SAP on IBM i Performance Update · PDF fileSAP on IBM i Performance Update ... Oracle SPARC Server M5-32 on the two-tier SAP SD standard application benchmark running SAP ... IBM i

SAP on IBM i Performance Update

Ron Schmerbauch

SAP on IBM i Development

Rochester, Minnesota, USA

[email protected]

Hardware performance for SAP on IBM i

CPU sizing/performance of Power 8

Single threaded performance comparison

Memory highlights

Storage considerations

Software performance for SAP on IBM i

IBM i 7.2

SAP BW performance 7.2 vs 7.1

HANA considerations

Summary

Agenda

© 2014 International Business Machines Corporation 2

313

883

996

443

365

458424

0

200

400

600

800

1000

1200

SD

Benchmark

Users per

Core

2.17X more users

IBM

E870

POWER8 8p/80c/640t

Dell

PowerEdge R920

E7-4890 v2 4p/60c/120t

Oracle

T5-8

T5 8p/128c/1024t

IBM

S824

POWER8

4p/24c/192t

2.17x ERP users per core with Power E870 versus the competition on

SAP Sales and Distribution 2-Tier Benchmark Nearly 1000 Users per Core with POWER8 based E870

Dell

PowerEdge R730

E5-2699 v3 2p/36c/72t

3.18X more users

(1) IBM Power Enterprise System E870 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 8 processors / 80 cores / 640 threads, POWER8; 4.19GHz, 2048 GB memory, 79,750 SD benchmark users, running AIX® 7.1 and DB2® 10.5 Certification #: 2014034 Source: http://www.sap.com/benchmark. (2) IBM Power System S824 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 4 processors / 24 cores / 96 threads, POWER8; 3.52GHz, 512 GB memory, 21,212 SD benchmark users, running AIX® 7.1 and DB2® 10.5, Certification # 2014016. Source: http://www.sap.com/benchmark All results valid as of October 3, 2014 (3) Dell PowerEdge R730, on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 2 processors/36 cores/72 threads, Intel Xeon Processor 2699v3; 2.30 GHz, 256 GB memory; 16,500 SD benchmark users, running RHEL 7 and SAP ASE 16; Certification # 2014033. Source: http://www.sap.com/benchmark. (4) Oracle SPARC Server M5-32 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 32 processors/192 cores/1536 threads, SPARC M5; 3.60 GHz, 4,096 GB memory; 85,050 SD benchmark users, running Solaris® 11 and Oracle 11g; Certification # 20013009. Source: http://www.sap.com/benchmark. (5) Dell PowerEdge R920 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 4 processors/60 cores/120 threads, Intel Xeon Processor 4890 v2; 2.80 GHz, 1024 GB memory; 25,451 SD benchmark users, running SUSE Linux Enterprise Server 11 and SAP ASE 16; Certification # 2014011. Source: http://www.sap.com/benchmark. (6) Oracle SPARC Server M6-32 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 32 processors/384 cores/3072 threads, SPARC M6; 3.60 GHz, 16 TB memory; 140,000 SD benchmark users, running Solaris® 11 and Oracle 11g; Certification # 20014008. Source: http://www.sap.com/benchmark. (7) Oracle SPARC Server T5-8 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 8 processors/128 cores/1024 threads, SPARC T5; 3.60 GHz, 2,048 GB memory; 40,000 SD benchmark users, running Solaris® 11 and Oracle 11g; Certification # 2013008. Source: http://www.sap.com/benchmark. SAP and all SAP logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries. All other product and service names mentioned are the trademarks of their respective companies.

SAP SD Standard Application Benchmark Results, 2-Tier: SD Benchmark Users per Core SAP enhancement package 5 for SAP ERP 6.0

Source: http//www.sap.com//benchmark

Oracle

M5-32

M5 32p/192c/1536t

Oracle

M6-32

M6 32p/384c/3072t


http://www.sap.com/benchmark







Power 8 capacity scaling - validation testing vs. Power 7

Hardware comparison CPW

rating

per core

SAPS

rating,

per core

NetWeaver

Workload -

Dialog

steps/hr

per core

Workload

Users

Workload

CPU % and

Response

Time

P7 795

(1 socket, 6 core lpar)

6010 2159 23560 400 97%

.665 sec

P8 S824

(1 socket, 8 core lpar)

10844 4383 42144 950 97%

.650 sec

P8 improvement 1.80x 2.03x 1.79x 2.375x

Power 7 LPAR – 795 6core LPAR (of 48) 9119-FHB 4702, 3.7GHz 288500 cpw 103650 saps 141363 ds/hr

Power 8 LPAR - S824 8core LPAR (of 16) 8286-42A EPXF, 4.14GHz 173500 cpw 70120 saps 337153 ds/hr

Same workloads, both using IBM i 7.2 - on separate HW boxes.

