Tuesday Mar 26, 2013

SPARC M5-32 Produces SAP SD Two-Tier Benchmark World Record for SAP Enhancement Package 5 for SAP ERP 6.0

Oracle's SPARC M5-32 server produced a world record result on the two-tier SAP Sales and Distribution (SD) Standard Application Benchmark using SAP Enhancement package 5 for SAP ERP 6.0.

  • The SPARC M5-32 server achieved 85,050 users running the two-tier SAP Sales and Distribution (SD) Standard Application Benchmark using SAP Enhancement package 5 for SAP ERP 6.0.

  • The SPARC M5-32 solution was run with Oracle Solaris 11 and used the Oracle Database 11g.

Performance Landscape

SAP-SD 2-Tier Performance Table (in decreasing performance order). SAP ERP 6.0 Enhancement Pack 5 for SAP ERP 6.0 results (new version of the benchmark as of May, 2012).

System OS
Database
Users SAPS SAP
ERP/ECC
Release
Date
SPARC M5-32 Server
32x SPARC M5 @3.6 GHz, 4 TB
Solaris 11
Oracle 11g
85,050 472,600 EHP5 for SAP
ERP 6.0
25-Mar-13
IBM Power 780
12xPOWER7+ @3.72 GHz, 1536 GB
AIX 7.1
DB2 10
57,024 311,720 EHP5 for SAP
ERP 6.0
3-Oct-12
IBM Power 760
8xPOWER7+ @3.41 GHz, 1024 GB
AIX 7.1
DB2 10
25,488 139,220 EHP5 for SAP
ERP 6.0
5-Feb-13

SAP ERP 6.0 Enhancement Pack 4 for SAP ERP 6.0 Results
(Old version of the benchmark, obsolete at the end of April, 2012)

System OS
Database
Users SAPS SAP
ERP/ECC
Release
Date
IBM Power 795
32xPOWER7 @4 GHz, 4 TB
AIX 7.1
DB2 9.7
126,063 688,630 EHP4 for SAP
ERP 6.0
15-Nov-10
SPARC Enterprise Server M9000
64xSPARC64 VII @2.88 GHz, 1152 GB
Solaris 10
Oracle 10g
32,000 175,600 EHP4 for SAP
ERP 6.0
18-Nov-09

Complete benchmark results may be found at the SAP benchmark website http://www.sap.com/benchmark.

Configuration Summary and Results

Hardware Configuration:

1 x SPARC M5-32 server with
32 x 3.6 GHz SPARC M5 processors (total of 32 processors / 192 cores / 1536 threads)
4 TB memory
1 x Sun Storage 2540-M2 (12 x 300 GB 5K RPM 3.5" SAS-2 disk & 2 GB cache)
Flash Storage

Software Configuration:

Oracle Solaris 11
SAP enhancement package 5 for SAP ERP 6.0
Oracle Database 11g Release 2

Certified Results (published by SAP)

Performance: 85,050 benchmark users
SAP Certification: 2013009

Benchmark Description

The SAP Standard Application SD (Sales and Distribution) Benchmark is a two-tier ERP business test that is indicative of full business workloads of complete order processing and invoice processing, and demonstrates the ability to run both the application and database software on a single system. The SAP Standard Application SD Benchmark represents the critical tasks performed in real-world ERP business environments.

SAP is one of the premier world-wide ERP application providers, and maintains a suite of benchmark tests to demonstrate the performance of competitive systems on the various SAP products.

See Also

Disclosure Statement

Two-tier SAP Sales and Distribution (SD) standard application benchmarks, SAP Enhancement package 5 for SAP ERP 6.0 as of 3/26/13:

SPARC M5-32 (32 processors, 192 cores, 1536 threads) 85,050 SAP SD users, 32 x 3.6 GHz SPARC M5, 4 TB memory, Oracle Database 11g, Oracle Solaris 11, Cert# 2013009. IBM Power 780 (12 processors, 96 cores, 384 threads) 57,024 SAP SD users, 12 x 3.72 GHz IBM POWER7+, 1536 GB memory, DB210, AIX7.1, Cert#2012033. IBM Power 760 (8 processors, 48 cores, 192 threads) 25,488 SAP SD users, 8 x 3.41 GHz IBM POWER7+, 1024 GB memory, DB2 10, AIX 7.1, Cert#2013004.

