Better Networking at Conferences better-networking-at-conferences P.S.We don’t really mind if you share ☺
  8. 8. 08 AUTUMN 13 Focus on: Tech13 Conference OracleScene First came the 12c launch in July. Then came Oracle OpenWorld in September. Tech13 – Oracle Explored As you would expect, Tech13 will offer three and a half days of world- class technical education for IT professionals and practitioners. And for 2013 we have our best conference programme yet! There will be inspirational sessions from more than 150 leading speakers including: Jonathan Lewis, Tom Kyte, Pete Finnigan, Carl Dudley, Joel Goodman, Mark Rittman, Adam Leventhal, Larry Carpenter and many more. The conference programme is a great opportunity to listen to, and interact with, some of the leading thinkers and practitioners in the Oracle workplace today. We have keynote addresses from Andrew Sutherland, Dermot O’Kelly, Tom Kyte, Mike Durran, Chris Armes and Regis Louis, case studies, roundtables, deep product exploration, hands-on labs and a wide range of collaborative and networking opportunities. Register today
  9. 9. Focus on: Tech13 Conference 09 The next big thing is... Hands-on learning For those looking for in-depth knowledge, our Masterclasses suite will provide attendees with an enhanced learning experience. Sessions include: • Understanding Optimizer Statistics • Oracle Database Backup and Recovery Workshop Exciting entertainment We like to treat Tech13 attendees like VIPs. And this year we are exclusively hiring out the National Football Museum for our delegates to enjoy an evening of entertainment. “The Database stream will have some great content. Last year it was all about – what’s coming? This year – there will be some excellent demonstrations, and it’s a great chance for delegates to ask questions about different features.” Joel Goodman, Oracle 1 DECEMBER 2 DECEMBER 3 DECEMBER 4 DECEMBER M A N C H E ST E R C E N T R A L 1 - 4 D E C 2 0 1 3 More content added Tech13 will deliver more targeted content than ever seen before at a UKOUG event. Three days just wasn’t enough! Seven extra tracks on offer on Super Sunday. Super Sunday includes: • Hands-on Engineered Systems • Analytics – Pulling the Threads Together • Two Database Technology tracks • Real World Lesson from ADF Ninjas • Extending APEX out of the Box • WebLogic As well as the traditional style of conference presentation Tech13 also offers: Interactive sessions Have your say in the Roundtable discussions. Topics include: • Licensing Management • Virtualisation QA • Debate: Does Oracle Ignore Hints? Highlights of this year’s Technology Conference
  10. 10. We aim to make Tech13 the best Oracle technology conference in the world! It will be held in Manchester, at Manchester Central, an iconic business resort in the heart of the city. Centrally located in the UK and easy to reach using city’s extensive multi- modal transport network. 10 AUTUMN 13 Focus on: Tech13 Conference OracleScene 13 New conference M A N C H E STE R C E NTR A L Petersfield M2 3GX NEW VENUE: Manchester Central
  11. 11. Focus on: Tech13 Conference 11 New location Manchester Tech13 is all about hands-on learning, deep product exploration, countless opportunities to build relationships with a community of Oracle experts and peers. But if you need a break from the conference, Manchester has a lot to offer. From events and attractions to fine dining and a great night out; discover things to do in Manchester. 1. Explore the people’s history Reopened in 2010 after a £12.5 million development, The People’s Museum, Manchester’s only national museum, is dedicated to telling a 200-year tale of British democracy. Dry and dusty it isn’t. Interactive exhibits bring political history to life, while a brand new wing, fused to the Grade II-listed Pump House and complete with a sunny riverside café, lets light flood inside. 2. Nightlife in Manchester The Manchester nightlife scene is cool, eclectic and undeniably charming. With so many bars in Manchester the choice can be overwhelming, so we have handpicked a few places: • The Whim Wham Cafe “Epicurean Eatery and Gin Saloon”offering up traditional British food-fare and a thoughtfully stocked bar. • The Briton’s Protection Winner of ‘The Best Pub in Manchester’title for the past 2 years, this tardis-like boozer next to The Bridgewater Hall boasts over 200 whiskeys, great ales, a beer garden and two roaring fires. • Tib Street Tavern There are no gimmicks, it’s not attempting to be a theme bar nor does it offer fancy cuisine, just a good old fashioned selection of beers, spirits and wines from around the Continent. 3. Make inroads to the ‘cultural corridor’ Infamous for its thundering traffic, Oxford Road nevertheless packs in so many galleries, museums and theatres that locals talk of it as Manchester’s ‘cultural corridor’. At the northern end of the main arterial route connecting the centre to south Manchester is Cornerhouse, the city’s acclaimed contemporary art complex, while at the other is the red-brick Whitworth Art Gallery. In between are theatres, the Manchester Museum and the University of Manchester’s campus, a quadrangle of Gothic buildings designed by Alfred Waterhouse. 4. Fascinated by science or intrigued by industrial life? Manchester’s Museum of Science and Industry has a range of captivating exhibits for you to feast your eyes on. Afterwards take in the spectacular views of the stars within the planetarium. The Museum of Science is a fantastic day out if you are planning on extending your trip. 5. Support one (or both) of Manchester’s FCs? Both sets of fans are well-catered for by their teams with tours, museums and of course the obligatory shops, selling the latest strip. Manchester United fans can take the popular behind-the-scenes stadium tour and trot down players’tunnel or visit the impressive museum. No devoted football fan should miss the National Football Museum, which holds both the FIFA and FA collections, including the hallowed ball used in that World Cup final (1966), while changing exhibitions will add interest to permanent displays. Register today at
  12. 12. 12 AUTUMN 13 Technology OracleScene Fine Tuning your Data Stack for Success applications. To understand how these elements work together to provide optimal IT solutions, let’s compare each component of the Oracle stack to talented sections of a world-class philharmonic orchestra. Consider how each section of an orchestra – be it the strings, brass, or woodwind instruments – can give an inspiring musical performance in its own right, so too can each layer of the Oracle stack provide best-of- breed performance for its specific functions. When these components are combined together, however, the true benefit of Oracle’s comprehensive solutions becomes apparent, much like the combined instruments in an orchestra create magical symphonies. When writing a symphony, a composer must imagine how all of the instruments under his or her direction will sound together; while each section does contribute some incredible sounds, the full force of the piece only shines through when played in unison. Likewise, the Oracle stack is developed by a single development organisation and the technology behind it has been designed to optimise the entire stack. Vertically integrated components are finely tuned to work optimally together between, within, and across every layer. They are engineered together, undergo A recent article in the McKinsey Quarterly revealed a concern, shared by many CEOs, that while their competition puts more and more pressure on them to keep up, they are too preoccupied with ‘keeping the lights on’IT projects to embark on truly innovative technology initiatives. There is no doubt that businesses need to focus on keeping their IT resources up and running, but their investment in technology shouldn’t stop there. It’s no surprise that organisations that view IT tools as more than just a set of resources that fill one-dimensional operational roles will experience a higher level of innovation in their organisation. Companies that simplify IT will find the time to take advantage of their technology resources and ultimately create added value internally and for their customers. For business leaders currently struggling with complexity in their organisations, this course of action is extremely attractive. To add to the issue of complexity in the IT space today, greater challenges are expected in the near future in the form of Big Data. While businesses everywhere have been exposed to the Big Data explosion, estimates indicate that 85 per cent of the world’s current data warehouses will not be able to meet expected information volume and complexity demands by 2014. For companies, the choices they make today can either prepare them to take advantage of this information explosion and get a jump on the competition, or keep them lagging behind. At Oracle, our goal is to help customers simplify their IT experience. We offer complete hardware and software stack solutions that include storage platforms, servers, virtual machines, operating systems, databases, middleware and Is your IT department focused on making your business more competitive, or are you caught up trying to maintain the status quo with your technology infrastructure? Dermot O’Kelly, Senior Vice President Oracle UK, Ireland and Israel Technology
  13. 13. Technology: Dermot O’Kelly 13 testing together, certified to work together, upgraded together and supported together. These comprehensive engineered systems deliver extreme performance, and provide cost-effective solutions for customers who do not have to invest their resources integrating each layer of the stack. That being said, just like guest musicians can join an orchestra missing a few players, so too can third party data centre components be integrated into the Oracle stack. Because our solutions are built on open source platforms, there is no customer lock-in. If they choose to, customers can swap out any layer of the stack and replace it with another vendor’s products. With flexible, high-performance solutions like these, Oracle makes IT simpler, and helps users avoid the compromises associated with most proprietary systems. Businesses that adopt our products can take the complexity out of their technology solutions, and therefore reduce the portion of their IT budget that goes into maintaining day-to- day operations. Managing the performance, configuration, provisioning, and patching of IT systems from a single application, for example, allows users to invest time and effort into growing their top line. Oracle is in a unique position to offer customers these complete solutions because it is the only provider that owns all the layers in its technology stack. To achieve this, Oracle has spent over US$24 billion in RD since 2004, investing in product innovation and cross-layer integration in the course of developing its next-generation technologies. Ultimately, what sets the Oracle stack apart is that it has been engineered for choice. Customers who take advantage of this flexibility will simplify their IT infrastructure, thus paving the way for them to drive innovation and achieve measurable success. Let’s make music together! Join us at an Oracle Day coming to a city near you where you will hear direct from companies that have simplified IT to drive innovation. Learn how technology is fuelling new business models and opportunities. Note from Oracle: The UK Oracle User Group will be joining us at our events. We are very proud of our independent user group communities and would encourage you to come along to these events to meet the UK Oracle User Group to find out more about the value user group membership offers our customers. Find out more: ABOUT THE AUTHOR Dermot O’Kelly Senior Vice President Oracle UK, Ireland and Israel Technology Dermot O’Kelly is Senior Vice President for Oracle UK, Ireland and Israel region and is responsible for driving Oracle’s operations, growth and profitability across these geographies. He also leads the close alignment of Oracle’s key accounts, and is the Country Leader for the UK.
