IBM Technical Brief. IBM System z9 ziip Measurements: SAP OLTP, BI Batch, SAP BW Query, and DB2 Utility Workloads. Authors:

Size: px
Start display at page:

Download "IBM Technical Brief. IBM System z9 ziip Measurements: SAP OLTP, BI Batch, SAP BW Query, and DB2 Utility Workloads. Authors:"

Transcription

1 IBM Technical Brief IBM System z9 ziip Measurements: SAP OLTP, BI Batch, SAP BW Query, and DB2 Utility Workloads Authors: Seewah Chan Veng K. Ly Mai N. Nguyen Howard E. Poole Michael R. Sheets Akira Shibamiya Document Owner: Michael R. Sheets IBM Americas EAS Advanced Technical Support Version: 1.00 Date: October 2, 2006 Filename: SAP_zIIP_System z.pdf Copyright IBM Corporation 2006 All Rights Reserved

2 TABLE OF CONTENTS 1 INTRODUCTION REMINDERS TRADEMARKS FEEDBACK ACKNOWLEDGEMENTS WORKLOAD AND MEASUREMENT BACKGROUND SAP OLTP Workload BI Batch Workload SAP BW Query Workload DB2 Utilities Workload CONFIGURATIONS Hardware System and Software Environment MEASUREMENT RESULTS SAP OLTP Workload BI Batch Workload SAP BW Query Workload DB2 Utilities Workload ANALYSIS SAP OLTP Workload BI Batch Workload SAP BW Query Workload...21 Copyright IBM Corporation 2006 All Rights Reserved Page 2 of 28

3 9.4 DB2 Utilities Workload CONCLUSIONS REFERENCES TABLES Table 1: SAP BW Query Workload Steps...6 FIGURES Figure 1: SAP OLTP Configuration...9 Figure 2: BI Batch Configuration...10 Figure 3: SAP BW Query Configuration...11 Figure 4: ziip Effects - SAP OLTP...20 Figure 5: ziip Effects - BI Batch...21 Figure 6: ziip Effects - SAP BW Query...22 Figure 7: ziip Effects - DB2 Utilities Load Table...23 Figure 8: ziip Effects - DB2 Utilities Load Partition...24 Figure 9: ziip Effects - DB2 Utilities Reorg Tablespace...24 Figure 10: ziip Effects - DB2 Utilities Reorg Partition...25 Figure 11: ziip Effects - DB2 Utilities Rebuild Index...25 Figure 12: ziip Effects - DB2 Utilities Rebuild Logical Partition of a Non-Partitioning Index and Reorg Index...26 Copyright IBM Corporation 2006 All Rights Reserved Page 3 of 28

4 1 Introduction IBM recently introduced the System z9 Integrated Information Processor (ziip) as a specialty engine to help improve resource utilization and lower the cost of portions of eligible workloads. ziips require at least DB2 for z/os V8, a z9 system, z/os 1.6, and, for maximum exploitation in an SAP environment, DB2 Connect. They can also imply specific application software levels such as in the case of SAP. ziips are especially intended for data serving workloads. More information about ziips and data serving can be found at [1, 2]. More information about DB2 for z/os V8 can be found at [3, 4, 5]. This paper describes some measurements done in IBM s Laboratories utilizing z9 ziip hardware and software. Four types of DB2 V8 workloads were measured: SAP OLTP exploiting Distributed Relational Database Architecture (DRDA using DB2 Connect) Business Intelligence (BI) batch direct attached SAP BW query exploiting DRDA DB2 Utility The purposes of these measurements were to: test these functions quantify the degree of redirect to ziips in different environments see if there are any CP costs 2 Reminders Neither this document nor any part of it may be copied or reproduced in any form or by any means or translated into another language, without the prior consent of the IBM Corporation. IBM makes no warranties or representations with respect to the content hereof and specifically disclaims any implied warranties of merchantability or fitness of any particular purpose. IBM assumes no responsibility for any errors that may appear in this document. The information contained in this document is subject to change without any notice. IBM reserves the right to make any such changes without obligation to notify any person of such revision or changes. IBM makes no commitment to keep the information contained herein up to date. 3 Trademarks SAP, R/3, and all SAP product and service names mentioned herein are trademarks or registered trademarks of SAP AG in Germany and several other countries around the world. IBM, IBM ^, iseries, pseries, xseries, zseries, DB2, Universal Database, z/os, z/vm, DB2, Enterprise Storage Server, TotalStorage, FICON, ESCON, and PR/SM are trademarks or registered trademarks of the International Business Machines Corporation in the United States and other countries. Other products may be trademarks or registered trademarks of their respective companies. 4 Feedback Please send comments or suggestions for changes to msheets@us.ibm.com Copyright IBM Corporation 2006 All Rights Reserved Page 4 of 28

5 5 Acknowledgements The authors would like to recognize the contributions of Gopal Krishnan and Terry Purcell. 6 Workload and Measurement Background As part of IBM s continuous testing and measuring of its products, we wanted to investigate the effects of running different workloads with and without ziips. We did not tune these measurements as one might for a benchmark. Initially we started this effort utilizing prototype hardware and software. But the results shown here represent the final designs and implementation. 6.1 SAP OLTP Workload One of the workloads we measured was SAP AG s mysap Database Server doing OLTP type work. We used DB2 Connect to utilize DRDA connected to external AIX SAP application servers submitting work to a DB2 database server from AIX SAP application servers. For this set of measurements, we used a standard Sales and Distribution (SD) workload. However, we did not run the SAP Sales and Distribution Standard Application Benchmark. While there may be some similarity in business functions performed, the measurements we did were stress tests not SAP certified benchmarks. 6.2 BI Batch Workload We also looked at a unique BI batch workload that was run direct attached. That is, there was no use of DRDA. The native SQL came from a script run in the same z/os as DB2. We used a specially developed query set. While this set is based on field experience and observation of customer environments, it was not intended to represent typical customer queries. Rather it was intended to be complex, challenging, and stressful to DB2. It exhibits higher query parallelism than most customers will experience. While the SAP application is not used, SAP BW data, structure, tables and database are. The query load was run against an SAP BW database with F fact table and dimension tables. The queries are run as a locally attached (via DSNTEP4, not DRDA) job. It is a single stream where each query is run sequentially. There are 79 queries, with three additional environmental settings: set SQLID=SAPR3, set current degree=any, and set current timestamp. There is a mixture of CPU bound queries and I/O bound queries. The F fact table was manually partitioned via DB2 to get more parallelism increasing the ziip redirect. It is important to note that the partitioning scheme mirrors that provided by SAP for the F fact table. After partitioning the F fact table, the DB2 Explain tool indicates 57 of the 79 queries are parallel. Copyright IBM Corporation 2006 All Rights Reserved Page 5 of 28

6 6.3 SAP BW Query Workload This third workload employs multiple streams (or multiple concurrent users) submitting queries to DB2 database server via DRDA from AIX SAP application servers. Each user submits three queries per loop (see Table 1 below). These queries result in ten SQL select statements. Each loop has nine of these selects going against F table and one going against the Operational Data Store (ODS) table. About 500+ SAP aggregates are used to provide better average response time. The ODS is a non-star schema type table. In terms of CPU time, the ODS query dominates. Since SAP aggregates are not supported for ODS, the DB2 Materialized Query Table (MQT) provides effective aggregate functions. Further, partitioning of the MQTs reduces the average response time to sub-second. The result is that this workload has very short-running queries that are less complex than the above BI Batch Workload s 79 queries. Step Operation 1 Query 1: Select a year 2 Within the year, drill down to a specific country 3 Within the country, drill down to a specific sales organization 4 Within the sales organization, drill down to distribution channel 5 Within the distribution channel, drill down to type, version, and material number 6 Change material number and customer number 7 Start new query including formulas 8 Query 2: Perform an ODS query limited to customer, distribution channel, sales organization 9 Query 3: Similar to query 1 for country, sales organization, and distribution channel 10 Within this query, expand to material level two 11 Expand to level three Table 1: SAP BW Query Workload Steps We did not run the SAP BW Standard Application Benchmark. While there may be some similarity in business functions performed, the measurements we did were stress tests not SAP certified benchmarks. 6.4 DB2 Utilities Workload The fourth set of workload runs are related to DB2 V8 Utilities [6]. DB2 build index processing in the Load, Reorg, and Rebuild Index utility can be redirected to ziip in DB2 for z/os V8. These measurements were conducted to understand the performance characteristics of Build index redirect, such as the percentage of redirect CPU time of Load table, Load partition, Reorg tablespace, Reorg partition, Rebuild Index, and Reorg Index with different number of indexes as well as the additional cost of redirect. We also had the opportunity with these measurements to test the ProjectCPU function to see what costs that might have. All of these measurements involved a non-compressed table with 20 million rows with 10 partitions and 1, 2, or 6 indexes. Note that some customers have some tables that are much larger such as SAP customers. Copyright IBM Corporation 2006 All Rights Reserved Page 6 of 28