Two-tier SAP Sales and Distribution (SD) standard application benchmarks, SAP Enhancement package 4 for SAP ERP 6.0 as of 3/26/13:

IBM Power 795 (32 processors, 256 cores, 1024 threads) 126,063 SAP SD users, 32 x 4 GHz IBM POWER7, 4 TB memory, DB2 9.7, AIX7.1, Cert#2010046. SPARC Enterprise Server M9000 (64 processors, 256 cores, 512 threads) 32,000 SAP SD users, 64 x 2.88 GHz SPARC64 VII, 1152 GB memory, Oracle Database 10g, Oracle Solaris 10, Cert# 2009046.

SAP, R/3, reg TM of SAP AG in Germany and other countries. More info www.sap.com/benchmark

SPARC T5-2 Achieves JD Edwards EnterpriseOne Benchmark World Records

Oracle produced World Record batch throughput for single system results on Oracle's JD Edwards EnterpriseOne Day-in-the-Life benchmark using Oracle's SPARC T5-2 server running Oracle Solaris Containers and consolidating JD Edwards EnterpriseOne, Oracle WebLogic servers and the Oracle Database 11g Release 2. There are two workloads tested: online plus batch workload and batch-only workload.

Online plus batch workload:

  • The SPARC T5-2 server delivered a result of 12,000 online users at 180 msec average response time while concurrently executing a mix of JD Edwards EnterpriseOne long and short batch processes at 198.5 UBEs/min (Universal Batch Engines per minute).

  • The SPARC T5-2 server online plus batch throughput is 2.7x higher than the IBM Power 770 server, both running 12,000 online users.

  • The SPARC T5-2 server online plus batch throughput is 6x higher per chip than the IBM Power 770 server. The SPARC T5-2 server has 2 chips and the IBM Power 770 has 4 chips, both ran 12,000 online users.

  • The SPARC T5-2 server online plus batch throughput is 3x higher per core than the IBM Power 770 server. Both servers have 32 cores and ran 12,000 online users.

Batch-only workload:

  • The SPARC T5-2 server delivered throughput of 880 UBEs/min while executing the batch-only workload (Long and Short batch processes).

  • The SPARC T5-2 server batch-only throughput is 2.7x faster per chip than the IBM Power 770 server. The SPARC T5-2 server has 2 chips and the IBM Power 770 has 4 chips.

  • The SPARC T5-2 server batch-only throughput is 1.4x higher per core than the IBM Power 770 server. Both servers have 32 cores.

  • The SPARC T5-2 server batch-only throughput is 61% faster than the Cisco multiple system solution.

  • The SPARC T5-2 server batch-only throughput is 5x faster per chip than the Cisco UCS B200/B250 M2 servers. The SPARC T5-2 server has 2 chips and the Cisco 3 server solution has 6 chips.

  • The SPARC T5-2 server batch-only throughput is 18x higher per core than the Cisco UCS B200/B250 M2 servers. The SPARC T5-2 server has 32 cores while the Cisco solution utilized 36 cores.

Both workloads:

  • The SPARC T5-2 server offers a 5.4x cost savings for the application server when compared to the IBM Power 770 application server.

  • The SPARC T5-2 server running Oracle Solaris Containers and consolidating JD Edwards EnterpriseOne, Oracle WebLogic servers and the Oracle Database 11g Release 2 utilized a maximum 65% of the available CPU power, leaving headroom for additional processing.

  • The database server in a shared-server configuration allows for optimized CPU resource utilization and significant memory savings on the SPARC T5-2 server without sacrificing performance.

Performance Landscape

JD Edwards EnterpriseOne Day in the Life (DIL) Benchmark
Consolidated Online with Batch Workload
System Rack
Units (U)
Batch
Rate
(UBEs/min)
Online
Users
Users/
U
UBEs/
Core
UBEs/
Chip
Version
SPARC T5-2 (2 x SPARC T5, 3.6 GHz) 3 198.5 12000 4000 6.2 99 9.0.2
IBM Power 770 (4 x POWER7, 3.3 GHz) 8 65 12000 1500 2.0 16 9.0.2

Batch Rate (UBEs/min) — Batch transaction rate in UBEs per minute.

JD Edwards EnterpriseOne Batch Only Benchmark
System Rack
Units (U)
Batch
Rate
(UBEs/min)
UBEs/
U
UBEs/
Core
UBEs/
Chip
Version
SPARC T5-2 (2 x SPARC T5, 3.6 GHz) 3 880 267 25 440 9.0.2
IBM Power 770 (4 x POWER7, 3.3 GHz) 8 643 81 20 161 9.0.2
2 x Cisco B200 M2 (2 x X5690, 3.46 GHz)
1 x Cisco B250 M2 (2 x X5680, 3.33 GHz)
3 546 182 15 91 9.0.2

Configuration Summary

Hardware Configuration:

1 x SPARC T5-2 server with
2 x SPARC T5 processors, 3.6 GHz
512 GB memory
4 x 300 GB 10K RPM SAS internal disk
2 x 300 GB internal SSD
4 x Sun Flash Accelerator F40 PCIe Card (4 x 93 GB)

Software Configuration:

Oracle Solaris 10 1/13
Oracle Solaris Containers
JD Edwards EnterpriseOne 9.0.2
JD Edwards EnterpriseOne Tools (8.98.4.2)
Oracle WebLogic Server 11g (10.3.4)
Oracle HTTP Server 11g
Oracle Database 11g Release 2 (11.2.0.3)

Benchmark Description

JD Edwards EnterpriseOne is an integrated applications suite of Enterprise Resource Planning (ERP) software. Oracle offers 70 JD Edwards EnterpriseOne application modules to support a diverse set of business operations.

Oracle's Day in the Life (DIL) kit is a suite of scripts that exercises most common transactions of JD Edwards EnterpriseOne applications, including business processes such as payroll, sales order, purchase order, work order, and manufacturing processes, such as ship confirmation. These are labeled by industry acronyms such as SCM, CRM, HCM, SRM and FMS. The kit's scripts execute transactions typical of a mid-sized manufacturing company.

  • The workload consists of online transactions and the UBE – Universal Business Engine workload of 61 short and 4 long UBEs.

  • LoadRunner runs the DIL workload, collects the user’s transactions response times and reports the key metric of Combined Weighted Average Transaction Response time.

  • The UBE processes workload runs from the JD Enterprise Application server.

    • Oracle's UBE processes come as three flavors:
      • Short UBEs < 1 minute engage in Business Report and Summary Analysis,
      • Mid UBEs > 1 minute create a large report of Account, Balance, and Full Address,
      • Long UBEs > 2 minutes simulate Payroll, Sales Order, night only jobs.
    • The UBE workload generates large numbers of PDF files reports and log files.
    • The UBE Queues are categorized as the QBATCHD, a single threaded queue for large and medium UBEs, and the QPROCESS queue for short UBEs run concurrently.

Oracle's UBE process performance metric is Number of Maximum Concurrent UBE processes at transaction rate, UBEs/minute.

Key Points and Best Practices

Four Oracle Solaris processors sets were used with Oracle Solaris Containers assigned to the processor sets as follows:

  • one JD Edwards EnterpriseOne Application server, two Oracle WebLogic Servers 11g Release 1 each coupled with an Oracle Web Tier HTTP server instances (online workload), each in an Oracle Solaris Container (three total),

  • one JD Edwards EnterpriseOne Application server (for batch only workload) in an Oracle Solaris Container,

  • Oracle Database 11g Release 2.0.3 database in an Oracle Solaris Container,

  • the Oracle database log writer.

Other items of note:

  • Each Oracle WebLogic vertical cluster, with twelve managed instances, was configured in a dedicated webserver container in order to load balance users' requests and to provide the infrastructure to support high number of users with ease of deployment and high availability.

  • The database redo logs were configured on the raw disk partitions.

  • The mixed batch workload of 44 short UBEs and 8 long UBEs was executed concurrently with the 12,000 online application users, producing a sustained rate of 198.5 UBE/min.

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 03/26/2013

SPARC T5-2 (SPARC T5-2 Server base package, 2xSPARC T5 16-core processors, 32x16GB-1066 DIMMS, 4x600GB 10K RPM 2.5. SAS-2 HDD,2x300GB SSDs, 4x Sun Flash Accelerator F40 PCIe Cards, 2x Power Cables) List Price $98,190. IBM Power 770 (IBM Power 770:9917 Model MMC, 2x3.3GHz 16-core, 32x one processor activation, 2xCEC Enclosure with IBM Bezel, I/O Backplane and System Midplane,2x Service Processor, 16x 0/64GB DDR3 Memory (4x16GB) DIMMS-1066MHz Power7 CoD Memory, 24x Activation of 1 GB DDR3 Power7 Memory, 10x Activation of 100GB DDR3 Power7 Memory, 2x Disk/Media Backplane. 2x 300GB SAS 15K RPM 2.5. HDD (AIX/Linux only), 1x SATA slimline DVD-RAM drive, 4x AC Power Supply 1925W) List Price $532,143. Source: ibm.com, collected 03/18/2013.