  14. 14. Technology 15 Histograms in 12c they can see it is necessary.This tends to reduce the occurrence of “random”bad plans appearing due to the variability of the samples that Oracle might take to generate a histogram – and it may be that the best time to create some histograms is early in the morning rather than late at night when the automatic job starts to run. However, with 12c you’re likely to find that it is safe to let the automatic stats collection job generate some of your histograms – so it’s nice to know that you can set up table-level preferences for stats collections that will tell the automatic job exactly which histograms to collect. Here’s an example of how: begin dbms_stats.set_global_prefs(‘METHOD_OPT’,’for all columns size 1’); dbms_stats.set_table_prefs( ownname = ‘test_user’, tabname = ‘t1’, pname = ‘method_opt’, pvalue = ‘for all columns size 1 for columns n1 size 5 n3 size 254’ ); end; / The first call (which you need only do once) tells Oracle that the general strategy is to collect no histograms. The second call tells Oracle that for table T1 you want a histogram of up to 5 buckets on column n1, a histogram of up to 254 buckets on column n3, and no histograms on any other column. In 12c Oracle has introduced some terrific enhancements to histogram creation: histograms are allowed to have more buckets, you can create accurate frequency histograms at a much lower resource cost, and there are two new types of histogram that eliminate various problems with the old “height-balanced” histograms. Background Before talking about histograms in 12c, it’s worth making a brief comment about a couple of features that appeared in 11g; these are the “approximate NDV”mechanism and the “table preferences” procedure (which only works properly in 11.2). If you enable the “approximate NDV” mechanism through a call to dbms_stats.set_param() or – the preferred method – dbms_stats. set_global_prefs(), then the various calls to gather stats will use a new mechanism to gather column stats whenever you leave the estimate_percent at its default value of “auto_sample_size”.This allows Oracle to collect accurate statistics using a 100% sample size without having to do the expensive count(distinct ()) on each column that it had to do in earlier releases of Oracle. In particular, of course, the automatic overnight stats collection job ought to do a better job with the newer, faster, more accurate mechanism. A piece of advice that I’ve often given to clients has been to disable automatic histogram generation, and then do something to create histograms programmatically for the few special cases where One of the most challenging problems a DBA has to face is defining a strategy for how and when to create statistics for a database; and one of the hardest parts of the problem is dealing with histograms which can be very expensive to create while still leaving the DBA facing unstable execution plans. Jonathan Lewis, Freelance Consultant, JL Computer Consultancy
  15. 15. 16 AUTUMN 13 Technology: Jonathan Lewis OracleScene Having set the background – let’s move on to the things that 12c does better. Enhancements The first enhancement is simple – you can specify up to 2,000 buckets for a histogram in 12c. The default, if you don’t change it, is still 254. Frankly this is likely to be sufficient in many cases, and I don’t think many people will need to increase the value beyond 500 (In part for reasons I’ll mention shortly). If you are tempted to set the bucket count up to 2,000 remember that the calls to gather stats save the old stats to some history tables before overwriting them – and I have seen several complaints in the past from people who have noticed that their sysaux tablespace has become huge with wri$_optstat_histgrm_ history (the histogram history) as the single largest table in the tablespace. The first enhancement is simple – you can specify up to 2,000 buckets for a histogram in 12c. The next enhancement is a relatively simple change to efficiency and precision for frequency histograms – Oracle has overloaded the “approximate NDV”mechanism to accumulate enough information to create a frequency histogram at the same time as it counts the number of distinct values for a column. The original mechanism used to create a hash table of 2^64 buckets, but applied a cunning algorithm that allowed it to represent the full data set (approximately) by recording a maximum of just 16,384 hash values. In 12c Oracle records the number of rows that appeared for each of these hash values, and a rowid to point back to a row with the actual value. If, at the end of the build the number of hash values hasn’t exceeded the number of buckets requested for the histogram, the contents of the hash table can be converted into a very accurate frequency histogram by looking up each of the rowids, and then sorting the hash table by lookup value. The time required to create a frequency histogram is the time to do a full tablescan plus the time to look up the N (up to 2,000) rows by rowid. In 12c Oracle records the number of rows that appeared for each of these hash values, and a rowid to point back to a row with the actual value. New features But we don’t just get a faster, more accurate, frequency histogram – Oracle gets even smarter with the contents of the hash table. Imagine you have asked for a histogram of 50 buckets, and Oracle finds that the hash table holds 3,000 hash values – far more than you had hoped for. Possibly you requested 50 buckets because you knew that almost all the data would be one of 50 critical values. Oracle will compare the total number of rows in the table, with the number of rows in the largest 50 hash buckets, and if the left-over rows in the other 2,950 buckets account (in this case) for less than 1/50th of the data Oracle will build a “Top-N”frequency histogram – setting up histogram information that reflects the top 50 buckets and allows for “a little spare data”. The algorithm is a little more subtle than the description I’ve given as the histogram has to reflect the low and high values for the column, so Oracle may have to inject a couple of “rogue” buckets into the histogram, but the basic principle is this: if you have a data set which you know has N popular values and a load of rows that represent no more than 1/Nth of the data, then create a histogram with N (plus a couple for safety) buckets, and you will get a histogram that is a good model for your data. The final new feature of 12c is the “Hybrid”histogram – so- called because it combines features of the frequency histogram and the height-balanced histogram. The hybrid histogram is something that Oracle will generate if you’ve requested a histogram and the data pattern doesn’t allow Oracle to create a frequency or Top-N histogram. It can be expensive to create a hybrid histogram because Oracle has to sample the data and run a query that uses an analytic sum() function – the larger the sample becomes the more expensive the query. Older versions of Oracle also run an analytic query (using the ntile() function) against the data, so the amount of work to generate a hybrid histogram is not different from the height- balanced version. The difference is in what Oracle does with the data. This is best described with a small example that gives the flavour of the mechanism. Imagine that (after sorting) I have the following 20 values for a column: 32 33 33 33 33 33 33 33 33 34 34 34 35 35 35 35 35 35 35 36 To create a height-balanced histogram of 4 buckets Oracle would record every fifth entry (20 rows / 4 buckets) = 5 rows per bucket, plus the first entry to show that the first bucket wasn’t completely full of 33s, so we’d get the following selection as our histogram: 32 33 33 33 33 33 33 33 33 34 34 34 35 35 35 35 35 35 35 36 32 33 34 35 36 Unfortunately Oracle only considers a value to be special (popular) if it appears in at least two consecutive positions in this list. So, from the perspective of the height-balanced histogram there’s nothing particularly skewed about this data – even though we can see that 33 and 35 occur rather more frequently than the other values. A hybrid histogram of 4 buckets starts off with the same unit of 5 values as its basic measure but doesn’t follow the fixed bucket size. After counting through five values it gets to a 33, but then keeps going until it reaches the last 33, remembering the size of the bucket and the number of 33s; then it counts 5 values which takes it to a 35, and keeps going to the end of the 35s –
  16. 16. Technology: Jonathan Lewis 17 and so on, to give us the following: 32 33 33 33 33 33 33 33 33 34 34 34 35 35 35 35 35 35 35 36 In this case we run out of data before we run out of buckets – which isn’t’too surprising since the algorithm tends to make buckets larger than the expected size as it works along the data. You’ll notice that I’ve also highlighted the first value in the list, again to show that the first bucket does have some values in it that don’t match the high value. The data stored as this histogram looks like this: (32,1) (33,8) (35,7) (36,1) As you can see from this little example, I’ve managed to capture both of my popular values, even though in this case they appear only as single end-point values in the histogram. As a general rule, the hybrid histogram tends to capture more popular values for a similar amount of work than the equivalent height- balanced histogram – and thanks to the end-point counts it captures better information. Hybrid histograms are a significant step forward from height- balanced histograms; bear in mind, though, that it’s still a data sample – the smaller the sample the more variable the effect may be on your execution plans, the larger the sample the longer it will take to build the histogram. Summary In 12c a histogram can have up to 2,000 buckets. This is probably overkill, so the default is still 254 buckets, and you should be a little cautious about much large bucket counts. Remember the impact this would have on the “optimizer history”tables. The new “Hybrid”histograms – enabled automatically if you use the “auto sample size”option – are still as expensive to generate as the old “height-balanced”histograms, but the quality of information they hold is vastly superior. If