7 7 Configurations 7.1 Hardware System z DB Server: These measurements were done on a two different IBM System z9 Enterprise Class systems. The SAP OLTP, BI Batch, and SAP BW Query measurements were done on a z9 S with four ziips configured and a total of 128 GB installed. That is, the regular CPs and ziip CPs were in the same book. This is the maximum number of ziips that can be configured on this model. The DB server resided on a z/os LPAR with 4 dedicated CPs plus 4 ziips and 16 GB central memory. The base measurements without ziip were done with the ziip processors varied offline. The DB2 V8 Utility measurements were done on a z9 S partition with four CPs and a total of 8 GB of storage. Initial runs were made without ziips configured. Later two ziips were configured and varied online. More ziips could have been configured on this model. The ziip CPs were not necessarily in the same book as the regular CPs. Each of these systems had the requisite service levels installed to support ziips. Customers should consult their local IBM representative or [7] to make sure they have the proper service levels on their system. Database DASD: The SAP OLTP and BI Batch workload measurements used one Enterprise Storage server (ESS or Shark) Model 800 ( ) with a total of 8 FICON attachments to the System z. The was a Turbo I model with 16 8-packs of 15K RPM 36.4GB drives. The actual capacity was volumes. It had 16 GB regular cache, 2 GB non-volatile storage (NVS) and 8 short wave FICON ports. Additionally there were 8 ESCON ports. However, ESCON was not used for these runs. The 8 FICON ports were connected via two MCDATA FICON directors to the z The host used FICON Express4, however, this Model 800 used only 2 Gbps per port. We exploited the new Modified Indirect Data Address Word (MIDAW) CCW function for improved I/O responsiveness. The SAP BW Query measurements used one dual frame IBM TotalStorage server Model DS8300 ( /92E) with a total of eight FICON attachments to the System z. The /92E coupled unit has 48 8-packs of 15K RPM 73 GB drives. The actual capacity was volumes. It had 128 GB regular cache, 4 GB non-volatile storage (NVS) and 32 long wave FICON ports. However, only eight FICON ports were used for the test and they were direct connected to the z The host used FICON Express4, however, this DS8300 used only 2 Gbps per port. We exploited the new Modified Indirect Data Address Word (MIDAW) CCW function for improved I/O responsiveness. DB2 V8 Utility measurements used one ESS Model 800 ( ) with 4 FICON Express2 attachment to the System z. The was a Turbo I model. The database and workload were spread on 34 volumes with additional 14 volumes for workfiles. There were 16GB regular cache, 2 GB non-volatile storage and eight short wave FICON ports. Copyright IBM Corporation 2006 All Rights Reserved Page 7 of 28

8 pseries Application Servers: The SAP OLTP workload used three p690s each with GHz processors and 128 GB of memory. No pseries servers were used with the BI Batch workloads. The SAP BW Query workload used one p570 with GHz processors and one p690 with GHz processors - each has 128 GB of memory. No pseries servers were used with the DB2 V8 Utility workloads. pseries Presentation Servers: The SAP OLTP workload used one p620 with four 600 MHz processors and 4 GB of memory. No pseries servers were used with the BI Batch workloads. The SAP BW Query workload used one p630 with four 1.0 GHz processors and 8 GB of memory as well as one p55a with four 1.9 GHz processors and 32 GB of memory. No pseries servers were used with the DB2 V8 Utility workloads. Network: Gigabit Ethernet networks were used for all network connections. Each of the p690/p570 application servers was connected to the z9-704 via OSA-Express2 adapter. Below are conceptual views of the configurations used. SAP OLTP is shown in Figure 1 Business Intelligence (BI) is shown in Figure 2 SAP BW Query is shown in Figure 3 DB2 Utility is similar to Figure 2 Copyright IBM Corporation 2006 All Rights Reserved Page 8 of 28

9 ziip Measurements: SAP OLTP Configuration w/ 16 8-packs of 15K RPM 36.4 GB drives. 16 GB regular cache + 2 GB non-volatile storage (NVS) Capacity: volumes Enterprise Storage Server: ( ) Two MCDATA FICON Directors 8 FICON connections 8 FICON connections Hardware: DB2 v8 Database Server: One z/os LPAR w/ 4 dedicated CPs plus 4 ziips, 16 GB central memory Software: zos DB2 v8.1 GA Z9-704 (2094) 128 GB Gigabit Ethernet Switch 3 p690 Application Servers (Each) GHz processors/128 GB memory - AIX SAP Kernel DB2 Connect v8.1 FixPak 9 1 p620 (4 600 MHz processors/4 GB memory) Presentation Server Figure 1: SAP OLTP Configuration Copyright IBM Corporation 2006 All Rights Reserved Page 9 of 28

10 ziip Measurements: BI Batch Configuration w/ 16 8-packs of 15K RPM 36.4 GB drives. 16 GB regular cache + 2 GB non-volatile storage (NVS) Capacity: volumes Enterprise Storage Server: Model 800 ( ) 8 FICON connections Two MCDATA FICON Directors 8 FICON connections Hardware: DB2 v8 Database Server: One z/os LPAR w/ 4 dedicated CPs plus 4 ziips, 16 GB central memory Software: zos DB2 v8.1 GA Z9-704 ( ) 128 GB Workstation Batch Job Submission: BI Parallel Queries Figure 2: BI Batch Configuration Copyright IBM Corporation 2006 All Rights Reserved Page 10 of 28

11 ziip Measurements: SAP BW Query Configuration /92E w/ 48 8-packs of 15K RPM 73 GB drives. 128 GB regular cache + 4 GB non-volatile storage (NVS) Capacity: volumes IBM TotalStorage Server Model DS8300: ( /92E) 8 FICON direct connections Hardware: DB2 v8 Database Server: One z/os LPAR w/ 4 dedicated CPs plus 4 ziips, 16 GB central memory Software: zos DB2 v8.1 GA BW 3.5 database Z9-704 ( ) 128 GB Gigabit Ethernet Switch 2 p620 (4 CPs) Presentation Servers p GHz processors/8 GB RAM p55a GHz processors/32 GB RAM Each w/ - SAP standard BW benchmark driver 2 Application Servers p GHz processors/ 128 GB RAM P GHz processors/ 128 GB RAM Each w/ - AIX SAP Kernel 6.40 w/ FDPR - DB2 Connect v8.1 FixPak 11 Figure 3: SAP BW Query Configuration Copyright IBM Corporation 2006 All Rights Reserved Page 11 of 28

12 7.2 System and Software Environment z/os MVS z/os release DB2 V8.1 GA with PTFs UK16616 and UK At the time this paper was written these were the most up to date DB2 implementation for Parallel Query and DB2 Utility ziip exploitation. DB2 Connect: V8.1 FixPak 9 Because the ziip environment continues to evolve, customers should consult their local IBM representative, [8], and DB2 Information APAR II14219 to make sure they have the proper service levels on their system. AIX AIX: 5.2 SAP Application Levels NetWeaver 04 SR1 BW 3.5 SP15 SAP disp + work information SAP kernel release 6.40 kernel make variant 6.40 compiled on AIX A4A4C00 compiled for 64 BIT compile time Feb :40:10 update level 0 patch number 58 source id Copyright IBM Corporation 2006 All Rights Reserved Page 12 of 28

13 8 Measurement Results During the course of this effort many measurement runs were done. Some were to get familiar with the environment and the workload. Some were for tuning. Some were for experimentation with different designs and implementations. It is beyond the scope of this paper to show them all. Listed below are the measurement results selected as utilizing the final design/implementation and the most useful, given the time constraints of this effort. The DB2 accounting information, shown in the BI Batch results, is from DB2 Performance Expert for z/os. More information is at [9]. 8.1 SAP OLTP Workload No ziip ziip Runid S60508G2 S60508G1 # of SD users 4,104 4,104 # of regular CPs 4 4 # of ziip Processors 0 4 Regular CP busy % % % ziip CP busy % 0.00% % Total CP busy % % % ETR ms/tr ms/tr regular CP ms/tr ziip Av response time (sec) ITR (based on CPs) Copyright IBM Corporation 2006 All Rights Reserved Page 13 of 28