SPARC T5 Systems Produce Oracle TimesTen Benchmark World Record

The Oracle TimesTen In-Memory Database is optimized to run on Oracle's SPARC T5 processor platforms running Oracle Solaris 11. In this series of tests, systems with the new SPARC T5 processor were significantly faster than systems based on other processors. Two tests were run to explore TimesTen performance: a Mobile Call Processing test (based on customer workload) and Oracle's TimesTen Performance Throughput Benchmark (TPTBM). TimesTen version 11.2.2.4 was used for all tests.

  • On the TimesTen Performance Throughput Benchmark (TPTBM), SPARC T5-8 server produced a world record 59.9 million read transactions per second.

  • On the Mobile Call Processing test, the SPARC T5 processor achieves 2.4 times more throughput than the Intel Xeon E7-4870 processor. The two-chip SPARC T5-2 server is 22% faster than an x86 server with four Intel E7-4870 2.4 GHz processors.

  • On the TimesTen Performance Throughput Benchmark (TPTBM) read-only workload, the SPARC T5 processor achieves 2.2 times higher throughput than the Intel Xeon E7-4870 processor. On the same workload, the two-chip SPARC T5-2 server produces 10% more throughput than an x86 server with four Intel E7-4870 processors and has almost twice the performance of a 2-chip Intel E5-2680 system.

  • With the TPTBM read-only workload, the SPARC T5-8 server delivers 3.8x more throughput than a SPARC T5-2 Server, showing excellent scalability.

  • The SPARC T5 processor delivers over twice the performace of the previous generation SPARC T4 processor and over 4x the performace of the SPARC T3 processor, all in the same amount of space.

  • The SPARC T5-2 server delivers 2.4x the performace of the SPARC T4-2 server in the same 3U space. This is better performance than that of the SPARC T4-4 server which occupies 5U.

Performance Landscape

Mobile Call Processing Test Performance

Processor Tps
SPARC T5, 3.6 GHz 367,600
Intel Xeon E7-4870, 2.4 GHz 302,000
SPARC T4, 2.85 GHz 230,500

All systems measured using Oracle Solaris 11 and Oracle TimesTen In-Memory Database 11.2.2.4.1

TimesTen Performance Throughput Benchmark (TPTBM) Read-Only

System Processor Chips Tps Tps/
Chip
SPARC T5-8 SPARC T5, 3.6 GHz 8 59.9M 7.5M
SPARC T5-2 SPARC T5, 3.6 GHz 2 15.9M 7.9M
x86 Intel Xeon E7-4870, 2.4 GHz 4 14.5M 3.6M
SPARC T4-4 SPARC T4, 3.0 GHz 4 14.2M 3.6M
x86* Intel Xeon E5-2680, 2.7 GHz 2 8.5M 4.3
SPARC T4-2 SPARC T4, 2.85 GHz 2 6.5M 3.3M
SPARC T3-4 SPARC T3, 1.65 GHz 4 7.9M 1.9M
T5440 SPARC T2+, 1.4 GHz 4 3.1M 0.8M

All systems measured using Oracle Solaris 11 and Oracle TimesTen In-Memory Database 11.2.2.4.1

*Intel E5-2680 using Oracle Linux and Oracle TimesTen In-Memory Database 11.2.2.4.1

TimesTen Performance Throughput Benchmark (TPTBM) Update-Only

Processor Tps
SPARC T5, 3.6 GHz 1,031.7K
Intel Xeon E7-4870, 2.4 GHz 988.1K
Intel Xeon E5-2680, 2.7 GHz * 944.3K
SPARC T4, 3.0 GHz 678.0K

All systems measured using Oracle Solaris 11 and Oracle TimesTen In-Memory Database 11.2.2.4.1

*Intel E5-2680 using Oracle Linux and Oracle TimesTen In-Memory Database 11.2.2.4.1

Configuration Summary

Hardware Configurations:

SPARC T5-8 server
8 x SPARC T5 processors, 3.6 GHz
2 TB memory
1 x 8 Gbs FC Qlogic HBA
1 x 6 Gbs SAS HBA
2 x 300 GB internal disks
Oracle Solaris 11
TimesTen 11.2.2.4.1
1 x Sun Fire X4275 server configured as COMSTAR redo head (log)

SPARC T5-2 server
2 x SPARC T5 processors, 3.6 GHz
512 GB memory
1 x 8 Gbs FC Qlogic HBA
1 x 6 Gbs SAS HBA
2 x 300 GB internal disks
Oracle Solaris 11
TimesTen 11.2.2.4.1
1 x Sun Fire X4275 server configured as COMSTAR redo head (log)

SPARC T4-4 server
4 x SPARC T4 processors, 3.0 GHz
1 TB memory
1 x 8 Gbs FC Qlogic HBA
1 x 6 Gbs SAS HBA
6 x 300 GB internal disks
Oracle Solaris 11
TimesTen 11.2.2.4.1
Sun Storage F5100 Flash Array (80 x 24 GB flash modules)
1 x Sun Fire X4275 server configured as COMSTAR redo head (log)