you’ve enabled the 11g “approximate NDV”then Oracle can create accurate frequency histograms and “Top N” frequency histograms very cheaply; combine this with setting “table preferences”and you may find that you can allow Oracle to deal with almost all your histogram requirements in the overnight “autostats”collection. Remember, though, that the picture of the data you want Oracle to see during the day may not be the same as the actual data content at the moment the auto stats job is running. Despite all the improvements, you may still find a few cases where you need to include extra calls to the dbms_stats package in your application code (typically during batch runs), and some of these calls may even need to use the set_column_stats() procedure to construct an artificial histogram. ABOUT THE AUTHOR Jonathan Lewis Freelance Consultant, JL Computer Consultancy Jonathan Lewis is a freelance consultant whose experience with Oracle goes back over 24 years. He specialises in physical database design, the strategic use of the Oracle database engine and solving performance issues. Jonathan is the author of ‘Oracle Core’ and ‘Cost Based Oracle – Fundamentals’ both published by Apress, and ‘Practical Oracle 8i – Designing Efficient Databases’ published by Addison-Wesley, and has contributed to three other books about Oracle. He is one of the best-known speakers on the UK Oracle circuit, as well as being very popular on the international scene – having worked or lectured in 50 different countries – and further details of his published papers, presentations and tutorials can be found through his blog at
  17. 17. The latest, greatest incarnation of the Oracle database, 12c, is chockfull of interesting and exciting new features. This article introduces the reader to just three of them available within SQL alone. Melanie Caffrey, Senior Development Manager, Oracle 18 AUTUMN 13 Technology OracleScene Row Limiting, PL/SQLWITH SQL, andTemporalValidity: Three in 12c Top-N queries can now employ simpler syntax with the use of the new row_ limiting clause, which allows you to limit the rows returned by a query. You can now define a PL/SQL function in the WITH clause of a query, and reference it within the query, as well as any of the query’s subqueries. And last, but not least, temporal validity enables you to have data visible depending on its time- based validity, as determined by the periods of time for which it is considered valid. Any application can now support a situation where it is important to have visible only the data within a table considered valid as of a specified time period, instead of all the data within a table (the current default). To demonstrate the power of the row limiting clause, I’ll start by querying a table called EMPLOYEE. First the table is queried with no row limiting clauses and then the table is queried with the new row limiting FETCH FIRST and ROWS ONLY clauses, for comparison purposes (Figure 1). ashton%ORA12CR1 select employee_id, first_name, last_name, salary 2 from employee 3 order by salary desc; EMPLOYEE_ID FIRST_NAME LAST_NAME SALARY ----------- ------------------------------ ------------------------------ ---------- 6573 Lori Dovichi 6571 Thomas Jeffrey 300000 28 Emily Eckhardt 100000 6569 michael peterson 90000 1234 Donald Newton 80000 37 Frances Newton 75000 7895 Matthew Michaels 70000 6572 Theresa Wong 70000 6570 mark leblanc 65000 6567 Roger Friedli 60000 6568 Betsy James 60000 11 rows selected. ashton%ORA12CR1 set feedback on ashton%ORA12CR1 set lines 32000 ashton%ORA12CR1 select employee_id, first_name, last_name, salary 2 from employee 3 order by salary desc nulls last 4 FETCH FIRST 5 ROWS ONLY; EMPLOYEE_ID FIRST_NAME LAST_NAME SALARY ----------- ------------------------------ ------------------------------ ---------- 6571 Thomas Jeffrey 300000 28 Emily Eckhardt 100000 6569 michael peterson 90000 1234 Donald Newton 80000 37 Frances Newton 75000 5 rows selected. FIGURE 1
  18. 18. Technology: Melanie Caffrey 19 Notice how with just a few extra keywords, FETCH FIRST 5 ROWS ONLY, the query is limited to the Top-N records we are interested in. This is a syntactical simplification over, for example, the RANK and DENSE_RANK analytic functions also used within Oracle SQL. This is not to say that RANK and DENSE_ RANK should not be used. (I am a huge fan of analytic functions.) This is simply a statement that many Top-N queries that are interested in, say, fetching the first five rows, and maybe the next five rows, for example, could make good use of the new row limiting clause syntax available in 12c. The FETCH clause specifies the number of rows or percentage of rows to return. Comparing the second query with the first query, you can see that the omission of this clause results in all rows returned. The second query fetches only the top five salary earners from the employee table (additionally, the second query omits any records with null salary values, so that only real salary values are returned). To then fetch the next top 5 salary earners, consider the following query (Figure 2). The differences between the third query and the second query are the OFFSET and FETCH NEXT clauses. The OFFSET clause specifies the number of rows to skip before the row limiting begins. The way to look at this is to read it as “skip the first five salary earners, and return the next five salary earners only”. Effectively calling PL/SQL within SQL has been a challenge for many application developers since PL/SQL was first written. One of the challenges has consistently been how to effectively write SQL that calls schema-level stored functions and procedures and ensure that, overall, the code performs well and is maintainable. As far as name resolution goes, a PL/SQL function declared within a SQL query has precedence over schema- level stored functions and procedures. The following query demonstrates how you can declare a function within the WITH clause of an SQL query (Figure 3). This type of query is particularly useful in a read-only database where the values are not likely to change often while the query is running. Keep in mind that if the query in which you specify this clause is ashton%ORA12CR1 select employee_id, first_name, last_name, salary 2 from employee 3 order by salary desc nulls last 4 OFFSET 5 ROWS FETCH NEXT 5 ROWS ONLY; EMPLOYEE_ID FIRST_NAME LAST_NAME SALARY ----------- ------------------------------ ------------------------------ ---------- 7895 Matthew Michaels 70000 6572 Theresa Wong 70000 6570 mark leblanc 65000 6567 Roger Friedli 60000 6568 Betsy James 60000 5 rows selected. FIGURE 2 ashton%ORA12CR1 WITH 2 FUNCTION get_prod_name (sku VARCHAR2) RETURN VARCHAR2 IS 3 start BINARY_INTEGER; 4 length BINARY_INTEGER; 5 BEGIN 6 start := INSTR(sku, ‘***.’); 7 length := INSTR(SUBSTR(sku, start + 4), ‘.’) – 1; 8 RETURN SUBSTR(sku, start + 4, length); 9 END; 10 SELECT get_prod_name(product_sku) 11 FROM product_catalog; 12 / FIGURE 3 not a top-level SELECT statement, but a sub-query, then if the top-level statement is a SELECT statement, it must have either a PL/SQL function or procedure declared within a WITH clause or it must specify the WITH_PLSQL hint. If the top-level statement is an INSERT, UPDATE, DELETE or MERGE statement, then it must have the WITH_PLSQL hint. This hint is not an optimizer hint. It is used specifically for declaring PL/SQL within SQL. Temporal Validity is sure to be a key audit feature as it allows you to specify valid time periods or intervals within a table’s data. A heap table can become a table that supports temporal validity with the addition of the PERIOD FOR clause, which allows you to specify a valid start and end period for which rows are valid and should, therefore, be visible by an application, for example. Consider the following CREATE TABLE statement for a table named EMPLOYEE_VALID, (a table that is similar to the EMPLOYEE table used in this article’s first three queries (Figure 4). ashton%ORA12CR1 create table employee_valid (employee_id NUMBER, 2 first_name VARCHAR2(30), 3 last_name VARCHAR2(30), 4 hire_date DATE, 5 departed_date DATE, 6 salary NUMBER(9,2), 7 manager NUMBER, 8 department_id NUMBER, 9 PERIOD FOR emp_valid_time (hire_date, departed_date)); Table created. FIGURE 4
  19. 19. 20 AUTUMN 13 Technology: Melanie Caffrey OracleScene The EMPLOYEE_VALID table has a valid period of time for which data is now considered when it falls between the time period specified by the emp_valid_ time period created with the PERIOD clause of the above CREATE TABLE statement. A subset of the records from this table reveals the following employee start and end periods (Figure 5). This table can now be queried using Oracle Flashback Technology to perform AS OF and VERSIONS BETWEEN queries similar to the following (Figure 6). ashton%ORA12CR1 select last_name, first_name, hire_date, departed_date 2 from employee_valid; LAST_NAME FIRST_NAME HIRE_DATE DEPARTED_ ------------------------------ ------------------------------ --------- --------- Eckhardt Emily 07-JUL-04 01-JUL-05 Newton Frances 14-SEP-05 07-DEC-06 Newton Donald 24-SEP-06 03-OCT-10 3 rows selected. FIGURE 5 ashton%ORA12CR1 select last_name, first_name, hire_date, departed_date 2 from employee_valid 3 AS OF PERIOD FOR emp_valid_time TO_DATE(‘07-JUL-2004’, ‘DD-MON-YYYY’); LAST_NAME FIRST_NAME HIRE_DATE DEPARTED_ ------------------------------ ------------------------------ --------- --------- Eckhardt Emily 07-JUL-04 01-JUL-05 1 row selected. ashton%ORA12CR1 select last_name, first_name, hire_date, departed_date 2 from employee_valid 3 VERSIONS PERIOD FOR emp_valid_time BETWEEN 4 TO_DATE(‘15-SEP-2005’, ‘DD-MON-YYYY’) AND 5 TO_DATE(‘15-JAN-2006’, ‘DD-MON-YYYY’); LAST_NAME FIRST_NAME HIRE_DATE DEPARTED_ ------------------------------ ------------------------------ --------- --------- Newton Frances 14-SEP-05 07-DEC-06 1 row selected. ashton%ORA12CR1 select last_name, first_name, hire_date, departed_date 2 from employee_valid 3 AS OF PERIOD FOR emp_valid_time TO_DATE (‘01-OCT-2006’, ‘DD-MON-YYYY’); LAST_NAME FIRST_NAME HIRE_DATE DEPARTED_ ------------------------------ ------------------------------ --------- --------- Newton Frances 14-SEP-05 07-DEC-06 Newton Donald 24-SEP-06 03-OCT-10 2 rows selected. FIGURE 6 Summary So, in summary, each of these three new 12c SQL features has their place in terms of providing a simpler way to code an often asked-for type of query. The new row limiting clause greatly simplifies the syntax required for creating, for example, a ranked Top-N query. You can now declare and use a PL/SQL function or procedure within a SQL query which cuts down on necessary namespace resolution. And, last but not least, you can give your table data temporal validity and be sure that certain table rows are only visible when they are valid (and not visible when they are not). ABOUT THE AUTHOR Melanie Caffrey Senior Development Manager, Oracle Melanie Caffrey is a senior development manager for Oracle Corporation. She is co-author of several technical publications including Expert PL/SQL Practices for Oracle Developers and DBAs and Expert Oracle Practices: Oracle Database Administration from the Oak Table (Apress), and the SQL 101 series of articles for Oracle Magazine. Save 10% With a 2 year fixed membership Call the membership team on: +44 (0)20 8545 9670