14 8.2 BI Batch Workload No ziip ziip Runid S60719W3 S60719W2 # of regular CPs 4 4 # of ziip Processors 0 4 Batch elapsed time (from job output) Batch CPU time (regular CP from job output) 4,347 4,282 2,196 1,123 DB2 acctg: Elapsed time (class 1) 4, ,278.4 CPU time (class 2) 2, Parallel Task CPU time (class 2) 1, IIPCP CPU (class1) 1, IIP CPU time (class 1) 0 1,079.7 Batch CPU - class 2 CPU time WLM info for batch Service Class: CPU (sec) IIP (sec) CPU - IIP (sec) SAP BW Query Workload No ziip ziip Runid S60719P2 S60720P1 # of BW users # of regular CPs 4 4 # of ziip Processors 0 4 Regular CP busy % % % ziip CP busy % 0.00% % Total CP busy % % % ETR ms/tr ms/tr regular CP ms/tr ziip Av response time (sec) ITR (based on CPs) Copyright IBM Corporation 2006 All Rights Reserved Page 14 of 28

15 8.4 DB2 Utilities Workload 1. Load Table 6 indexes No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) indexes No ziip No ziip ziip Elapsed time (sec) 116 No 120 Total CPU time (sec) 136 Measurement 117 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available 27 1 index No ziip No ziip ziip Project CPU No Yes Elapsed time (sec) 100 No 103 Total CPU time (sec) 70 Measurement 60 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available Load Partition 6 indexes No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) indexes No ziip No ziip ziip Copyright IBM Corporation 2006 All Rights Reserved Page 15 of 28

16 Elapsed time (sec) 28 No 30 Total CPU time (sec) 27 Measurement 11 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available 18 1 index No ziip No ziip ziip Elapsed time (sec) 10 No 10 Total CPU time (sec) 7.2 Measurement 6.0 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available Reorg Tablepsace 6 indexes No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) indexes No ziip No ziip ziip Elapsed time (sec) 242 No 246 Total CPU time (sec) 124 Measurement 107 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available 27 1 index No ziip No ziip ziip Elapsed time (sec) 239 No 244 Total CPU time (sec) 94 Measurement 86 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available 15 Copyright IBM Corporation 2006 All Rights Reserved Page 16 of 28

17 4. Reorg Partition 6 indexes No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) indexes No ziip No ziip ziip Elapsed time (sec) 37 No 39 Total CPU time (sec) 24 Measurement 10 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available 17 1 index No ziip No ziip ziip Elapsed time (sec) 24 No 24 Total CPU time (sec) 9.4 Measurement 8.5 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available Rebuild Index 6 indexes No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) indexes No ziip No ziip ziip Copyright IBM Corporation 2006 All Rights Reserved Page 17 of 28

18 Elapsed time (sec) 66 No 71 Total CPU time (sec) 100 Measurement 78 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available 29 1 index No ziip No ziip ziip Elapsed time (sec) 22 No 22 Total CPU time (sec) 50 Measurement 48 IIPCP CPU time (sec) 0 Data 0 IIP CPU time (sec) 0 Available Rebuild a logical partition of non-partitioning index 1 index No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) Reorg Index 1 index No ziip No ziip ziip Elapsed time (sec) Total CPU time (sec) IIPCP CPU time (sec) IIP CPU time (sec) Copyright IBM Corporation 2006 All Rights Reserved Page 18 of 28

19 9 Analysis Analysis of various aspects of these measurements was done. In some cases this analysis used standard Large Systems Performance Reference (LSPR) techniques and methodology [10]. It is assumed the reader is familiar with the terms therein, especially External Throughput Rate (ETR), Internal Throughput Rate (ITR) and the corresponding ratios (ETRR and ITRR). In this paper we defined ITR as being based on regular CP utilization. That is, it did not take into account ziip utilizations. 9.1 SAP OLTP Workload This workload exploited ziips due to its use of DRDA. There was no use of the ProjectCPU=Yes function. See DB2 Utilities Workload on page 22 for more information on this. There was enough ziip CPs so that all ziip eligible work was redirected. There were no stored procedures used. No DB2 accounting information was collected because of the volume of data generated by the tool for this workload. The results show several interesting items. Some are shown graphically in Figure 4 below. First, in the ziip case, the total milliseconds CP time per transaction was about 1% less. This indicates that ziip exploitation is about as efficient in terms of total CP time as non-ziip. Second, the response time of the ziip run was much better. This is due to the fact that the ziip measurement had a total of eight CPs of relatively low utilization compared to the non-ziip case which had only four total CPs at a much higher utilization. However, in both cases, the response time was sub-second. The regular CP busy time dropped to almost half. Similarly, ITR based on regular CPs about doubled. This reflects the fact that the configuration with ziips has about twice the throughput potential. Finally, the ziip measurement showed that 49.8% of the entire LPAR was redirected to the ziips. This includes both captured, uncaptured, and LPAR management times. However, this was not what we would expect of a typical production environment. More normal production environments should expect lower redirect percentages. See Conclusions on page 27. Copyright IBM Corporation 2006 All Rights Reserved Page 19 of 28

20 Figure 4: ziip Effects - SAP OLTP 9.2 BI Batch Workload In this workload there was no use of DRDA. There was no use of the ProjectCPU=Yes function. See DB2 Utilities Workload on page 22 for more information on this. There was enough ziip CPs so that all ziip eligible work was redirected. In addition, there were no stored procedures used. This is because the batch application was directly attached to DB2 in the same z/os image. The exploitation of ziips in this particular workload came entirely from its use of parallel queries. Because this workload was batch, and therefore had relatively few transactions, we ran DB2 accounting. The results show several interesting items. Some are shown graphically in Figure 5: ziip Effects - BI Batch below. Once again, the total CPU time is virtually the same with and without ziips. In this case the total elapsed time was only slightly less with ziips. In contrast to the previous workload, which was CPU bound, this workload is dominated by waiting for database I/O. The regular CP busy fell 49% with ziips. From the DB2 accounting data, the No ziip run s IIPCP CPU estimate for ziip potential redirect was very close to the ziip run s IIP actual CPU time. Finally, the ziip measurements show that 49.04% of the entire LPAR was redirected to the ziips. Because this data is from WLM, it doesn t count uncaptured time. However, this was not what we would expect of a typical production environment. More normal production environments should expect lower redirect percentages. See Conclusions on page 27. Copyright IBM Corporation 2006 All Rights Reserved Page 20 of 28

21 Figure 5: ziip Effects - BI Batch 9.3 SAP BW Query Workload SAP BW environments have the potential to exploit ziips for two reasons. First, more recent SAP releases use DRDA. Further, SAP BW environments also have the potential to have Parallel Queries. There was no use of the ProjectCPU=Yes function. See DB2 Utilities Workload on page 22 for more information on this. There was enough ziip CPs so that all ziip eligible work was redirected. There were no stored procedures used. No DB2 accounting information was collected because of the volume of data generated by the tool for this workload. The results show several interesting items. Some are shown graphically in Figure 6: ziip Effects - SAP BW Query below. Again, in the ziip case, the milliseconds CPU per transaction was virtually the same - about 0.5% less. This indicates that ziip exploitation is about as efficient in terms of total CPU use as non-ziip. Second, the response time of the ziip run was 43% better. Again, like the SAP OLTP measurement, the ziip measurement had a total of eight CPs of relatively low utilization compared to the non-ziip case which had only four total CPs at a much higher utilization. The regular CP busy time dropped 46%. However, because with ziips there was more throughput, the ITR based on regular CPs doubled. This reflects the fact that the configuration with ziips has about twice the throughput potential. Finally, the ziip measurement showed that 50.1% of the entire LPAR was redirected to the ziips. This includes both captured, uncaptured, and LPAR management times. This is very close to the same redirect as the SAP OLTP workload. Normally one might expect BW with DRDA to get more redirect than regular SAP with DRDA. However, as mentioned in SAP BW Query Workload on page 6, this SAP BW Query workload consists of very short queries that get little opportunity to exploit ziips. Copyright IBM Corporation 2006 All Rights Reserved Page 21 of 28

22 However, these results are not what we would expect of a typical production environment. More normal production environments should expect lower redirect percentages. However, parallelism is more likely chosen when queries include partitioned tablespaces, and SAP BW provides the capability to partition both the F and E fact tables. Thus, customers do have the opportunity to increase the likelihood that their queries will run in parallel. See Conclusions on page 27. Figure 6: ziip Effects - SAP BW Query 9.4 DB2 Utilities Workload Shown graphically below in Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, and Figure 12 are ratios of CPU, elapsed time, and regular CP busy of measurements with ziips relative to measurements without ziips for the utility measurements shown previously. The amount of redirect for this configuration is: 5 to 20% for Rebuild Index 10% to 20% for Load or Reorg of a partition with one index only, or Load of entire table, or Reorg of entire tablespace 40% for Rebuild Index of logical partition of non partitioning index 40% to 50% for Reorg Index 30% to 60% for Load or Reorg of a partition with more than one index Variations in percentage redirect for various utilities are primarily determined by the percentage of CPU time consumed by build index processing, which is redirected, to the total CPU time for a given utility. For example, Load or Reorg with more than one partition spends most of the CPU time in build index phase and consequently is in a position to gain the biggest redirect percentage, especially with more indexes. Copyright IBM Corporation 2006 All Rights Reserved Page 22 of 28