SPARC T4-2 server
2 x SPARC T4 processors, 2.85 GHz
256 GB memory
1 x 8 Gbs FC Qlogic HBA
1 x 6 Gbs SAS HBA
4 x 300 GB internal disks
Oracle Solaris 11
TimesTen 11.2.2.4.1
Sun Storage F5100 Flash Array (40 x 24 GB flash modules)
1 x Sun Fire X4275 server configured as COMSTAR head

SPARC T3-4 server
4 x SPARC T3 processors, 1.6 GHz
512 GB memory
1 x 8 Gbs FC Qlogic HBA
8 x 146 GB internal disks
Oracle Solaris 11
TimesTen 11.2.2.4.1
1 x Sun Fire X4275 server configured as COMSTAR head

Intel Server x86_64
2 x Intel Xeon E5-2680 processors, 2.7 GHz
256 GB memory
4 x SSD SAS disks (log)
1 x 600 GB internal disks
Oracle Linux
TimesTen 11.2.2.4.1

Sun Server X2-4
4 x Intel Xeon E7-4870 processors, 2.4 GHz
512 GB memory
1 x 8 Gbs FC Qlogic HBA
6 x 146 GB internal disks
Oracle Solaris 11
TimesTen 11.2.2.4.1
1 x Sun Fire X4275 server configured as COMSTAR redo head (log)

Benchmark Descriptions

TimesTen Performance Throughput BenchMark (TPTBM) is shipped with TimesTen and measures the total throughput of the system. The benchmark workloads can be reads, inserts, updates, and delete operations, or a mix of them as required.

Mobile Call Processing is a customer-based workload for processing calls made by mobile phone subscribers. The workload has a mixture of read-only, update, and insert-only transactions. The peak throughput performance is measured from multiple concurrent processes executing the transactions until a peak performance is reached via saturation of the available resources.

Key Points and Best Practices

The Mobile Call Processing test utilized Oracle Solaris processor sets in all environments for optimum performance. This features isolates running processes from other processes in the system. Combined with parameters to limit memory pages to the lgroup within the processor set and isolating the processor set to a single processor within the system.

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 26 March 2013.

SPARC T5-8 Delivers Oracle OLAP World Record Performance

Oracle's SPARC T5-8 server delivered world record query performance with near real-time analytic capability using the Oracle OLAP Perf Version 3 workload running Oracle Database 11g Release 2 on Oracle Solaris 11.

  • The maximum query throughput on the SPARC T5-8 server is 1.6x higher than that of the 8-chip Intel Xeon E7-8870 server. Both systems had sub-second response time.

  • The SPARC T5-8 server with the Oracle Database demonstrated the ability to support at least 600 concurrent users querying OLAP cubes (with no think time), processing 2.93 million analytic queries per hour with an average response time of 0.66 seconds per query. This performance was enabled by keeping the entire cube in-memory utilizing the 4 TB of memory on the SPARC T5-8 server.

  • Assuming a 60 second think time between query requests, the SPARC T5-8 server can support approximately 49,450 concurrent users with the same 0.66 sec response time.

  • The SPARC T5-8 server delivered 4.3x times the maximum query throughput of a SPARC T4-4 server.

  • The workload uses a set of realistic BI queries that run against an OLAP cube based on a 4 billion row fact table of sales data. The 4 billion rows are partitioned by month spanning 10 years.

  • The combination of the Oracle Database with the Oracle OLAP option running on a SPARC T5-8 server supports live data updates occurring concurrently with minimally impacted user query executions.

Performance Landscape

Oracle OLAP Perf Version 3 Benchmark
Oracle cube base on 4 billion fact table rows
10 years of data partitioned by month
System Queries/
hour
Users* Average Response
Time (sec)
0 sec think time 60 sec think time
SPARC T5-8 2,934,000 600 49,450 0.66
8-chip Intel Xeon E7-8870 1,823,000 120 30,500 0.19
SPARC T4-4 686,500 150 11,580 0.71

Configuration Summary and Results

SPARC T5-8 Hardware Configuration:

1 x SPARC T5-8 server with
8 x SPARC T5 processors, 3.6 GHz
4 TB memory
Data Storage and Redo Storage
1 x Sun Storage F5100 Flash Array (with 80 FMODs)
Oracle Solaris 11.1
Oracle Database 11g Release 2 (11.2.0.3) with Oracle OLAP option