  20. 20. UKOUG Oracle Days 21 W ith over 10 years of experience, MapMechanics is a recognised expert in the provision, processing and enhancement of map data for Oracle systems and currently has over 60,000 resources being scheduled globally with our map data on a daily basis. From small and medium sized businesses to enterprise level companies, MapMechanics can help you to get the most out of map data for these Oracle systems: • Oracle Real-time Scheduler • Oracle Mobile Workforce Management • Oracle Transportation Management • Oracle Business Intelligence Enterprise Edition • Siebel Field Service • Oracle Spatial MAPMECHANICS Map data solutions for your Oracle application “We have found MapMechanics is very well equipped to provide exactly the mapping and demographic data we need. The MapMechanics team are responsive and helpful when it comes to advising on the best products for our requirements, and good at supplying the data in an appropriate format. – For more information Call +44 (0)20 8568 7000 or email Find us on Twitter @mapmechanicsfrom MapMechanics Oracle advert half 170x123_Layout 1 24/07/2013 14:26 Page 1 Join hundreds of your peers at the Oracle Days 2013 to see how you can plan your journey for the future of your IT delivery and drive business success. Highlights include: • Oracle customers such as Virgin Media and An Post will share first-hand how they have simplified IT to transform their business with Oracle technology. • First opportunity to hear direct from Oracle all new product news and announcements from Oracle OpenWorld. • Meet your peers who are facing the same challenges as you at the networking drinks reception. UKOUG will have a stand, with competitions, prizes and giveaways. So come and see us there! Join UKOUG at Oracle Days 2013 In London, Dublin and Manchester this November Register your place today! London 5th November Dublin 12th November Manchester 14th November Platinum sponsor
  21. 21. Learn more at The world is changing faster than ever, creating new opportunities for those ready to seize them. At Cognizant, we combine business consulting, next-generation technology and business process expertise to help hundreds of organizations across the globe not only run better, but run different. It’s a commitment that has helped make us a mainstay on Fortune’s ‘World’s Most Admired Companies’ for three years running. So if you want to continue to stand apart, we have three words of advice: don’t stand still. The future belongs to those who challenge the present.
  22. 22. REDEFINING ENTERPRISE TRANSFORMATIONS For more information visit Email: 1 Kingdom Street, London W2 6BD, UK Phone: +44 207 297 7600 Fax: +44 207 121 0102| | End-to-end Oracle Services Consulting | Implementation | Upgrade | Application Support | Testing Cognizant is a leading provider of information technology, consulting, and business process outsourcing services, dedicated to helping the world’s leading organisations build stronger businesses. Cognizant's Oracle Solutions Practice offers a wide range of services and solutions covering Oracle's Application and Technology products. Our unique onsite-offshore customer engagement model, passion for client satisfaction and innovation along with deep industry expertise has elevated us to a leadership position in the global business and IT services marketplace. Fusion Applications Implementing the world’s largest Oracle Fusion Applications Footprint the UK Accredited Oracle Fusion Localisation Partner in Oracle certified implementation methodologies Packaged solution for Casual Dining Industry Fusion Middleware Expertise in End-to-end Oracle Fusion Middleware services Innovation center for Oracle Fusion Middleware AIA Beta testing, PIP co-development with Oracle  Proven tools provide enhanced value to customers PeopleSoft Experienced in delivering numerous Oracle’s PeopleSoft 9.1 engagements globally Asia’s largest end-to-end Implementation of PeopleSoft for 162,700+ users Beta Testing Partner with Oracle for PeopleSoft 9.2 E-Business Suite Extensive experience delivering E-Business Suite R12 engagements  R12.x Beta Testing Partner for Oracle Oracle E-Business Suite release 12 customer proof of concepts Engineered Systems Global delivery experience across industries Dedicated CoE comprising Exadata architects and migration specialists  Repository of solution frameworks, tools, accelerators and enhanced documentation library  Close collaboration with Oracle in developing the infrastructure and joint GTM JD Edwards  Delivery Expertise in JD Edwards EnterpriseOne and WorldSoft  Domain expertise in Manufacturing, Distribution, SCM, CRM, Financial Management, HR Payroll  Rich talent pool across Functional, Technical and CNC administration
  23. 23. 24 OracleScene AUTUMN 13 JD Edwards Drive Value From Your ERP A PISA helps organisations achieve their Enterprise Resource Planning (ERP) goals with minimal service costs by identifying value gaps within the ERP to realise additional returns on investment. Dale Kaplan, Vice President – Sales, Marketing and Strategic Alliances, CAPSCIENT Organisations that have implemented JD Edwards ERP software have the potential to realise and drive additional value from their implementations by focusing on how those systems are being used, operated and maintained. New software implementation projects define various value driver metrics as goals to measure project success. Due to over-riding project pressures (i.e. time and budget constraints), it is not uncommon that these goals become less of a focus as system operational (i.e. getting the system installed quickly) and functional considerations take precedence. At project go-live and after, the measurement of progress against the goals is often not performed. Investing in large scale process improvement initiatives to drive ERP value is costly and requires time and resources, preventing many organisations from taking these next steps. A PISA provides the solution. Many JD Edwards ERP system integrators have formulated service offerings to meet these type of challenges. Referred to as Post Implementation System Assessments (PISA), the services provided are efficient, focused and inexpensive as they drive significant value from the ERP system. The success of the PISA is based on the unbiased low-cost focused approach to assessing the value gaps. With that said, identifying the value gaps can be challenging, requiring skilled expert product-knowledgeable resources to evaluate current system usage, functionality usage gaps and their alignment to best business processes. Resources have to have deep product and latest version release knowledge which internal company resources may not have. Other constraints include internal resource time and work conflicts that cause unavailability for these type of projects, minimal JD Edwards product expertise and lack of leverage and authority to affect change within the organisation. Those closest to the system are typically not the best choice for evaluating it critically, as they most likely were the ones involved in its implementation and may have a more protective and biased viewpoint over how the system is used. The PISA review is performed by an independent third party with an unbiased perspective which is critical to the PISA success. For organisations that have older ERP systems or that are on back releases and are planning upgrades, understanding where and how to drive additional value from their current state to that of the new system prior to implementing it, is a significant consideration and which provides the business case justification for moving the company forward to the new release. For upgrades, the new release value proposition and net change documentation is available from Oracle, however it still has to be correlated to the current usage of the system to justify the upgrade value. The PISA process can facilitate identification of the key areas where improvements are needed. As an example, in an upgrade scenario, current processes that worked may not work in the new system or the new system may bring improved efficient processes and ways to do business that should be utilised with a Post Implementation System Assessment (PISA)