23 The ProjectCPU function in WLM enables estimation of how much redirect can occur as reported in IIPCP CPU time when ziip is not available. When ziip is available, there is no difference between ProjectCPU No and Yes. When there is no ziip processor but ProjectCPU is requested in order to estimate the potential redirect, or when ziip is present, a new execution unit switch from utility TCB to an enclave SRB will take place during index build phase of DB2 z/os Load, Reorg, and Rebuild Index for each block of index entries. This newly added processing to enable real or simulated CPU time redirect will typically result in less than 10% increase in the job total CPU time and elapsed time. Most of this added CPU time is actually redirected to ziip when available. This extra CPU and elapsed time tends to go up as the ratio of the number of ziips to the number of CPs goes down. In other words, increasing the ratio of the number of ziips to the number of CPs closer to one tends to result in less extra CPU and elapsed time and a maximized redirect percentage. Even including the ProjectCPU costs, we saw a net reduction in regular CP time. This ranged from 4% to 57% reduction. As one might suspect, more indexes result in more utility CPU time. All of the ziip measurements were with four regular CPs and two ziips. In all the six index measurements, the IIPCP CPU time indicates that there was some overflow to regular CPs and that these measurements could have exploited even more than two ziips. This set of measurements is not intended to represent customer environments. As mentioned earlier, some customers have some tables that are much larger. In some cases these customers have had to carefully plan when to do this sort of utility work. ziips might help this as well as possibly reduce the costs of this work. See Conclusions on page 27. Figure 7: ziip Effects - DB2 Utilities Load Table Copyright IBM Corporation 2006 All Rights Reserved Page 23 of 28

24 Figure 8: ziip Effects - DB2 Utilities Load Partition Figure 9: ziip Effects - DB2 Utilities Reorg Tablespace Copyright IBM Corporation 2006 All Rights Reserved Page 24 of 28

25 Figure 10: ziip Effects - DB2 Utilities Reorg Partition Figure 11: ziip Effects - DB2 Utilities Rebuild Index Copyright IBM Corporation 2006 All Rights Reserved Page 25 of 28

26 Figure 12: ziip Effects - DB2 Utilities Rebuild Logical Partition of a Non-Partitioning Index and Reorg Index Copyright IBM Corporation 2006 All Rights Reserved Page 26 of 28

27 10 Conclusions These measurements show exploitation of ziips in several environments. In all these measurements, the amount of redirect to ziips is more than any possible additional overheads due to ProjectCPU. In some cases there were very significant redirects. One was almost 60%. The number of ziips configured to a system is limited to less than or equal to the number of regular CPs configured on the system. But this is at a system level. Individual LPARs on that system can exploit more ziips than they have regular CPs. These measurements were run in a laboratory environment. We expect real customer DB2 systems would be running more functions related to the care and feeding of a production DB2 database. These functions can range from monitoring and tracing options to service and administrative programs such as DFHSM. These functions do not exploit ziips, and any environment using them will have their potential ziip redirect percentage of an LPAR diluted. Similarly, running application unrelated to DB2 can dilute observed redirect LPAR percentages. As such, customers should understand that these results are limited in scope. They should not be used for sizing. We expect that customer workloads will show different results. For an estimate of your workloads ziip exploitation potential, please contact your local IBM account team. 11 References [1] IBM Corp IBM System z9 Integrated Information Processor (IBM ziip): Frequently Asked Questions from [2] IBM Corp Why Data Serving on a Mainframe from ftp://ftp.software.ibm.com/common/ssi/rep_wh/n/zsw01811usen/zsw01811usen.pdf [3] IBM Corp DB2 UDB for z/os V8: Through the Looking Glass and What SAP Found There (SG ) or from [4] IBM Corp DB2 UDB for z/os Version 8: Everything You Ever Wanted to Know,... and More (SG ) or from [5] General information about DB2 UDB for z/os: [6] General information about DB2 Utilities Suite for z/os: [7] IBM Corp. [8] IBM Corp. Copyright IBM Corporation 2006 All Rights Reserved Page 27 of 28

28 [9] General information about DB2 Performance Expert for z/os on World Wide Web page: [10] IBM Corp Large Systems Performance Reference (SC ) or Copyright IBM Corporation 2006 All Rights Reserved Page 28 of 28

ziip and zaap Software Update

ziip and zaap Software Update ziip and zaap Software Update Overview The System z9 and z10 Integrated Information Processor (ziip) is the latest specialty engine for the IBM System z mainframe. The ziip is designed to help improve

More information

WebSphere Application Server Base Performance

WebSphere Application Server Base Performance WebSphere Application Server Base Performance ii WebSphere Application Server Base Performance Contents WebSphere Application Server Base Performance............. 1 Introduction to the WebSphere Application

More information

Maximizing offload to ziip processors with DB2 9 for z/os native SQL stored procedures

Maximizing offload to ziip processors with DB2 9 for z/os native SQL stored procedures Maximizing offload to ziip processors with DB2 9 for z/os native SQL stored procedures Richard Corrihons IBM Customer Center - PSSC Montpellier, France Introduction This document is based on what has been

More information

Leveraging ziip with DB2 for z/os V8

Leveraging ziip with DB2 for z/os V8 Leveraging ziip with DB2 for z/os V8 Gopal Krishnan DB2 for z/os Performance IBM Silicon Valley Lab, San Jose, CA Session 2218 Trademarks AIX* CICS* DB2* DB2 Connect DB2 Universal Database DRDA* FICON*

More information

IBM zenterprise System:

IBM zenterprise System: IBM zenterprise System: SAP Enqueue Replication into System z Performance Test Report Authors: Seewah Chan Harald Duvenbeck Dr. Paul Lekkas Rose L. Manz Wolfgang Reichert Michael Sheets Stefan Sell Document

More information

IBM Technical Brief. IBM zenterprise System : DB2 11 for z/os with SAP Performance Report. Version 1.0. December 16, 2013.

IBM Technical Brief. IBM zenterprise System : DB2 11 for z/os with SAP Performance Report. Version 1.0. December 16, 2013. IBM Technical Brief IBM zenterprise System : DB2 11 for z/os with SAP Performance Report Version 1.0 December 16, 2013 Authors: Brenda Beane Seewah Chan Akiko Hoshikawa Dr. Paul Lekkas Michael Sheets Document

More information

CA IDMS 18.0 & 18.5 for z/os and ziip

CA IDMS 18.0 & 18.5 for z/os and ziip FREQUENTLY ASKED QUESTIONS CA IDMS 18.0 & 18.5 for z/os and ziip Important October 2013 update ziip (IBM System z Integrated Information Processor) is a specialty mainframe processor designed to help free

More information

Practical Capacity Planning in 2010 zaap and ziip

Practical Capacity Planning in 2010 zaap and ziip Practical Capacity Planning in 2010 zaap and ziip Fabio Massimo Ottaviani EPV Technologies February 2010 1 Introduction When IBM released zaap (2004) and ziip(2006) most companies decided to acquire a

More information

IBM InfoSphere Data Replication s Change Data Capture (CDC) for DB2 LUW databases (Version ) Performance Evaluation and Analysis

IBM InfoSphere Data Replication s Change Data Capture (CDC) for DB2 LUW databases (Version ) Performance Evaluation and Analysis Page 1 IBM InfoSphere Data Replication s Change Data Capture (CDC) for DB2 LUW databases (Version 10.2.1) Performance Evaluation and Analysis 2014 Prasa Urithirakodeeswaran Page 2 Contents Introduction...