Sun Server X2-8 Hardware Configuration:

1 x Sun Server X2-8 with
8 x Intel Xeon E7-8870 processors, 2.4 GHz
512 GB memory
Data Storage and Redo Storage
3 x StorageTek 2540/2501 array pairs
Oracle Solaris 10 10/12
Oracle Database 11g Release 2 (11.2.0.2) with Oracle OLAP option

SPARC T4-4 Hardware Configuration:

1 x SPARC T4-4 server with
4 x SPARC T4 processors, 3.0 GHz
1 TB memory
Data Storage
1 x Sun Fire X4275 (using COMSTAR)
2 x Sun Storage F5100 Flash Array (each with 80 FMODs)
Redo Storage
1 x Sun Fire X4275 (using COMSTAR with 8 HDD)
Oracle Solaris 11 11/11
Oracle Database 11g Release 2 (11.2.0.3) with Oracle OLAP option

Benchmark Description

The Oracle OLAP Perf Version 3 benchmark is a workload designed to demonstrate and stress the ability of the OLAP Option to deliver fast query, near real-time updates and rich calculations using a multi-dimensional model in the context of the Oracle data warehousing.

The bulk of the benchmark entails running a number of concurrent users, each issuing typical multidimensional queries against an Oracle cube. The cube has four dimensions: time, product, customer, and channel. Each query user issues approximately 150 different queries. One query chain may ask for total sales in a particular region (e.g South America) for a particular time period (e.g. Q4 of 2010) followed by additional queries which drill down into sales for individual countries (e.g. Chile, Peru, etc.) with further queries drilling down into individual stores, etc. Another query chain may ask for yearly comparisons of total sales for some product category (e.g. major household appliances) and then issue further queries drilling down into particular products (e.g. refrigerators, stoves. etc.), particular regions, particular customers, etc.

While the core of every OLAP Perf benchmark is real world query performance, the benchmark itself offers numerous execution options such as varying data set sizes, number of users, numbers of queries for any given user and cube update frequency. Version 3 of the benchmark is executed with a much larger number of query streams than previous versions and used a cube designed for near real-time updates. The results produced by version 3 of the benchmark are not directly comparable to results produced by previous versions of the benchmark.

The near real-time update capability is implemented along the following lines. A large Oracle cube, H, is built from a 4 billion row star schema, containing data up until the end of last business day. A second small cube, D, is then created which will contain all of today's new data coming in from outside the world. It will be updated every L minutes with the data coming in within the last L minutes. A third cube, R, joins cubes H and D for reporting purposes much like a view might join data from two tables. Calculations are installed into cube R. The use of a reporting cube which draws data from different storage cubes is a common practice.

Query users are never locked out of query operations while new data is added to the update cube. The point of the demonstration is to show that an Oracle OLAP system can be designed which results in data being no more than L minutes out of date, where L may be as low as just a few minutes. This is what is meant by near real-time analytics.

Key Points and Best Practices

  • Update performance of the D cube was optimized by running update processes in the FX class with a priority greater than 0. The maximum lag time between updates to the source fact table and data availability to query users (what was referred to as L in the benchmark description) was less than 3 minutes for the benchmark environment on the SPARC T5-8 server.

  • Building and querying cubes with the Oracle OLAP option requires a large temporary tablespace. Normally temporary tablespaces would reside on disk storage. However, because the SPARC T5-8 server used in this benchmark had 4 TB of main memory, it was possible to use main memory for the OLAP temporary tablespace. This was done by using files in /tmp for the temporary tablespace datafiles.

  • Since typical BI users are often likely to issue similar queries, either with the same, or different, constants in the where clauses, setting the init.ora parameter "cursor_sharing" to "force" provides for additional query throughput and a larger number of potential users.

  • Assuming the normal Oracle initialization parameters (e.g. SGA, PGA, processes etc.) are appropriately set, out of the box performance for the OLAP Perf workload should be close to what is reported here. Additional performance resulted from (a)using memory for the OLAP temporary tablespace (b)setting "cursor_sharing" to force.

  • For a given number of query users with zero think time, the main measured metrics are the average query response time and the query throughput. A derived metric is the maximum number of users the system can support, with the same response time, assuming some non-zero think time. The calculation of this maximum is from the well-known "response-time law"

      N = (rt + tt) * tp

    where rt is the average response time, tt is the think time and tp is the measured throughput.

    Setting tt to 60 seconds, rt to 0.66 seconds and tp to 815 queries/sec (2,934,000 queries/hour), the above formula shows that the SPARC T5-8 server will support 49,450 concurrent users with a think time of 60 seconds and an average response time of 0.66 seconds.