  24. 24. JD Edwards: Dale Kaplan 25 to drive additional value that may mean revision to current business processes. To summarise some of the main hindrances to IT departments not executing on PISA type projects are related to time, cost, resource constraints, other projects having a priority over funds and resources, as well as a lack of deep in-house expertise in the most recent releases of the ERP software. The PISA service offering is a focused system review engagement of a company’s existing ERP operations against best-practice processes and configurations of both the company’s existing software release and can include a review against the most recent software release. The PISA identifies areas for improvement and additional configurations, including needed modifications to ensure best practices. It takes into account cost reduction through improved software efficiencies derived by increased system functionality usage and alignment of new processes to the business. Industry best practices are also a key focus area that drives additional incremental value. The PISA process typically utilises templates, questionnaires, structured interviews, checklists, ERP implementation best practice knowledgebases, new version and product net change information. The PISA provides the necessary support, direction and plan for execution of any system changes and evaluates the system by module by identifying module utilisation, and provides the guidance needed for implementation of the recommendations. Customers that are upgrading their ERP can also derive value from the PISA. As part of the PISA evaluation, net changes to the proposed new system version release and identification of value drivers to support upgrade investments can be included. The PISA supports the scope and planning phases of upgrades (Diagram 1). A PISA is a focused system review that drives ERP value by providing specific best-practice process and configuration improvements for your JD Edwards system. The initial PISA steps include the Review Phase, in which the original ERP project implementation documentation is reviewed (if available), and the design, testing, training and other components of the implementation that were originally performed are evaluated. This process allows for quick identification of areas for further review and analysis. User requests and reported help desk incidents are reviewed to quickly identify system configuration issues and areas where additional end-user training is needed. The PISA vendor’s expert knowledge of ERP software, specifically Oracle, JD Edwards and known risk areas within each of the modules within an implementation, allows vendor SI consultants to focus the PISA efforts on those risk areas as a priority. Recommendations to mitigate those risks can then be quickly developed as part of the PISA report. If they do exist as gaps in the implementation the consultant can provide mitigation strategies upfront. The PISA approach is to analyse the current ‘state of the union’through focused and planned discussion with management, key users and process owners, as well as system and documentation reviews. The PISA is divided into three tracks, that run parallel to each other to ensure congruence with the PISA goals, namely Project Management, Functional and Technical. At the start of the PISA, the Project Manager establishes a PISA index metric against which to determine variance from best practices and which is used to report system usage factors. Recommended actions for improvement against the index are provided (see Diagram 2). The factors used to evaluate the PISA index may vary by PISA SI vendor, but should include coverage of both the functional and technical areas of the system and cover objective and subjective measurements. A PISA final report is prepared and presented to management. The PISA approach is flexible and can include any specific items of customer concern, for example review of other related integrations or systems that may impact the main system being reviewed. Overriding the approach to the PISA is the vendor’s implementation DIAGRAM 1: PROCESS DIAGRAM: POST IMPLEMENTATION SYSTEM ASSESSMENT (EXAMPLE FOR JD EDWARDS ERP)
  25. 25. ABOUT THE AUTHOR Dale Kaplan Vice President – Sales, Marketing and Strategic Alliances, CAPSCIENT Dale Kaplan, currently holds the position of Vice President, Sales, Marketing and Alliances, at CAPSCIENT, a leading JD Edwards consultancy and value systems integration firm specialising in Oracle services. Prior to joining CAPSCIENT he led KPIT/SYSTIME’s Global Alliances, JD Edwards Global Solutions and Service Delivery, and Organisational Development and Strategy departments. He held senior management positions at Arthur Andersen LLP in the areas of JD Edwards Business Consulting, Computer Risk Management and Financial Audit. He qualified as a Chartered Accountant and holds a Bachelor of Business Science with major in Finance from the University of Cape Town, South Africa. Currently based out of Los Angeles, he can be reached at +1.310.944.5038 / / / 26 AUTUMN 13 JD Edwards: Dale Kaplan OracleScene methodology which should be robust enough to ensure quality project execution. The Functional Track (see Diagram 1) explores the following key areas of the ERP system: • Review of the original ERP implementation documentation that includes design, testing, training and readiness assessment type documentation. • Review of the current open user reported issues lists, review of system integrities and other system related issues that signal areas for improvement. Evaluation of these areas for user training deficiencies is key. • Review of existing configurations of the system (e.g. chart of accounts setups, constants, automatic accounting instructions, version, and other significant configurable items). • Interviews with management, business line leaders/process owners and users. • Review of technical items that impact functional users – like access security, menus and naming conventions. • Review of functional aspects related to RICE (Reports, Interfaces, Customisations and Extensions). • If upgrading, consideration for doing net change comparisons including demonstrations of new release versions can assist in highlighting areas for improvement. • Comparison to best practices and highlighting business process weaknesses with recommendations for improvement. • Assessment of the system utilisation percentage (i.e. to the pre-defined PISA index metric) to enable focus on areas that need the most attention. • Collaboration with the Technical Track consultants to identify similar items for improvement that cross application weaknesses and support the recommendations. • Preparation of a final PISA report. PISA consulting specialists drive and facilitate the PISA workshop sessions and provide documentation post meetings that include decisions taken, issues and risks identified, and actions to be performed with next steps. Along with meeting agendas and the like, the PISA process enables an expedited review of the ERP system saving time, money, cost, effort and reduction of the customer’s resource time. The second phase of the PISA is the Technical Track where the system is reviewed for technical weaknesses with focus on performance issues, system configuration issues, security and access, sizing, architecture and RICE elements. Reviews of patches, upgrade approaches, policies and procedures, backups, purging and archiving are also evaluated. Helpdesk issues and problems are reviewed to identify trends that could signal quick hits for improvement recommendations. For organisations that have heavily customised their systems, a detailed assessment of those RICE elements should be performed to evaluate if and how they can be retired and base vanilla JD Edwards functionality can be used in its place so as to reduce issues with upgrades and future patches. Review of the software development methodology followed and the functional and technical specifications and documentation prepared can also highlight weaknesses. Many of the areas mentioned under the Functional Track also require the involvement of the PISA technical consultant to assist in formulating the recommendation. Project oversight and governance by an assigned executive Project Sponsor should be included at various stages of the PISA project, typically around each key milestone and review checkpoint to review deliverables and provide feedback. Given the short duration of a PISA project, at minimum weekly status meeting should be held, but bi-weekly or even daily “scrum” type status meetings are recommended. On completion of the PISA a final report with recommendations for both functional and technical areas is presented to the organisation and a plan for the execution of the recommendations is developed. With a structured approach to a PISA, with defined agenda, workplan, processes and deliverables, performed by skilled ERP PISA experts, organisations can quickly understand how to drive more value from their ERP investments. Implementing the recommendations identified by the PISA will realise the returns and benefits that were originally planned to be attained from the implementation of the system. DIAGRAM 2: EXAMPLE OF PISA METRIC ANALYSIS
  26. 26. 28 AUTUMN 13 Technology OracleScene Oracle Advanced Analytics in Oracle 12c New Features of By now you may have heard that Oracle 12c has been released. In addition to the new version of the database we also get a new version of SQL Developer (version 4). Brendan Tierney, Consultant, For those of you who are interested in data mining, predictive analytics or data science, Oracle 12c has a number of new features for Oracle Data Miner that is part of the Advanced Analytics Option. The purpose of this article is to outline these new features and to give some examples of how they can be used. The article has two main parts. The first part looks at new features that come with the Oracle Data Miner tool that is part of SQL Developer 4. The second part looks at the new in-database features that come with the Oracle 12c database. The Oracle Advanced Analytics Option consists of Oracle Data Mining and Oracle R Enterprise (ORE). ORE is a separate product that has a separate install for both the server and client side. There has been no new release of ORE that coincided with the releases of 12c and SQL Developer 4. So this article will not be looking at what ORE can do. New Features in the Oracle Data Miner Tool The new features of the Oracle Data Miner tool, which comes as part of SQL Developer 4, can be grouped into two categories. The first category contains the new features that are available to all users of the tool (11.2g and 12c). The second category contains the new features that are only available in 12c. The new features of each of these categories will be explained below. Category 1 – Common new features for 11.2g and 12c Database users There is a new View Data feature that allows you to drill down to view the table object and to view nested tables. A new Graph Node that allows you to create graphs such as line, bar, scatter and boxplots for data at any stage of a workflow. You can specify any of the attributes from the data source for the graphs. You don’t seem to be limited to the number of graphs you can create.