Goal: Maximize throughput of each workload, compare "per core" results.

CPU intensive workload, single thread comparison P7+ 740 4.2GHz; P8 S824 4.1GHz

1 core dedicated partition

Each P8 box is smaller (faster) than it's P7 equivalent.

Each unit in P8 SMT8 mode is completed faster than P7 SMT4 mode.


Memory Buffer

DRAM Chips

DDR Interfaces

POWER

8

Link

Scheduler &

Management

16MB

Memory

Cache

POWER8 Memory Buffer Chip

“L4 cache”

Intelligence Moved into Memory • Scheduling logic, caching structures

• Energy Mgmt, RAS decision point

–Moved to Memory Buffer from CPU

Processor Interface • More robust RAS

• “On-the-fly” lane isolation/repair

Performance Value • End-to-end fast path and data retry (latency)

• Cache latency/bandwidth, partial updates

• Cache write scheduling, prefetching, energy


Is Disk I/O a concern in your SAP system?

HANA

Performance SSD(3rd gen) / FlashStorage "in memory"

Read or write requests? CPU bound or I/O bound?

GUI

time

Wait

time

Roll

time

Load and

gen. time

Enqueue

time

Processing

time

Database

time

SAP Response time


Customer Example with FlashStorage

Partition ID : 001 Feature Code . : 4700-4700

Average Average

Itv Average Reads Writes

End I/O /Sec /Sec /Sec

----- --------- -------- --------

23:40 7,947.3 3,080.6 4,866.7

23:45 7,576.0 4,061.2 3,514.7

23:50 5,496.1 1,982.2 3,513.8

23:55 5,637.2 2,153.2 3,483.9

--------- -------- --------

Average: 8.674.6 3,439.6 5,235,0

Performance Tools – Resource Interval Report:

Avg. Response

times per I/O:

HDD: 3.5 ms

SSD/Flash: 0.5 ms

Delta 3.0 ms

Total number of reads per day from 5770-PT1: 297,181,440

Total number of transactions from ST03 (all SIDs): 9,833,504

Number of reads per transaction: 30*)

Expected average benefit: 90 ms per transaction*)

*) Reads from non-SAP work like backup or RTVDSKINF not considered – reduces benefit even more


Tools for Disk I/O Assessment

SSD Analyzer Tool for IBM i

General recommendation based on average disk

read time collected by IBM i Performance Monitor

ABAP SQL Monitor – SAP Note 1885926

Identification of expensive database access through

ABAP applications, available for SAP NetWeaver ≥ 7.0

I/O share in SQL statements – SAP Note 1869494

List of SQL statements with long execution times with

their I/O share (report ZDB4GETIOTIMESHARE)


http://www.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS3780

https://service.sap.com/sap/support/notes/1885926











Memory highlights



IBM i 7.2


HANA considerations

Summary

Agenda


DB2 for i - I/O Cost Model Enhancement

• Motivation

– Query optimizer needs to know how much time it’s going to take to access an object (scan a table, probe an index) to create a plan

– Original I/O Cost model assumption: 25 msec access time per I/O for all HW

– I/O hardware landscape is changing rapidly (e.g. faster and smaller HDDs, external storage, SSDs, Flash)

– I/O implementation and execution algorithms diverged from model over time

– Big Data paradigm means more dependency on I/O cost model correctness

• Costing Change @ 7.2

– New method to sample actual access times rather than hard-code a fixed time; Every system tracks unique I/O performance metrics over time

– Sampling enables optimizer to distinguish unique performance characteristics of current and future internal, external, and solid state storage devices

• Result

– DB2 has more accurate I/O detail when costing plans

– If data is moved to faster I/O HW, DB2 access plans may change

IBM i 7.2 validation – BW complex query workload vs. i 7.1 Configuration

– 2 LPARs with identical HW configuration (CPU/Mem/Disk), both on same P7 server

– Same number of SAP work process jobs to service users

– One LPAR IBM i 7.1 (latest TR), the other with IBM i 7.2 GA version

Test

– A script of 18 complex BW queries, each somehow unique from the others.