More information

IBM Americas Solutions Advanced Technical Skills

IBM Americas Solutions Advanced Technical Skills IBM Americas Solutions Advanced Technical Skills IBM Technical Brief IBM System zenterprise, System Storage, and DB2 10 for z/os : SAP Banking Services 7.0 150 Million Accounts Measurements Authors: Seewah

More information

IBM TotalStorage Enterprise Storage Server Delivers Bluefin Support (SNIA SMIS) with the ESS API, and Enhances Linux Support and Interoperability

IBM TotalStorage Enterprise Storage Server Delivers Bluefin Support (SNIA SMIS) with the ESS API, and Enhances Linux Support and Interoperability Hardware Announcement February 17, 2003 IBM TotalStorage Enterprise Storage Server Delivers Bluefin Support (SNIA SMIS) with the ESS API, and Enhances Linux Support and Interoperability Overview The IBM

More information

IBM TotalStorage Enterprise Storage Server Model RAID 5 and RAID 10 Configurations Running Oracle Database Performance Comparisons

IBM TotalStorage Enterprise Storage Server Model RAID 5 and RAID 10 Configurations Running Oracle Database Performance Comparisons IBM TotalStorage Enterprise Storage Server Model 800 - RAID 5 and RAID 10 Configurations Running Oracle Database Performance Comparisons May 2003 IBM Systems Group Open Storage Systems Laboratory, San

More information

IBM Systems. Oracle and the ziip processor. G. Tom Russell IBM Canada Ltd. MVS Oracle SIG April 30, 2008 Redwood Shores, CA

IBM Systems. Oracle and the ziip processor. G. Tom Russell IBM Canada Ltd. MVS Oracle SIG April 30, 2008 Redwood Shores, CA Oracle and the ziip processor G. Tom Russell IBM Canada Ltd. Tom_Russell@ca.ibm.com MVS Oracle SIG April 30, 2008 Redwood Shores, CA Agenda IBM System z What is a ziip processor? z/os software support

More information

IBM Education Assistance for z/os V2R2

IBM Education Assistance for z/os V2R2 IBM Education Assistance for z/os V2R2 Item: RSM Scalability Element/Component: Real Storage Manager Material current as of May 2015 IBM Presentation Template Full Version Agenda Trademarks Presentation

More information

VM & VSE Tech Conference May Orlando Session M70

VM & VSE Tech Conference May Orlando Session M70 VM & VSE Tech Conference May 2000 - Orlando Session M70 Bill Bitner VM Performance 607-752-6022 bitner@vnet.ibm.com Last Updated: April10,2000 RETURN TO INDEX Disclaimer Legal Stuff The information contained

More information

Enterprise Workload Manager Overview and Implementation

Enterprise Workload Manager Overview and Implementation Enterprise Workload Manager Overview and Implementation Silvio Sasso IBM ITS Delivery for z/os sisa@ch.ibm.com 2006 IBM Corporation Trademarks The following are trademarks of the International Business

More information

Reduce Costs: Getting the Most out of ziips and zaaps with DB2 for z/os

Reduce Costs: Getting the Most out of ziips and zaaps with DB2 for z/os Reduce Costs: Getting the Most out of ziips and zaaps with DB2 for z/os Session Number: 8415 February 28, 2011 Greg Dyck DB2 for z/os Development IBM Silicon Valley Lab, San Jose, CA Disclaimer Copyright

More information

Planning Considerations for Running zaap Work on ziips (ZAAPZIIP) IBM. Kathy Walsh IBM. Version Date: December 3, 2012

Planning Considerations for Running zaap Work on ziips (ZAAPZIIP) IBM. Kathy Walsh IBM. Version Date: December 3, 2012 Planning Considerations for Running zaap Work on ziips (ZAAPZIIP) IBM Kathy Walsh IBM Version Date: December 3, 2012 This document can be found on the web, www.ibm.com/support/techdocs Under the category

More information

SAP NetWeaver BW Performance on IBM i: Comparing SAP BW Aggregates, IBM i DB2 MQTs and SAP BW Accelerator

SAP NetWeaver BW Performance on IBM i: Comparing SAP BW Aggregates, IBM i DB2 MQTs and SAP BW Accelerator SAP NetWeaver BW Performance on IBM i: Comparing SAP BW Aggregates, IBM i DB2 MQTs and SAP BW Accelerator By Susan Bestgen IBM i OS Development, SAP on i Introduction The purpose of this paper is to demonstrate

More information

IBM Client Center z/vm 6.2 Single System Image (SSI) & Life Guest Relocation (LGR) DEMO

IBM Client Center z/vm 6.2 Single System Image (SSI) & Life Guest Relocation (LGR) DEMO Frank Heimes Senior IT Architect fheimes@de.ibm.com 12. Mär 2013 IBM Client Center z/vm 6.2 Single System Image (SSI) & Life Guest Relocation (LGR) DEMO IBM Client Center, Systems and Software, IBM Germany

More information

z/vm Large Memory Linux on System z

z/vm Large Memory Linux on System z December 2007 z/vm Large Memory Linux on System z 1 Table of Contents Objectives... 3 Executive summary... 3 Summary... 3 Hardware equipment and software environment... 4 Host hardware... 5 Network setup...

More information

... IBM Advanced Technical Skills IBM Oracle International Competency Center September 2013

... IBM Advanced Technical Skills IBM Oracle International Competency Center September 2013 Performance benefits of IBM Power Systems and IBM FlashSystem for JD Edwards EnterpriseOne IBM Power 780 server with AIX and IBM FlashSystem 820 flash storage improves batch performance in a client proof

More information

SAP SD Benchmark with DB2 and Red Hat Enterprise Linux 5 on IBM System x3850 M2

SAP SD Benchmark with DB2 and Red Hat Enterprise Linux 5 on IBM System x3850 M2 SAP SD Benchmark using DB2 and Red Hat Enterprise Linux 5 on IBM System x3850 M2 Version 1.0 November 2008 SAP SD Benchmark with DB2 and Red Hat Enterprise Linux 5 on IBM System x3850 M2 1801 Varsity Drive

More information

Stored Procedure Monitoring and Analysis

Stored Procedure Monitoring and Analysis Stored Procedure Monitoring and Analysis Paul Bartak, IBM DB2 Advisor Agenda What are stored procedures? Benefits of stored procedures Stored procedure analysis Issues and solutions Monitoring stored procedures

More information

WSC Experiences with IPSec on the ziip Processor

WSC Experiences with IPSec on the ziip Processor WSC Experiences with IPSec on the ziip Processor v Walt Caprice IBM Washington Systems Center Gaithersburg, MD Trademarks AIX* CICS* DB2* DB2 Connect DB2 Universal Database DRDA* FICON* GDPS* HiperSockets

More information

Expert Stored Procedure Monitoring, Analysis and Tuning on System z

Expert Stored Procedure Monitoring, Analysis and Tuning on System z Expert Stored Procedure Monitoring, Analysis and Tuning on System z Steve Fafard, Product Manager, IBM OMEGAMON XE for DB2 Performance Expert on z/os August 16, 2013 13824 Agenda What are stored procedures?

More information

Leveraging ziip, zaap Specialty Engines with DB2 for z/os

Leveraging ziip, zaap Specialty Engines with DB2 for z/os IBM Software Group Leveraging ziip, zaap Specialty Engines with DB2 for z/os SHARE in Boston Session Number TDZ 2296A Greg Dyck DB2 for z/os Development gad@us.ibm.com August 2010 1 TOPICS ziip & zaap

More information

Why is the CPU Time For a Job so Variable?

Why is the CPU Time For a Job so Variable? Why is the CPU Time For a Job so Variable? Cheryl Watson, Frank Kyne Watson & Walker, Inc. www.watsonwalker.com technical@watsonwalker.com August 5, 2014, Session 15836 Insert Custom Session QR if Desired.

More information

What are the major changes to the z/os V1R13 LSPR?

What are the major changes to the z/os V1R13 LSPR? Prologue - The IBM Large System Performance Reference (LSPR) ratios represent IBM's assessment of relative processor capacity in an unconstrained environment for the specific benchmark workloads and system

More information

HANA Performance. Efficient Speed and Scale-out for Real-time BI

HANA Performance. Efficient Speed and Scale-out for Real-time BI HANA Performance Efficient Speed and Scale-out for Real-time BI 1 HANA Performance: Efficient Speed and Scale-out for Real-time BI Introduction SAP HANA enables organizations to optimize their business

More information

- Benchmark White Paper - Java CICS TS V2.2 Application

- Benchmark White Paper - Java CICS TS V2.2 Application - - Java CICS TS V2.2 Application Page 1/10 Table of Contents 1. Requirements 3 1.1 Application architecture 3 1.2 Project objectives 3 1.3 Technical configuration 4 1.3.1 Hardware and logical configuration

More information

IBM TotalStorage Enterprise Storage Server Model 800

IBM TotalStorage Enterprise Storage Server Model 800 A high-performance resilient disk storage solution for systems across the enterprise IBM TotalStorage Enterprise Storage Server Model 800 e-business on demand The move to e-business on demand presents

More information

Best Practices. Deploying Optim Performance Manager in large scale environments. IBM Optim Performance Manager Extended Edition V4.1.0.

