IMS v12 Performance


Published on

Published in: Technology
1 Like
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide
  • IBM IOD 2011 10/17/11 Prensenter name here.ppt 10/17/11 19:36
  • IBM IOD 2011 10/17/11 Prensenter name here.ppt 10/17/11 19:36
  • IMS v12 Performance

    1. 1. IMS V12 Performance Enhancements Dave Viguers IMS Performance
    2. 2. <ul>Please Note: </ul><ul>IBM ’s statements regarding its plans, directions, and intent are subject to change or withdrawal without notice at IBM’s sole discretion. Information regarding potential future products is intended to outline our general product direction and it should not be relied on in making a purchasing decision. The information mentioned regarding potential future products is not a commitment, promise, or legal obligation to deliver any material, code or functionality. Information about potential future products may not be incorporated into any contract. The development, release, and timing of any future features or functionality described for our products remains at our sole discretion. </ul><ul>Performance is based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput or performance that any user will experience will vary depending upon many factors, including considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve results similar to those stated here. </ul>
    3. 3. <ul> Acknowledgements and Disclaimers : </ul><ul>© Copyright IBM Corporation 2011. All rights reserved. </ul><ul><ul><li>U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. </li></ul></ul><ul>IBM, the IBM logo,, Information Management, IMS, CICS, WebSphere Application Server and z/OS are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at Other company, product, or service names may be trademarks or service marks of others. </ul><ul>Availability . References in this presentation to IBM products, programs, or services do not imply that they will be available in all countries in which IBM operates. </ul><ul>The workshops, sessions and materials have been prepared by IBM or the session speakers and reflect their own views. They are provided for informational purposes only, and are neither intended to, nor shall have the effect of being, legal or other guidance or advice to any participant. While efforts were made to verify the completeness and accuracy of the information contained in this presentation, it is provided AS-IS without warranty of any kind, express or implied. IBM shall not be responsible for any damages arising out of the use of, or otherwise related to, this presentation or any other materials. Nothing contained in this presentation is intended to, nor shall have the effect of, creating any warranties or representations from IBM or its suppliers or licensors, or altering the terms and conditions of the applicable license agreement governing the use of IBM software. </ul><ul>All customer examples described are presented as illustrations of how those customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics may vary by customer. Nothing contained in these materials is intended to, nor shall have the effect of, stating or implying that any activities undertaken by you will result in any specific sales, revenue growth or other results. </ul>
    4. 4. Performance related Items in V12 <ul><li>Internal instruction path optimization
    5. 5. Logging enhancements
    6. 6. DBCTL TCB switch reduction
    7. 7. Fast Path Secondary Index support
    8. 8. APPC and OTMA synchronous shared queues enhancement
    9. 9. OTMA ACEE reduction
    10. 10. Dynamic buffer pool capability for full function
    11. 11. TCP/IP support for MSC
    12. 12. FP 64 bit buffer manager enhancements
    13. 13. And more..... </li></ul>
    14. 14. Measurement Methodology <ul><li>Setup and customize environment </li><ul><li>#CP's, IMS workload, Regions, etc. </li></ul><li>Execute measurement </li><ul><li>Run workload until stable and gather data </li></ul><li>Analyze results </li><ul><li>Typically IMSPA, APA and/or HIS </li></ul><li>Repeat </li><ul><li>Further tuning or adjustments
    15. 15. Be sure results are repeatable </li></ul></ul>
    16. 16. Measurement Environment <ul><li>IBM z10 and z196 processors </li><ul><li>Multiple LPAR's
    17. 17. ICF's </li></ul><li>8300 and 8700 DASD
    18. 18. ESCON and FICON channels
    19. 19. 128 MB/sec OSA cards
    20. 20. z/OS V1R11
    21. 21. IMS V11 an IMS V12 </li></ul>
    22. 22. Internal path optimization <ul><li>Updated compiler </li><ul><li>Take advantage of new hardware instructions </li></ul><li>Restructure some high usage modules </li><ul><li>Reduce cache misses </li></ul><li>Remove or change expensive instructions </li><ul><li>Sometimes one instruction can make a difference </li></ul><li>General review of many other processes </li></ul>
    23. 23. Base function results <ul><li>Several types of workloads </li><ul><li>Full function
    24. 24. Fast Path
    25. 25. SMQ
    26. 26. Data Sharing </li></ul><li>All performed equal or better than V11 </li><ul><li>Offset additional path of new function </li></ul></ul>