  27. 27. Technology: Brendan Tierney 29 A new SQL Node. This is a welcome addition, as there have been many times that I’ve needed to write some SQL or PL/SQL to do a specific piece of processing on the data that was not available with the other nodes. There are two important elements to this SQL node really. The first is that you can write SQL and PL/SQL code to do whatever processing you want to do. But you can only do it on the Data node you are connected to. The second is that you can use it to call some ORE code. This allows you to use the power of R and the extensive range of packages that are available to expand the analytic functionality that is available in the database. If there is a particular function that you cannot do in Oracle and it is available in R, you can now embed this function/code as an ORE object in the database. It can then be called using SQL and the SQL Node. WARNING: This particular feature will only work if you have ORE installed on your, or 12.1c. New Model Build Node features, include node level text specifications for text transformations, displays the heuristic rules responsible for excluding predictor columns and being able to control the amount of classification and regression test results that are generated. New Workflow SQL Script Deployment features. Up to now the workflow SQL script, I found to be of limited use. The development team have put a lot of work into generating a proper script that can be used by developers and DBAs. But there are some limitations still. You can use the script to run the workflow automatically in the database without having to use the ODM tool. But it can only be run in the schema that the workflow was generated. You will still have to do a lot of coding (although a lot less than you used to) to get your ODM models and workflows to run in another schema or database. This will output the deployment script to a file buried deep somewhere inside your SQL Developer directory. You can edit this location to have a different shorter location. Category 2 – New features for 12c Database users Now for the new features that are only visible when you are running ODM/SQL Dev 4 against a 12c database. No configuration changes are needed. The ODM tool checks to see what version of the database you are logging into. It will then present the available features based on the version of the database. New Predictive Query nodes allows you to build a node based on the new non-transient feature in 12c called Predictive Queries (PQs). In SQL Developer we get four types of Predictive Queries. These can be used for Anomaly Detection, Clustering, Feature Extraction and Classification. It is important to remember that the underlying model produced by these PQs do not exist in the database after the query has executed. The model is created, used on the data and then the model is deleted. The Clustering node has the new algorithm Expectation Maximization in addition to the existing algorithms of K-Means and O-Cluster. The Feature Extraction node has the new algorithm called Principal Component Analysis in addition to the existing Non- Negative Matrix Factorization algorithm. Text Transformations are now built into the model build nodes. These text transformations will
  28. 28. Wary of moving to Release 12? Do you find it difficult to justify the cost of an upgrade project against the benefits R12 can bring? Would you prefer to avoid heavy capex spend for hardware, for transition, for extensive re-training and third-party costs? Mokum’s hosted Oracle Release 12 platform could be the answer. Oracle and E-Business Suite are registered trademarks of Oracle and/or its affioliates. Other names may be trademarks of their respective owners Mokum Change Management Ltd Offices: London Manchester Dublin Tel: +44 (0)20 7536 9962 Email: The combination of Oracle’s virtualisation technology and Mokum’s proven R12 managed services expertise will reduce - or even remove - the need for capital expenditure and will minimise business disruption. Your organisation can make the move to a - supported platform at low cost, and in little time. Fancy joining the Mokum private cloud? Call or email right now for more information...
  29. 29. Technology: Brendan Tierney 31 be part of the Automatic Data Processing steps for the model build nodes. This is illustrated in the above images. The Generalized Linear Model that is part of the Classification Node has a Feature Selection option in the Algorithm Settings. The default setting is Ridge Regression. Now there is an additional option of using Feature Selection. New 12c In-database Features As part of the 12c there are a number of new in-database Advanced Analytics features. These new features can be called using SQL and PL/SQL. These new features are separate to the Advanced Analytics new features that come as part of the Oracle Data Miner tool, that is part of SQL Developer. The new Oracle 12c in-Database Advanced Analytics features include: • Using Decisions Trees for Text analysis is now possible. Up to now (11.2g) when you wanted to do text classification you had to exclude Decision Trees from the process. This was because the Decision Trees algorithm could not support nested data. • Additionally for text mining some of the text processing has been moved from having a separate step, to being part of the algorithms. • A number of additional features are available for clustering. These include a cluster distance (from the centroid) and details functions. • There is a new clustering algorithm (in addition to the K-Means and O-Cluster algorithms), called Expectation Maximization algorithm. This creates a density model that can give better results when data from different domains are combined for clustering. This algorithm will also determine the optimal number of clusters. • There are two new Feature Extraction methods that are scalable for high dimensional data, large number of records, for both structured and unstructured. This can be used to reduce the number of dimensions to use as input to the data mining algorithms. The first of these is called Singular Value Decomposition (SVD) and is widely used in text mining. The second method can be considered a special scoring method of SVD is called Principal Component Analysis (PCA). With this method it produces projections that are scaled with the data variance. • A new feature of the GLM algorithm is that it will perform a feature section step. This is used to reduce the number of predictors used by the algorithm and allow for faster builds. This will make the outputs more understandable and the model more transparent. This feature is not default so you will need to set this to ‘on’if you want to use it with the GLM algorithm. • In previous versions of the database, there could be some performance issues that relate to the data types used. In 12c these has been addressed for BINARY_DOUBLE and BINARY_ FLOAT. So if you are using these data types you should now see faster scoring of the data in 12c. • There is a new in-database feature called Predictive Queries. This allows on-the-fly models that are temporary models that are formed as part of an analytics clause. These models cannot be tuned and you cannot see the details of the model produced. They are formed for the query and do not exist afterwards. SELECT cust_id, age, pred_age, age-pred_age age_diff, pred_det FROM (SELECT cust_id, age, pred_age, pred_det, RANK() OVER (ORDER BY ABS(age-pred_age) DESC) rnk FROM (SELECT cust_id, age, PREDICTION(FOR age USING *) OVER () pred_age, PREDICTION_DETAILS(FOR age ABS USING *) OVER () pred_det FROM mining_data_apply_v)) WHERE rnk = 5; • There is a new function called PREDICTION_DETAILS. This allows you to see what the algorithm used to make the prediction. For example, if we want to score a customer to see if they will churn, we can use the PREDICTION and PREDICTION_PROBABILITY functions to do this and to see how strong this prediction is. With PREDICTION_DETAILS we can now see what attributes and values the algorithm used to make that particular prediction. The output is in XML format. These are the new in-database Advanced Analytics (Data Mining) features. Apart from the new algorithms or changes to them, most of the other changes give greater transparency into what the algorithms/models are doing. This is good as it allows us to better understand and see what is happening. Conclusion The new features of the Oracle Data Miner tool and the in-database new features show continued improvement in the Advanced Analytics Option. The improvements and new features for the Oracle Data Miner tool consist of new features for 11.2g and 12.1c users. In particular the Graph and SQL nodes are two very useful features. With the SQL node we can now include R code developed using ORE, in our Oracle data mining and data science projects. The in-database 12.1c improvements brings greater insight into what the functions and procedures are doing behind the scenes. Additionally, we get some new algorithms that increases the types of advanced analytics we can perform on our data. To read ‘About the Author’ please go to page 4.