Best Practices. Deploying Optim Performance Manager in large scale environments. IBM Optim Performance Manager Extended Edition V4.1.0. IBM Optim Performance Manager Extended Edition V4.1.0.1 Best Practices Deploying Optim Performance Manager in large scale environments Ute Baumbach (bmb@de.ibm.com) Optim Performance Manager Development

More information

Planning Considerations for HiperDispatch Mode Version 2 IBM. Steve Grabarits Gary King Bernie Pierce. Version Date: May 11, 2011

Planning Considerations for HiperDispatch Mode Version 2 IBM. Steve Grabarits Gary King Bernie Pierce. Version Date: May 11, 2011 Planning Considerations for HiperDispatch Mode Version 2 IBM Steve Grabarits Gary King Bernie Pierce Version Date: May 11, 2011 This document can be found on the web, www.ibm.com/support/techdocs Under

More information

DB2 Stored Procedures Monitoring, Analysis, and Tuning on System z

DB2 Stored Procedures Monitoring, Analysis, and Tuning on System z DB2 Stored Procedures Monitoring, Analysis, and Tuning on System z Charles Lewis, DB2 Advisor IBM System z Software Technical Professional September 11, 2013 Agenda What are stored procedures? Benefits

More information

QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER

QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER Hardware Sizing Using Amazon EC2 A QlikView Scalability Center Technical White Paper June 2013 qlikview.com Table of Contents Executive Summary 3 A Challenge

More information

... Performance benefits of POWER6 processors and IBM i 6.1 for Oracle s JD Edwards EnterpriseOne A performance case study for the Donaldson Company

... Performance benefits of POWER6 processors and IBM i 6.1 for Oracle s JD Edwards EnterpriseOne A performance case study for the Donaldson Company Performance benefits of POWER6 processors and IBM i 6.1 for Oracle s JD Edwards EnterpriseOne A performance case study for the Donaldson Company........ Jim Denton i ERP Development Jos Vermaere Executive

More information

IBM i Version 7.3. Systems management Disk management IBM

IBM i Version 7.3. Systems management Disk management IBM IBM i Version 7.3 Systems management Disk management IBM IBM i Version 7.3 Systems management Disk management IBM Note Before using this information and the product it supports, read the information in

More information

IBM TotalStorage Enterprise Tape Controller 3590 Model A60 enhancements support attachment of the new 3592 Model J1A Tape Drive

IBM TotalStorage Enterprise Tape Controller 3590 Model A60 enhancements support attachment of the new 3592 Model J1A Tape Drive Hardware Announcement IBM TotalStorage Enterprise Tape Controller 3590 Model A60 enhancements support attachment of the new 3592 Model J1A Tape Drive Overview New levels of performance and cartridge capacity

More information

Performance Tuning BI on SAP NetWeaver Using DB2 for i5/os and i5 Navigator

Performance Tuning BI on SAP NetWeaver Using DB2 for i5/os and i5 Navigator Performance Tuning BI on SAP NetWeaver Using DB2 for i5/os and i5 Navigator Susan Bestgen System i ERP SAP team System i TM with DB2 for i5/os TM leads the industry in SAP BI-D and SAP BW benchmark certifications

More information

Advanced Technical Skills (ATS) North America. John Burg Brad Snyder Materials created by John Fitch and Jim Shaw IBM Washington Systems Center

Advanced Technical Skills (ATS) North America. John Burg Brad Snyder Materials created by John Fitch and Jim Shaw IBM Washington Systems Center Advanced Technical Skills (ATS) North America zpcr Capacity Sizing Lab SHARE Sessions 2110/2111 March 17, 2010 John Burg Brad Snyder Materials created by John Fitch and Jim Shaw IBM Washington Systems

More information

PSR Testing of the EnterpriseOne Adapter for JD Edwards EnterpriseOne 8.12, OBIEE , DAC 7.9.6, and Informatica 8.6

PSR Testing of the EnterpriseOne Adapter for JD Edwards EnterpriseOne 8.12, OBIEE , DAC 7.9.6, and Informatica 8.6 PSR Testing of the EnterpriseOne Adapter for JD Edwards EnterpriseOne 8.12, OBIEE 1.1.3.4, DAC 7.9.6, and Informatica 8.6 Christian Smith Oracle Corporation January 29 Abstract This white paper documents

More information

IBM CICS Transaction Server V4.2

IBM CICS Transaction Server V4.2 IBM CICS Transaction Server V4.2 A Comparison of CICS QR and OTE Performance March 2012 IBM Hardware Acceleration Lab Nicholas C. Matsakis Wei K. Liu Greg Dyck Terry Borden Copyright IBM Corporation 2012

More information

Using EMC FAST with SAP on EMC Unified Storage

Using EMC FAST with SAP on EMC Unified Storage Using EMC FAST with SAP on EMC Unified Storage Applied Technology Abstract This white paper examines the performance considerations of placing SAP applications on FAST-enabled EMC unified storage. It also

More information

Framework for Doing Capacity Sizing on System z Processors

Framework for Doing Capacity Sizing on System z Processors Advanced Technical Skills (ATS) North America Framework for Doing Capacity Sizing on System z Processors Seattle Share: Session 2115 Bradley Snyder Email Address: bradley.snyder@us.ibm.com Phone: 972-561-6998

More information

IBM Lotus Domino 7 Performance Improvements

IBM Lotus Domino 7 Performance Improvements IBM Lotus Domino 7 Performance Improvements Razeyah Stephen, IBM Lotus Domino Performance Team Rob Ingram, IBM Lotus Domino Product Manager September 2005 Table of Contents Executive Summary...3 Impacts

More information

IBM TotalStorage 3592 Tape Drive Model J1A

IBM TotalStorage 3592 Tape Drive Model J1A Supports Business Continuity and Information Lifecycle Management in enterprise environments IBM TotalStorage 3592 Tape Drive Model J1A Highlights Overview The IBM TotalStorage 3592 Tape Drive Model J1A

More information

WebSphere Application Server 6.1 Base Performance September WebSphere Application Server 6.1 Base Performance

WebSphere Application Server 6.1 Base Performance September WebSphere Application Server 6.1 Base Performance WebSphere Application Server 6.1 Base Performance September 2008 WebSphere Application Server 6.1 Base Performance Table of Contents Introduction to the WebSphere Application Server performance tests...

More information

IBM TotalStorage Enterprise Storage Server Enhances Performance 15,000 rpm Disk Drives and Delivers New Solutions for Long Distance Copy

IBM TotalStorage Enterprise Storage Server Enhances Performance 15,000 rpm Disk Drives and Delivers New Solutions for Long Distance Copy Hardware Announcement April 23, 2002 IBM TotalStorage Enterprise Storage Server Enhances Performance 15,000 rpm Disk and Delivers New Solutions for Long Distance Copy Overview IBM continues to demonstrate

More information

SAP BPC 5.0 & 5.1 Performance and Reliability tips

SAP BPC 5.0 & 5.1 Performance and Reliability tips SAP BPC 5.0 & 5.1 Performance and Reliability tips Applies to: EPM-BPC 5.0 and 5.1 For more information, visit the Enterprise Performance Management homepage. Summary This is short list of hints to solve

More information

A. Specify NUMTCB=10 and allow 1 WLM managed stored procedure address space per sysplex for AE1.

A. Specify NUMTCB=10 and allow 1 WLM managed stored procedure address space per sysplex for AE1. Volume A~B: 103 Questions Volume A Question No : 1 An external stored procedure, assigned to application environment AE1, should run in parallel to a maximum of 10 concurrent procedures. Which action will

More information

z/vm 6.2 Live Guest Relocation with Linux Middleware

z/vm 6.2 Live Guest Relocation with Linux Middleware 2013-12-09 Linux on System z - z/vm 6.2 Live Guest Relocation with Linux Middleware Dr. Juergen Doelle Michael Johanssen IBM Research and Development Boeblingen Germany Trademarks IBM, the IBM logo, and

More information

What's New in the DBA Cockpit with SAP NetWeaver 7.0

What's New in the DBA Cockpit with SAP NetWeaver 7.0 What's New in the DBA Cockpit with SAP NetWeaver 7.0 Applies to: Database monitoring and administration of SAP systems running on DB2 for Linux, UNIX, and Windows using the latest DBA Cockpit that has

More information

z Processor Consumption Analysis Part 1, or What Is Consuming All The CPU?

z Processor Consumption Analysis Part 1, or What Is Consuming All The CPU? z Processor Consumption Analysis Part 1, or What Is Consuming All The CPU? Peter Enrico Email: Peter.Enrico@EPStrategies.com z/os Performance Education, Software, and Managed Service Providers Enterprise

More information

An Energy, Memory, and Performance Analysis Case Study on the IBM System z10 BC

An Energy, Memory, and Performance Analysis Case Study on the IBM System z10 BC An Energy, Memory, and Performance Analysis Case Study on the IBM System z10 BC June 5, 2009 Elisabeth Stahl Michael Buechele John P. Rankin IBM Systems and Technology Group Executive Overview As part