    27. 27. Logging <ul><li>What changed? </li><ul><li>OLDS and WADS now extended format
    28. 28. OLDS and SLDS Striping for more bandwidth
    29. 29. OLDS buffers above the bar </li><ul><li>Real was already 64 bit
    30. 30. Virtual storage now 64 bit instead of ECSA </li></ul><li>WADS channel program rewrite </li><ul><li>Primarily to benefit mirroring environment
    31. 31. Also benefits non-mirrored systems </li></ul></ul></ul>
    32. 32. OLDS Buffers <ul><li>Buffers and control blocks moved from ECSA to 64 bit
    33. 33. ECSA savings should be slightly more than the buffer size time the number of buffers </li><ul><li>Test with 2516 buffers of 24K = 64MB reduction in ECSA
    34. 34. So 100 buffers would be about 2.5MB savings
    35. 35. And 5000 would be around 120MB savings
    36. 36. And so on </li></ul></ul>
    37. 37. OLDS Striping <ul><li>Can now allocate multiple stripes </li><ul><li>OLDS and SLDS </li></ul><li>No longer constrained to a single device
    38. 38. Could help heavy BMP times
    39. 39. May help online also with heavy logger demand
    40. 40. Measured over 350 MB per second on DS8700 </li><ul><li>Note: DS8700 achieved about 35% more bandwidth than DS8300 </li></ul></ul>
    41. 41. Striping Results
    42. 42. WADS enhancement <ul><li>Channel program rewrite </li><ul><li>Reduce impact on mirroring environments </li><ul><li>Metro and Global </li></ul><li>May also help in non-mirrored systems
    43. 43. Response time may be lower </li></ul><li>V11 and V10 enhancement also </li><ul><li>Reduce I/O time impact if record not in cache
    44. 44. Added by APAR to V11 and V10
    45. 45. Requires OA33810 and DS8700 R6.1 microcode </li></ul></ul>
    46. 46. DBCTL TCB Switch Reduction <ul><li>AKA – threadsafe
    47. 47. Application must be defined as Threadsafe
    48. 48. Must use CICS TS V4.2
    49. 49. Early measurements by Hursley </li><ul><li>More than 20% cpu savings </li></ul><li>IMS measurements in progress </li></ul>
    50. 50. Fast Path Secondary Index <ul><li>DEDB's can now have indexes defined </li><ul><li>IMS will maintain integrity of updates </li></ul><li>FP DEDB's generally use less cpu than FF </li><ul><li>But many users require indexing thus </li><ul><li>Could not take advantage of DEDB's </li></ul></ul><li>Savings will depend upon may factors </li><ul><li>How many DB's could be converted
    51. 51. How much of the workload accesses those DB's
    52. 52. And on and on
    53. 53. But could be substantial </li></ul></ul>
    54. 54. FP 64 bit buffer manager <ul><li>If you really like using DBBF, DBFX, BSIZE and ECSA you may continue to do so. </li><ul><li>For us old timers that don't like change </li></ul><li>V12 64 bit is enhanced to: </li><ul><li>Pre-extend pools rather than wait for full condition
    55. 55. Compress pools when demand decreases </li></ul><li>Some ECSA still required </li><ul><li>For control blocks </li></ul></ul>
    56. 56. OTMA ACEE Reduction <ul><li>Prior to V12 </li><ul><li>Each client connected to IMS could potentially have the same users (ACEE's)
    57. 57. ACEE was created for each client </li></ul><li>V12 now eliminates duplicates </li><ul><li>One ACEE per user regardless of # clients
    58. 58. Less storage usage
    59. 59. Fewer hash table synonyms </li><ul><li>Less cpu to search </li></ul></ul></ul>
    60. 60. Synchronous SQ Enhancement <ul><li>IMS will use XCF instead of RRS </li><ul><li>Selected by AOS=X parameter </li></ul><li>Applies to OTMA CM1 and synchronous APPC </li><ul><li>Sync level none or confirm </li><ul><li>Syncpoint will still require RRS </li></ul></ul><li>Simplifies implementation
    61. 61. Improves performance </li></ul>
    62. 62. SMQ improvements <ul><li>FE/BE with BE processing 100% </li><ul><li>OTMA SYNCLVL=NONE </li><ul><li>63% reduction in cpu </li></ul><li>OTMA SYNCLVL=CONFIRM </li><ul><li>81% reduction </li></ul></ul><li>FE/BE both processing </li><ul><li>OTMA SYNCLVL=NONE </li><ul><li>27% reduction </li></ul></ul></ul>
    63. 63. Full Function Dynamic Buffer Pool <ul><li>Dynamically change number and size of buffers </li><ul><li>Update Pool command </li><ul><li>Increase or decrease
    64. 64. OSAM or VSAM </li></ul></ul><li>Probably want to make any change when volume is low </li><ul><li>However it can be done at any time
    65. 65. Might save an outage </li></ul><li>OSAM impact much lower than VSAM </li></ul>
    66. 66. Some other Performance Enhancements <ul><li>MSC can now use TCP/IP in addition to VTAM, CTC, and MTM
    67. 67. IMS Connect RACF userid caching
    68. 68. Additional pools support 64 bit real
    69. 69. Member OLC option to only process PSB
    70. 70. zIIP support for CQS
    71. 71. Logger and FP avoid storage movement </li></ul>
    72. 72. Summary <ul><li>V12 should perform equal or better than V11
    73. 73. Some functions perform significantly better </li><ul><li>Synchronous shared queues
    74. 74. FPSI
    75. 75. Logging capacity
    76. 76. DBCTL
    77. 77. MSC </li></ul><li>Virtual Storage reduction
    78. 78. V12 Performance paper coming soon! </li></ul>