  30. 30. Streamlining EPM Deployment with Planning in the Cloud In today’s challenging planning environment, reliance on spreadsheets is woefully inadequate and fraught with peril. Jennifer Toomey, Senior Principal Product Marketing Director – Business Analytics, Oracle Packaged planning and forecasting solutions can help alleviate these difficulties, yet may remain out of reach for many organisations which do not have adequate IT resources or capital budgets to justify the investment. New Cloud- based solutions, such as Oracle Planning and Budgeting Cloud Service, Oracle’s first Enterprise Performance Management (EPM) solution for the Cloud, opens up opportunities for organisations of all sizes to streamline planning and forecasting, accelerate deployment and reduce costs. Today’s Enterprise Planning Challenges – Drowning in Spreadsheets The planning environment has changed considerably for most organisations in today’s global economy. Global recession, erratic growth and recovery, market volatility all make accurate planning and forecasting a challenging process. In fact, a 2010 survey by CFO Research Services revealed that 40% of CFOs can only accurately forecast one quarter into the future. To make matters worse, a recent Oracle survey on planning and forecasting revealed that 93% of finance managers globally are drowning in spreadsheets. This was especially true in the area of budgeting, planning and forecasting with 75% of respondents indicating they used spreadsheets in this area. While spreadsheets are easy to use and are the preferred tool of finance professionals, they are not well-suited for enterprise- wide processes like planning. When organisations are overly reliant on spreadsheets to support planning 32 AUTUMN 13 Hyperion EPM OracleScene
  31. 31. activities, they face challenges around time, quality, flexibility and cost. Long planning cycles result in missed opportunities and obsolete plans. Data integrity and quality issues show up in the form of errors in Excel spreadsheets, version control problems, lack of audit trails. Disconnected processes create a lack of flexibility and responsiveness to changes. And finally, too much time spent gathering information and not enough time to analyse results in costly waste of resources with questionable benefits. Yet, despite these challenges with current tools, the adoption of packaged enterprise performance management software is still not widespread. This is evidenced by relatively low EPM penetration rates among ERP install base customers. A Shift Towards the Cloud New technologies like Cloud make it easier to get off of spreadsheets and adopt best practices, such as rolling forecasts and driving the planning process beyond finance to the broader community of line managers. Gaining the input and intelligence of line of business managers 40% Can Forecast Only 3 Months Ahead 40%: 3 months or less 6 months in advance 3 months in advance 1 month in advance 12 months in advance More than 12 months in advance Not sure 7% 5% 20% 29% 28% 0% 20% 40% 12% Source: CFO Research, the research group at CFO Publishing LLC FIGURE 1: CHALLENGES IN FORECASTING becomes much more feasible with a cloud-based model that can be easily and flexibly rolled out across the organisation. Whereas a couple of years ago, we still saw reluctance in the finance department around adopting cloud applications, largely driven by security and data confidentiality concerns, this is changing rapidly. A 2013 Gartner Financial Executives International (FEI) CFO Technology Study indicated that during the next four or more years, 84% (up from 53% in 2012) believe that half of their transactions will be delivered through SaaS. It appears that 34% in 2012 have made a SaaS directional decision, as there were no “Don’t Know”responses in 2013.1 As CFOs shift their attitudes towards Cloud, we are also seeing the SaaS model for EPM applications gaining traction. In the 2013 Gartner FEI study, budgeting, consolidations and reporting selections showed a continued movement to cloud where almost one in five companies see themselves using SaaS. In conversations with CFOs, some of the reasons they cite in favor of cloud-based finance systems include the resource allocation flexibility that comes with avoiding large upfront investments, the ability to avoid fixed capital investments during periods of corporate or economic uncertainty and timely access to the latest software capabilities. The fact that traditionally risk-averse CFOs are open to moving their mission-critical finance systems into the cloud should be a clear indication to all companies that cloud services are fast becoming the go-to strategy for finance executives looking to access the latest technologies quickly and cost-effectively to support their corporate objectives. Lowering the EPM Adoption Barrier – Oracle Hyperion Planning in the Cloud As we see our customers’attitudes and requirements shifting towards Cloud, we want to meet those changing needs. Thus, Oracle is making its Hyperion Planning solution available in the Cloud with a service called “Oracle Planning and Budgeting Cloud Service.”Oracle Planning and Budgeting Cloud Service removes the barriers to adoption of on-premise applications and makes it much easier for businesses of any size to deploy a world- class planning and budgeting solution in a matter of weeks. “We are extremely excited about the new Oracle Planning and Budgeting Cloud Service offering. We expect this solution to enable customers to get up and running quickly and streamline their planning, budgeting and forecasting processes, all with lower up- front costs and investment risk than a traditional on- premise solution.” Neil Sellers, Director, Qubix Process % Using Spreadsheets Financial Planning and Budget Control 75% Financial Reporting and Analysis 74% Scenario Planning 40% Tracking KPIs 53% Monitoring Profitability 59% Other 3% Source: Performance Management: An Incomplete Picture. Study carried out by Dynamic Markets for Oracle Corporation, April 2011 FIGURE 2: FINANCE MANAGERS DROWNING IN SPREADSHEETS Hyperion EPM: Jennifer Toomey 33 1 Van Decker, John, “Survey Analysis: CFOs’Top Imperatives From the 2013 Gartner FEI CFO Technology Study”, May 2, 2013, p.11.
  32. 32. On Demand Inquiries Reporting when you need to, not because you have to I just want to see my JD Edwards data... Visit booth 26 at the JD Edwards Conference from the 12th - 13th November =Getting Answers More Reports
  33. 33. 35 Most Hyperion EPM customers have deployed the applications on-premise. And over the past 5 years, Hyperion EPM applications have been available as a hosted solution through Oracle Managed Cloud Services (formerly called Oracle On Demand) or many of our hosting partners around the world. But now we are moving our EPM applications into the Oracle Cloud and making them available as SaaS-based solutions. Moving to cloud-based applications has the potential to remove the adoption barriers for many organisations that didn’t have the IT resources or budget for packaged software applications in the past. By way of background, Oracle Hyperion Planning is a market-leading application that accelerates planning, budgeting and forecasting. It’s built to leverage the powerful Oracle Essbase multi- dimensional analysis engine, includes a Web or Excel interface, powerful workflow and process management and pre-built functionality to address the most complex planning and budgeting requirements and best practices in this area. Hyperion Planning supports driver-based planning to help connect operational assumptions to financial outcomes, and it supports a hierarchical planning process that encompasses both corporate finance and the lines of business within an enterprise. Hyperion Planning has seen rapid adoption over the past 10 years with over 3700 organisations having deployed the solution. And many of these deployments have become quite large with many over 1000 users and a few with over 5000 users. The results that our customers have achieved with Hyperion Planning have been impressive. Most often cited is the ability to reduce budgeting and forecasting cycle time. Customers have also reduced reliance on spreadsheets and improved the accuracy of their forecasts. Oracle Planning and Budgeting Cloud Service provides all the power of Hyperion Planning as a cloud service with a subscription-based pricing model. The application can be deployed as a standalone cloud solution, or integrated with Oracle Fusion Financials Cloud Service. All the infrastructure, reporting, and data management components of Hyperion Planning are included as part of the Oracle Planning and Budgeting Cloud Service, so everything the customer needs Less Time Spent in Annual Budgeting and Planning Process 15% less time spent in manual processes supporting annual budgeting Oracle Business Analysis, Customer Value Index, Interim Results, April 2013 Before 4.5 After 3.9 NumberofMonths to run their planning process is part of the service. Oracle Planning and Budgeting Cloud Service is built upon the robust Oracle Cloud, and integrates with your other on-premise and Cloud applications. “The addition of Oracle Planning and Budgeting Cloud Service to Oracle’s Cloud portfolio offers us more choice in deployment. We found Oracle Planning and Budgeting Cloud Service particularly easy to use with good response times and all the functionality that we expect from a sophisticated planning application.” Steve Tevault, Database Administrator, Friendly’s Ice Cream Deploy How YOU Want Our goal is to provide the most flexible deployment options to our customers. That can include on-premise deployments, pure cloud-based deployment, or a hybrid approach where development is done in the Cloud, then testing and production deployment on-premise. Many companies are assessing Cloud options in parallel with traditional upgrades of on-premise solutions. But they are wary of the integration issues between Cloud and on-premise solutions and do not want to be locked into a particular approach. For many, the flexibility of deployment holds the key to the way forward, i.e. the ability to adopt public, private and hybrid Clouds as desired and to alter the mix when business circumstances dictate. But very few software vendors can hope to offer this range of choice and even fewer can leverage their own hardware and database technology to optimise performance. Oracle has been offering applications in the Cloud for more than 12 years and is one of the few vendors that can comprehensively support the disparate needs of small, medium and large multinational organisations in the Cloud with flexible, modern and secure applications. For organisations who own or are considering Hyperion Planning, there are a number of options to take advantage of the new Cloud offering. For new applications, customers can develop and deploy in the Cloud, or develop in the Cloud and then deploy on premise. Existing Hyperion Planning customers may wish to support new divisions or Hyperion EPM: Jennifer Toomey FIGURE 3: RESULTS ACHIEVED WITH ORACLE HYPERION PLANNING