More information

OuterBay. IBM TotalStorage DR550 Performance Measurements. Updated as of February 14, IBM Storage Products

OuterBay. IBM TotalStorage DR550 Performance Measurements. Updated as of February 14, IBM Storage Products IBM TotalStorage DR55 Performance Measurements 1 OuterBay IBM TotalStorage DR55 Performance Measurements Covering: Results of benchmark testing on the IBM TotalStorage DR55 Updated as of February 14, 25

More information

VM & VSE Tech Conference May Orlando Session G41 Bill Bitner VM Performance IBM Endicott RETURN TO INDEX

VM & VSE Tech Conference May Orlando Session G41 Bill Bitner VM Performance IBM Endicott RETURN TO INDEX VM & VSE Tech Conference May 2000 - Orlando Session G41 Bill Bitner VM Performance IBM Endicott 607-752-6022 bitnerb@us.ibm.com RETURN TO INDEX Legal Stuff Disclaimer The information contained in this

More information

HiperDispatch Logical Processors and Weight Management

HiperDispatch Logical Processors and Weight Management HiperDispatch Logical Processors and Weight Management Fabio Massimo Ottaviani EPV Technologies August 2008 1 Introduction In the last few years, the power and number of the physical processors available

More information

What Developers must know about DB2 for z/os indexes

What Developers must know about DB2 for z/os indexes CRISTIAN MOLARO CRISTIAN@MOLARO.BE What Developers must know about DB2 for z/os indexes Mardi 22 novembre 2016 Tour Europlaza, Paris-La Défense What Developers must know about DB2 for z/os indexes Introduction

More information

IBM TotalStorage Enterprise Storage Server Model 800

IBM TotalStorage Enterprise Storage Server Model 800 A high-performance disk storage solution for systems across the enterprise IBM TotalStorage Enterprise Storage Server Model 800 e-business on demand The move to e-business on demand presents companies

More information

Qlik Sense Performance Benchmark

Qlik Sense Performance Benchmark Technical Brief Qlik Sense Performance Benchmark This technical brief outlines performance benchmarks for Qlik Sense and is based on a testing methodology called the Qlik Capacity Benchmark. This series

More information

IBM Technical Brief. IBM zenterprise System and System Storage Server DS8870: SAP Bank Analyzer 8.0 AFI Current Accounts Tests.

IBM Technical Brief. IBM zenterprise System and System Storage Server DS8870: SAP Bank Analyzer 8.0 AFI Current Accounts Tests. IBM Technical Brief IBM zenterprise System and System Storage Server DS8870: SAP Bank Analyzer 8.0 AFI Current Accounts Tests Authors: Brenda W. Beane Seewah Chan Paul Lekkas Veng K. Ly Michael R. Sheets

More information

Microsoft SQL Server 2012 Fast Track Reference Configuration Using PowerEdge R720 and EqualLogic PS6110XV Arrays

Microsoft SQL Server 2012 Fast Track Reference Configuration Using PowerEdge R720 and EqualLogic PS6110XV Arrays Microsoft SQL Server 2012 Fast Track Reference Configuration Using PowerEdge R720 and EqualLogic PS6110XV Arrays This whitepaper describes Dell Microsoft SQL Server Fast Track reference architecture configurations

More information

IBM. Systems management Disk management. IBM i 7.1

IBM. Systems management Disk management. IBM i 7.1 IBM IBM i Systems management Disk management 7.1 IBM IBM i Systems management Disk management 7.1 Note Before using this information and the product it supports, read the information in Notices, on page

More information

InfoSphere Warehouse with Power Systems and EMC CLARiiON Storage: Reference Architecture Summary

InfoSphere Warehouse with Power Systems and EMC CLARiiON Storage: Reference Architecture Summary InfoSphere Warehouse with Power Systems and EMC CLARiiON Storage: Reference Architecture Summary v1.0 January 8, 2010 Introduction This guide describes the highlights of a data warehouse reference architecture

More information

IBM TotalStorage Enterprise Storage Server (ESS) Model 750

IBM TotalStorage Enterprise Storage Server (ESS) Model 750 A resilient enterprise disk storage system at midrange prices IBM TotalStorage Enterprise Storage Server (ESS) Model 750 Conducting business in the on demand era demands fast, reliable access to information

More information

Dell PowerEdge R910 SQL OLTP Virtualization Study Measuring Performance and Power Improvements of New Intel Xeon E7 Processors and Low-Voltage Memory

Dell PowerEdge R910 SQL OLTP Virtualization Study Measuring Performance and Power Improvements of New Intel Xeon E7 Processors and Low-Voltage Memory Dell PowerEdge R910 SQL OLTP Virtualization Study Measuring Performance and Power Improvements of New Intel Xeon E7 Processors and Low-Voltage Memory A Dell Technical White Paper Dell, Inc. Waseem Raja

More information

IBM System Storage TS1130 Tape Drive Models E06 and other features enhance performance and capacity

IBM System Storage TS1130 Tape Drive Models E06 and other features enhance performance and capacity IBM Europe Announcement ZG08-0543, dated July 15, 2008 IBM System Storage TS1130 Tape Drive Models E06 and other features enhance performance and capacity Key prerequisites...2 Description...2 Product

More information

System Performance: Sizing and Tuning

System Performance: Sizing and Tuning www.novell.com/documentation System Performance: Sizing and Tuning ZENworks Mobile Management 2.6.x November 2012 Legal Notices Novell, Inc., makes no representations or warranties with respect to the

More information

IBM Data Warehousing Balanced Configuration Unit for AIX, V1.1 accelerates development of data warehouse and business intelligence infrastructures

IBM Data Warehousing Balanced Configuration Unit for AIX, V1.1 accelerates development of data warehouse and business intelligence infrastructures Software Announcement IBM Data Warehousing Balanced Configuration Unit for AIX, V1.1 accelerates development of data warehouse and business intelligence infrastructures Overview The Data Warehousing Balanced

More information

IBM System Storage - DS8870 Disk Storage Microcode Bundle Release Note Information v1

IBM System Storage - DS8870 Disk Storage Microcode Bundle Release Note Information v1 Release Date: August 15, 2015 VRMF Level Data Results: VRMF level From: 87.50.5.0 VRMF Level To: 87.51.10.0 Report for: Code Bundle Contents All DS8870 This table includes code component reference information.

More information

Infor M3 on IBM POWER7+ and using Solid State Drives

Infor M3 on IBM POWER7+ and using Solid State Drives Infor M3 on IBM POWER7+ and using Solid State Drives IBM Systems & Technology Group Robert Driesch cooter@us.ibm.com This document can be found on the web, Version Date: January 31, 2014 Table of Contents

More information

Oracle s JD Edwards EnterpriseOne IBM POWER7 performance characterization

Oracle s JD Edwards EnterpriseOne IBM POWER7 performance characterization Oracle s JD Edwards EnterpriseOne IBM POWER7 performance characterization Diane Webster IBM Oracle International Competency Center January 2012 Copyright IBM Corporation, 2012. All Rights Reserved. All

More information

QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER

QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER Measuring Business Intelligence Throughput on a Single Server QlikView Scalability Center Technical White Paper December 2012 qlikview.com QLIKVIEW THROUGHPUT

More information

An Oracle White Paper. Released April 2013

An Oracle White Paper. Released April 2013 Performance and Scalability Benchmark: Siebel CRM Release 8.1.1.4 Industry Applications and Oracle 11.2.0.3 Database on Oracle's SPARC T5 Servers and Oracle Solaris An Oracle White Paper Released April

More information

Session: B10 Should Your Next Warehouse be on DB2 for z/os? Willie Favero IBM Silicon Valley Lab. 15 October :30 14:30 Platform: z/os

Session: B10 Should Your Next Warehouse be on DB2 for z/os? Willie Favero IBM Silicon Valley Lab. 15 October :30 14:30 Platform: z/os Session: B10 Should Your Next Warehouse be on DB2 for z/os? Willie Favero IBM Silicon Valley Lab 15 October 2008 13:30 14:30 Platform: z/os Disclaimer The information contained in this presentation has

More information

IBM System Storage TS7740 Virtualization Engine now supports three cluster grids, Copy Export for standalone clusters, and other upgrades

IBM System Storage TS7740 Virtualization Engine now supports three cluster grids, Copy Export for standalone clusters, and other upgrades IBM United States Announcement 107-392, dated July 10, 2007 IBM System Storage TS7740 Virtualization Engine now supports three cluster grids, Copy Export for standalone clusters, and other upgrades Key

More information

System z Performance : More Than You Think

System z Performance : More Than You Think System z Performance : More Than You Think SHARE Pittsburgh Session 15709 EWCP Steve Grabarits IBM August 6, 2014 1 Trademarks The following are trademarks of the International Business Machines Corporation

More information

IBM TotalStorage Enterprise Tape Library 3494

IBM TotalStorage Enterprise Tape Library 3494 Modular tape automation for heterogeneous computing environments IBM TotalStorage Enterprise Tape Library 3494 configured with drive frames, storage frames and up to two IBM TotalStorage Virtual Tape Servers.