    For more information about the "response-time law" see chapter 3 from the book "Quantitative System Performance" cited below.

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 03/26/2013.

SPARC T5-2 Obtains Oracle Internet Directory Benchmark World Record Performance

Oracle's SPARC T5-2 server running Oracle Internet Directory (OID, Oracle's LDAP Directory Server) on Oracle Solaris 11 achieved a record result for LDAP searches/second with 1000 clients.

  • The SPARC T5-2 server running Oracle Internet Directory on Oracle Solaris 11 achieved a result of 944,624 LDAP searches/sec with an average latency of 1.05 ms with 1000 clients.

  • The SPARC T5-2 server running Oracle Internet Directory demonstrated 2.7x better throughput and 39% better latency improvement over similarly configured OID and SPARC T4 benchmark environment.

  • The SPARC T5-2 server running Oracle Internet Directory demonstrates 39% better throughput and latency for LDAP searches on core-to-core comparison over an x86 system configured with two Intel Xeon X5675 processors.

  • Oracle Internet Directory achieved near linear scaling on the SPARC T5-2 server with 68,399 LDAP searches/sec with 2 cores to 944,624 LDAP searches/sec with 32 cores.

  • Oracle Internet Directory and the SPARC T5-2 server achieved up to 12,453 LDAP modifys/sec with an average latency of 3.9 msec for 50 clients.

Performance Landscape

Oracle Internet Directory Tests
System c/c/th Search Modify Add
ops/sec lat (msec) ops/sec lat (msec) ops/sec lat (msec)
SPARC T5-2 2/32/256 944,624 1.05 12,453 3.9 888 17.9
SPARC T4-4 4/32/256 682,000 1.46 12,000 4.0 835 19.0

In order to compare the SPARC T5-2 to a 12-core x86 system, only 1 processor and 12 cores was used in the SPARC T5-2.

Oracle Internet Directory Tests – Comparing Against x86
System c/c/th Search Compare Authentication
ops/sec lat (msec) ops/sec lat (msec) ops/sec lat (msec)
SPARC T5-2 1/12/96 417,000 1.19 274,185 1.82 149,623 3.30
x86 2 x Intel X5675 2/12/24 299,000 1.66 202,433 2.46 119,198 4.19

Scaling runs were also made on the SPARC T5-2 server.

Scaling of Search Tests – SPARC T5-2
Cores Clients ops/sec Latency (msec)
32 1000 944,624 1.05
24 1000 823,741 1.21
16 500 560,709 0.88
8 500 270,601 1.84
4 100 145,879 0.68
2 100 68,399 1.46

Configuration Summary

System Under Test:

SPARC T5-2
2 x SPARC T5 processors, 3.6 GHz
512 GB memory
4 x 300 GB internal disks
Flash Storage (used for database and log files)
1 x Sun Storage 2540-M2 (used for redo logs)
Oracle Solaris 11.1
Oracle Internet Directory 11g Release 1 PS6 (11.1.1.7.0)
Oracle Database 11g Enterprise Edition 11.2.0.3 (64-bit)

Benchmark Description

Oracle Internet Directory (OID) is Oracle's LDAPv3 Directory Server. The throughput for five key operations are measured — Search, Compare, Modify, Mix and Add.

LDAP Search Operations Test

This test scenario involved concurrent clients binding once to OID and then performing repeated LDAP Search operations. The salient characteristics of this test scenario is as follows:

  • SLAMD SearchRate job was used.
  • BaseDN of the search is root of the DIT, the scope is SUBTREE, the search filter is of the form UID=, DN and UID are the required attribute.
  • Each LDAP search operation matches a single entry.
  • The total number concurrent clients was 1000 and were distributed amongst two client nodes.
  • Each client binds to OID once and performs repeated LDAP Search operations, each search operation resulting in the lookup of a unique entry in such a way that no client looks up the same entry twice and no two clients lookup the same entry and all entries are searched randomly.
  • In one run of the test, random entries from the 50 Million entries are looked up in as many LDAP Search operations.
  • Test job was run for 60 minutes.

LDAP Compare Operations Test

This test scenario involved concurrent clients binding once to OID and then performing repeated LDAP Compare operations on userpassword attribute. The salient characteristics of this test scenario is as follows:

  • SLAMD CompareRate job was used.
  • Each LDAP compare operation matches user password of user.
  • The total number concurrent clients was 1000 and were distributed amongst two client nodes.
  • Each client binds to OID once and performs repeated LDAP compare operations.
  • In one run of the test, random entries from the 50 Million entries are compared in as many LDAP compare operations.
  • Test job was run for 60 minutes.