  34. 34. Hyperion EPM: Jennifer Toomey 37 subsidiaries that want their own Hyperion Planning application. They may also wish to move their existing applications to the Cloud, such as upgrading from an earlier Hyperion release. And, as described above, customers always have the option to move back on premise if needed or as business needs change. Future directions – EPM in the Cloud As our customers’attitudes shift towards leveraging the Cloud for even their Finance functions, Oracle is moving to provide more flexibility and choice in deployment options. Oracle Planning and Budgeting Cloud Service is the first of Oracle EPM applications to be offered in the Cloud. In the future you will see Oracle offering additional EPM applications in the Cloud, providing fast time to value, simplicity of configuration and pre-built content for deeper integration and easy extensibility. ABOUT THE AUTHOR Jennifer Toomey Senior Principal Product Marketing Director – Business Analytics, Oracle Jennifer Toomey is Senior Principal Product Marketing Director at Oracle Corporation, with over 18 years experience in the high tech and services industries. At Oracle, she focuses on Business Analytics and Cloud. Jennifer has a BA degree from Harvard University and an MBA from MIT’s Sloan School of Management. Our mobile app gives you access to all the information relating to Apps13 Tech13 including: EXPERIENCE UKOUG I N WHOLE N EW WAY Search for: UK Oracle User Group to download the app today • The agenda • Session information such as time, date and abstracts • Speaker information • Exhibitor information including contact details • Up to the minute social media feeds • Networking facilities with other attendees • Pre-event and on-site updates relating to the agenda and other activities on site • Useful maps to help you around the venue and local area • Access to your personalised event information Mobile app sponsored by: Going Mobile
  35. 35. 38 OracleScene Learning the Fusion Technology Stack The Tools of Fusion: Oracle JDeveloper and Oracle ADF In this column over the past couple of years, I’ve given you an insight into how we in Oracle use Oracle ADF to build our Fusion Applications. Grant Ronald, Director of Product Management, Oracle Application Development Tools Throughout those articles I’ve introduced Java, ADF, JDeveloper, Groovy, Mobile, DVT, MDS, CSS, HTML and an extra large helping of alphabetti spaghetti acronyms for good measure. And herein lies an issue with which I think many of us will empathise: Enterprise software development isn’t necessarily getting easier. Now, don’t get me wrong, I’m not saying our development environments have taken a step back into the dark ages. On the contrary, take it from someone who flipped switches on a PDP-11, the JDeveloper IDE is like something from Star Trek by comparison. But the fact is as we explore new frontiers, we look to innovate and those more complex and demanding requirements are asking more from both tools and developers. So how do you help developers to pick up new technologies and keep their skills up to date? Well that’s a topic very close to my heart. My team is responsible for both reactive and proactive enablement of Oracle ADF developers. That ranges from using social channels such as technical blog postings and YouTube recordings, up to Oracle University courses and all stops in between. In this article, I’ll look at the challenges developers face and some new ideas we’ve been looking at for educating our internal and external developers on JDeveloper and Oracle ADF: The Tools of Fusion. I don’t know what I need to know! Now, I am going to start by giving you a nugget of information that might seem obvious, but it has been a factor in many AUTUMN 13 Technology
  36. 36. Technology: Grant Ronald 39 of the customer escalations or problems that me and my team get involved in when in our reactive mode: you need time to learn. I know, not an earth shattering revelation but worth stating up front. Does that mean Oracle ADF is difficult to learn? Not particularly, but would you let someone who has only done 5 days of a database course be your enterprise-wide DBA? No, you wouldn’t; but they could still be productive so long as you give them the right job to do and they can develop their skills in a more “hands-on” mode. And Oracle ADF isn’t any different. You can attend a course or read a book, and you’ll be able to build applications, but you need time to develop those real-world skills and the more you develop those skills the more you get from the power of the framework and technologies. But where do you start and what are the base skills you need? Do I need to know Java? The most common question I hear is “do I need to know Java?”It’s like asking an Oracle Forms developer “do I need to know PL/SQL”. Oracle ADF is a Java framework so the clue is in the name and so yes, you should have some Java programming skills. The key point with Oracle ADF is how much of the Java programming language you have to know. Many of the concepts you might already know from PL/SQL, Ada, or whatever language you already know, can be easily mapped to Java. Data types, declaring variables, assigning values, conditional statements and loops are all pretty much of a muchness and are the staples of Oracle ADF coding, so it’s a case of picking up the new syntax and some of the nuances of the new language. And because Oracle ADF is taking care of so much of the low- level “plumbing”you’ll be excused from having to discover and learn many of the platform APIs (such as JDBC) because that functionality can be achieved by using the declarative elements of the framework. Also, the more advanced features of the Java programming language such as concurrently/multi-threading are unlikely to darken the door of most of your developers. So how much Java is enough? Personally, coming from a PL/SQL, ADA and Pascal background, I found that “Head First Java” published by O’Reilly gave me enough in a couple of days to get me up and running. Of course that is an absolute minimum and I was prone to write some pretty mangled Java code but there again, I was no different when I started out with PL/ SQL. There is no shortage of books or online material if you are completely new to the Java programming language and probably the main mind shift is thinking in a more object focused, rather than procedural way. But of course I’ve assumed you or your team don’t already know Java. On the other hand there are literally millions of Java developers so it may be that finding programmers with knowledge of the Java language isn’t the challenge. What about the Java platform? So, I’ve talked about the Java programming language, but Java is also a platform. The Java Enterprise Edition (Java EE) is a platform that includes hundreds of APIs to implement standards such as database connectivity, security and user interfaces. Do you have to learn those? As noted before, it’s not so much “do you have to know”but more “how much”. You may never have to write a line of JDBC, but knowing what it is and what it does will be a help. Java Server Faces (JSF) is probably the key Java EE skill on which you should have some knowledge. Whilst ADF Faces is based on JSF and protects you from much of the repetitious and boring stuff, the fact that a page has a lifecycle and that data can be kept in memory for different lengths of time are part of the workings of JSF that would be helpful for you to understand. And now I learn ADF? So the Java language and platform are skills that you need some experience in, but the focus of your learning is here: Oracle ADF. There are a number of building blocks on which you will focus your learning: • JDeveloper – you might be new to Java or maybe you are an Eclipse developer, but you will have to learn your way around the IDE. Where are the preferences, how does the debugger work, how do I refactor files, etc. • Architecture – you need to know the broad building blocks of both ADF and MVC (Model-View-Controller) architecture. Also you need to know how you might structure an application, package it up, how to reuse parts of it etc. • Declarative building blocks – don’t fight the framework, work with it. If the framework provides a declarative feature then use that rather than writing code yourself. In the end, life will be so much easier if you do, you just have to discover that declarative feature. • Oracle ADF APIs – at some point you will be writing code and if you want to interact with the various declarative building blocks then you will be coding against their APIs, so you need awareness of those APIs. What is there to help me learn? Learning anything is a blend of knowledge and experiences all coming from different sources and learning Oracle ADF is no different. Now, I won’t try to cover all sources of learning (although I’ll give you some essential links at the end) but what I will do is cover some of the more popular sources that I’ve seen used to good effect in the community. In my opinion, it is the mix of different types of delivery medium that can be the key to successful learning. And rather than writing long and unwieldy links in the middle of a sentence (which may be useless if you are reading a paper copy of this article) I suggest googling the terms in black italics. As way of a disclaimer I will point out I mention my own book and some books written by colleagues but I do so as genuine recommended reading rather than any selfish goal of knocking JK Rowling off her perch!