More information

Using Xcelsius 2008 with SAP NetWeaver BW

Using Xcelsius 2008 with SAP NetWeaver BW Using Xcelsius 2008 with SAP NetWeaver BW Applies to: Xcelsius 2008 Enterprise Service Pack 02 (and higher) SAP NetWeaver BW 7.0 Enhancement package 01 Service Pack 05 (and higher) Summary In this short

More information

IBM TotalStorage DS8800 Disk Storage Microcode Bundle v2 Release Note Information

IBM TotalStorage DS8800 Disk Storage Microcode Bundle v2 Release Note Information Release Date: November 18, 2011 VRMF Level Data Results: VRMF level From: N/A VRMF Level To: 86.20.98.0 Report for: Code Bundle Contents DS800 Code Bundle Level SEA or LMC Version: Used with dscli ver

More information

Understanding The Interaction Of z/os Workload Manager And DB2

Understanding The Interaction Of z/os Workload Manager And DB2 IBM Software Group Understanding The Interaction Of z/os Workload Manager And DB2 Ed Woods / IBM Corporation 2010 IBM Corporation Agenda Workload Manager Overview Important WLM Concepts And Terminology

More information

The Hidden Gold in the SMF 99s

The Hidden Gold in the SMF 99s The Hidden Gold in the SMF 99s Peter Enrico Email: Peter.Enrico@EPStrategies.com z/os Performance Education, Software, and Managed Service Providers Enterprise Performance Strategies, Inc. 3457-53rd Avenue

More information

Introduction to HiperDispatch Management Mode with z10 1

Introduction to HiperDispatch Management Mode with z10 1 Introduction to HiperDispatch Management Mode with z10 1 Donald R. Deese Computer Management Sciences, Inc. Hartfield, Virginia 23071-3113 HiperDispatch was introduced with IBM s z10 server, and is available

More information

Infor Lawson on IBM i 7.1 and IBM POWER7+

Infor Lawson on IBM i 7.1 and IBM POWER7+ Infor Lawson on IBM i 7.1 and IBM POWER7+ IBM Systems & Technology Group Mike Breitbach mbreit@us.ibm.com This document can be found on the web, Version Date: March, 2014 Table of Contents 1. Introduction...

More information

Austin HOTDUG. DB2 for z/os Selected Performance Topics

Austin HOTDUG. DB2 for z/os Selected Performance Topics Austin HOTDUG DB2 for z/os Selected Performance Topics Roger Miller IBM July 14, 2009 10:00 a.m. 11:00 a.m. Platform: DB2 for z/os This talk will present some of the latest challenges and achievements

More information

DB2 for z/os and OS/390 Performance Update - Part 1

DB2 for z/os and OS/390 Performance Update - Part 1 DB2 for z/os and OS/390 Performance Update - Part 1 Akira Shibamiya Orlando, Florida October 1-5, 2001 M15a IBM Corporation 1 2001 NOTES Abstract: The highlight of major performance enhancements in V7

More information

The Deployment of SAS Enterprise Business Intelligence Solution in a large IBM POWER5 Environment

The Deployment of SAS Enterprise Business Intelligence Solution in a large IBM POWER5 Environment February 2006 The Deployment of SAS Enterprise Business Intelligence Solution in a large IBM POWER5 Edward Hayes-Hall Brian Porter Harry Seifert Hsian-Fen Tsao Page 2 Introduction A user of a SAS solution

More information

Sub-capacity licensing for select IBM Passport Advantage eligible programs running on x86 servers helps improve flexibility and price/performance

Sub-capacity licensing for select IBM Passport Advantage eligible programs running on x86 servers helps improve flexibility and price/performance Software Announcement April 25, 2006 Sub-capacity licensing for select IBM Passport Advantage eligible programs running on x86 servers helps improve flexibility and price/performance Overview IBM continues

More information

Implementing SQL Server 2016 with Microsoft Storage Spaces Direct on Dell EMC PowerEdge R730xd

Implementing SQL Server 2016 with Microsoft Storage Spaces Direct on Dell EMC PowerEdge R730xd Implementing SQL Server 2016 with Microsoft Storage Spaces Direct on Dell EMC PowerEdge R730xd Performance Study Dell EMC Engineering October 2017 A Dell EMC Performance Study Revisions Date October 2017

More information

Oracle PeopleSoft Applications for IBM z Systems

Oracle PeopleSoft Applications for IBM z Systems Oracle PeopleSoft Applications for IBM z Systems Michael Curtis IBM Systems, ISV Enablement z Systems Technical Specialist mcurtis@us.ibm.com Susan Adamovich IBM Systems, ISV Enablement Oracle on z Systems

More information

DB2 9 for z/os V9 migration status update

DB2 9 for z/os V9 migration status update IBM Software Group DB2 9 for z/os V9 migration status update July, 2008 Bart Steegmans DB2 for z/os L2 Performance Acknowledgement and Disclaimer i Measurement data included in this presentation are obtained

More information

z990 Performance and Capacity Planning Issues

z990 Performance and Capacity Planning Issues z990 Performance and Capacity Planning Issues Cheryl Watson Session 2537; SHARE 104 in Anaheim March 2, 2005 Watson & Walker, Inc. home of Cheryl Watson's TUNING Letter, CPU Chart, BoxScore & GoalTender

More information

z/vm Data Collection for zpcr and zcp3000 Collecting the Right Input Data for a zcp3000 Capacity Planning Model

z/vm Data Collection for zpcr and zcp3000 Collecting the Right Input Data for a zcp3000 Capacity Planning Model IBM z Systems Masters Series z/vm Data Collection for zpcr and zcp3000 Collecting the Right Input Data for a zcp3000 Capacity Planning Model Session ID: cp3kvmxt 1 Trademarks The following are trademarks

More information

IMPROVING THE PERFORMANCE, INTEGRITY, AND MANAGEABILITY OF PHYSICAL STORAGE IN DB2 DATABASES

IMPROVING THE PERFORMANCE, INTEGRITY, AND MANAGEABILITY OF PHYSICAL STORAGE IN DB2 DATABASES IMPROVING THE PERFORMANCE, INTEGRITY, AND MANAGEABILITY OF PHYSICAL STORAGE IN DB2 DATABASES Ram Narayanan August 22, 2003 VERITAS ARCHITECT NETWORK TABLE OF CONTENTS The Database Administrator s Challenge

More information

Linux and z Systems in the Datacenter Berthold Gunreben

Linux and z Systems in the Datacenter Berthold Gunreben Linux and z Systems in the Datacenter Berthold Gunreben Build Service Engineer SUSE Linux Available Architectures for SUSE Linux Enterprise Server X86, z Systems, Power Common Source Code for all architectures

More information

Run SAP BPC in a VMware environment Version 1.00 December 2008

Run SAP BPC in a VMware environment Version 1.00 December 2008 How-to Guide SAP CPM How To Run SAP BPC in a VMware environment Version 1.00 December 2008 Applicable Releases: SAP BPC 5.x, SAP BPC 7.0 (MS), SAP BPC 7.0 (NW) (.Net tier only) Copyright 2007 SAP AG. All

More information

TABLE DISTRIBUTION IN HANA HANA. SAP Active Global Support, June 2012

TABLE DISTRIBUTION IN HANA HANA. SAP Active Global Support, June 2012 TABLE DISTRIBUTION IN HANA HANA SAP Active Global Support, June 2012 Table Distribution : Why Load Balancing Parallelization Table Partitioning - A non-partitioned table can support only 2 billion rows.

More information

Key Metrics for DB2 for z/os Subsystem and Application Performance Monitoring (Part 1)

Key Metrics for DB2 for z/os Subsystem and Application Performance Monitoring (Part 1) Key Metrics for DB2 for z/os Subsystem and Application Performance Monitoring (Part 1) Robert Catterall IBM March 12, 2014 Session 14610 Insert Custom Session QR if Desired. The genesis of this presentation

More information

Framework for Doing Capacity Sizing for System z Processors

Framework for Doing Capacity Sizing for System z Processors IBM Advanced Technical Support - WSC Framework for Doing Capacity Sizing for System z Processors Summer 2009 Share session: 2115 Bradley Snyder Email Address: bradley.snyder@us.ibm.com Phone: 972-561-6998

More information