– All necessary data cannot fit into memory, I/O is required

– Users log on and repeat the script 5 times, each user starts 1 sec after the previous

– Measurements taken only once all users are logged on and looping through queries

– Script randomness occurs by the time all users are logged on

Compare

– Overall CPU utilization, throughput and response times

– Individual query average response times

LPAR1:

BW

DEV QAS

7.1

LPAR2:

BW

7.2

For this workload, IBM i 7.2

- reduced CPU consumption by ~20%

- improved Response time by almost half

- OLTP workloads may not have as much room for improvement

0

10

20

30

40

50

60

70

80

90

100

N Nx2 Nx3 Nx4 Nx5

CP

U a

nd

RT

User load

7.1 CPU

7.2 CPU

7.1 RT

7.2 RT

IBM i 7.2 validation – BW complex query workload vs. i 7.1

IBM i 7.2 – BW complex query workload comparison

Best single query timing for each step, across all users and loops

- 7.2 consistently improves at least a little over 7.1


Slowest single query timing, across all users and loops

- Worst case scenario. 7.2 is overall much better than 7.1


Average response time for each query, all users, all loops

7.2 is consistently better than 7.1 on a busy system.




Memory highlights



IBM i 7.2


HANA considerations

Summary

Agenda


HANA considerations

Business Suite is OLTP

– Columnar DB primary strength is complex query performance

– Row based DB strength with OLTP applications

SAP BW benchmarks showed DB2 for i competitive with HANA

– Load and query aspects included in SAP's benchmark

– DB2 for i EVIs provide fast complex query performance on row-based tables

– Small number of CPUs are sufficient for most i BW systems

IBM i integration with HANA sidecar

– Pre-defined ERP reporting scenarios (CO-PA example)

– Original data provided by DB2 for i

– Least disruptive use of HANA - DB2 for i provides a safety net

– HANA in a Power LPAR (under construction)

Power 8 HW and Storage

More CPU capacity

Better single threaded performance, even with more threads

SSD and Flash are mainstream

IBM i 7.2

Functional and performance improvements in OS and DB

Notable improvements in complex DB workloads overall

Makes the "worst case query" easier on users

More performance coming soon from IBM and SAP

IBM EVI performance enhancement

SAP Core Data Services for all DB platforms

SAP extended support of Business Suite 7 to 2025

Power 8 and 7.2 provide the foundation for the coming years

Postition yourself for the future...


Questions? – Thank You!

Contact: [email protected]

www.linkedin.com/in/ronschmerbauch/

Visit us at http://www.ibm.com/developerworks/ibmi/sap


mailto:[email protected]

http://www.linkedin.com/in/ronschmerbauch/

http://www.ibm.com/developerworks/ibmi/sap

Legal Information – Please Note

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.

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.

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.

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.

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.

© 2014 International

Business Machines

Corporation

24

Legal Information – Trademarks, Copyrights Information contained in this material has not been submitted to any formal IBM review and is distributed on “as is” basis without any

warranty either expressed or implied. The use of this information is a customer responsibility.

IBM MAY HAVE PATENTS OR PENDING PATENT APPLICATIONS COVERING SUBJECT MATTER IN THIS DOCUMENT. THE

FURNISHING OF THIS DOCUMENT DOES NOT IMPLY GIVING LICENSE TO THESE PATENTS.

The client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific

configurations and operating conditions. It is the user’s responsibility to evaluate and verify the operation of any other products or

programs with IBM products and programs.

IBM, the IBM logo, ibm.com, AIX, BatchPipes, BladeCenter, DB2, DB2 Connect, Netezza, GDPS, GPFS, Netfinity, Intelligent Miner,

OpenPower, Parallel Sysplex, POWER, RACF, Redbooks, RETAIN, System Storage, System p, System x, System z, Tivoli,

WebSphere, zEnterprise, z9, z10, and z/OS are trademarks or registered trademarks of International Business Machines Corp.,

registered in many jurisdictions worldwide. A current list of IBM trademarks is available on the web at “Copyright and trademark

information” at: http://www.ibm.com/legal/us/en/copytrade.shtml

LinkedIn, the LinkedIn logo, the IN logo and InMail are registered trademarks or trademarks of LinkedIn Corporation and its affiliates in

the United States and/or other countries.

Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.

Adobe, the Adobe logo, Acrobat, PostScript, and Reader are either trademarks or registered trademarks of Adobe Systems

Incorporated in the United States and/or other countries.

Microsoft, Windows, Excel, Outlook, and PowerPoint are registered trademarks of Microsoft Corporation.

Oracle and Java are registered trademarks of Oracle and/or its affiliates.

UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.

SAP, R/3, SAP NetWeaver, ByDesign, SAP BusinessObjects Explorer, StreamWork, and other SAP products and services mentioned

herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries.

All other product and service names mentioned are the trademarks of their respective companies.

Data contained in this document serves informational purposes only. National product specifications may vary.

© 2014 International

Business Machines

Corporation

25

http://www.ibm.com/legal/us/en/copytrade.shtml

SAP on IBM i Performance Update · PDF fileSAP on IBM i Performance Update ... Oracle SPARC Server M5-32 on the two-tier SAP SD standard application benchmark running SAP ... IBM i

Documents