LDAP Modify Operations Test

This test scenario consisted of concurrent clients binding once to OID and then performing repeated LDAP Modify operations. The salient characteristics of this test scenario is as follows:

  • SLAMD LDAP modrate job was used.
  • A total of 50 concurrent LDAP clients were used.
  • Each client updates a unique entry each time and a total of 50 Million entries are updated.
  • Test job was run for 60 minutes.
  • Value length was set to 11.
  • Attribute that is being modified is not indexed.

LDAP Mixed Load Test

The test scenario involved both the LDAP search and LDAP modify clients enumerated above.

  • The ratio involved 60% LDAP search clients, 30% LDAP bind and 10% LDAP modify clients.
  • A total of 1000 concurrent LDAP clients were used and were distributed on 2 client nodes.
  • Test job was run for 60 minutes.

LDAP Add Load Test

The test scenario involved concurrent clients adding new entries as follows.

  • Slamd standard add rate job is used.
  • A total of 500,000 entries were added.
  • A total of 16 concurrent LDAP clients were used.
  • Slamd add's inetorgperson objectclass entry with 21 attributes (includes operational attributes).

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 26 March 2013.

SPARC T5-2 Scores Oracle FLEXCUBE Universal Banking Benchmark World Record Performance

Oracle's SPARC T5-2 server running Oracle FLEXCUBE Universal Banking Release 12 along with Oracle Database 11g Release 2 on Oracle Solaris 11 produced record results.

  • A SPARC T5-2 server running Oracle FLEXCUBE Universal Banking Release 12 and Oracle Real Application Clusters (RAC) Database 11g Release 2 processed 25 million accounts in 150 minutes for the End of Month workloads with an average utilization of 55% and 196 minutes utilizing 20 cores with an average cpu utilization of 85%.

  • A SPARC T5-2 server running Oracle FLEXCUBE Universal Banking Release 12 and Oracle Real Application Clusters (RAC) Database 11g Release 2 processed 25 million accounts in 56 minutes for the End of Day workload utilizing just 20 cores.

  • A SPARC T5-2 server running Oracle FLEXCUBE Universal Banking Release 12 achieved twice the throughput compared to a SPARC T4-4 server (which has twice the number of processors) for End of Month batch processing.

  • A SPARC T5-2 server running Oracle FLEXCUBE Universal Banking Release 12 achieved a record result processing 10.14 million accounts in 28 minutes for the End of Day workload with an average cpu utilization of 72% on a single server.

  • These results demonstrate how SPARC T5 processor systems along with Oracle Solaris 11 can benefit global, private and corporate financial institutions who are running Oracle FLEXCUBE Universal Banking. The uniquely co-engineered Oracle software and SPARC T5 processor based system unlock unique agile capabilities demanded by modern business environments.

  • The SPARC T5-2 system along with Oracle Solaris is able to provide a combination of uniquely essential characteristics that resonate with core values for a modern financial services institution.

  • The SPARC T5 processor based systems are capable of delivering higher performance and lower total cost of ownership (TCO) than older SPARC infrastructure, without introducing the unseen tax and risk of migrating applications away from older SPARC systems.

Performance Landscape

Oracle FLEXCUBE Universal Banking Release 12
End of Month Batch Processing
System Customer
Accounts
Time in Minutes Notes
SPARC T5-2 25M 150.66 RAC (two systems)
SPARC T5-2 10.14M 101.92 single instance
SPARC T4-4 10.14M 108.77 single instance
SPARC T4-4 5M 106.18 single instance, two chips

Oracle FLEXCUBE Universal Banking Release 12
End of Day Batch Processing
System Customer
Accounts
Time in Minutes Notes
SPARC T5-2 25M 56.05 RAC (two systems)
SPARC T5-2 10.14M 27.87 single instance

Configuration Summary

SPARC T5 Configuration:

1 x SPARC T5-2 with
2 x SPARC T5 processors, 3.6 GHz
512 GB memory
1 x SPARC T5-2 with
2 x SPARC T5 processors, 3.6 GHz
256 GB memory
Oracle Solaris 11 11/11
Oracle Database 11g Release 2 (RAC/ASM 11.2.0.3.0)
Oracle FLEXCUBE Universal Banking Release 12.0.1

SPARC T4 Configuration:

2 x SPARC T4-4, each with
4 x SPARC T4 processors, 3.0 GHz
512 GB memory
Oracle Solaris 11 11/11
Oracle Database 11g Release 2 (RAC/ASM 11.2.0.3.0)
Oracle FLEXCUBE Universal Banking Release 12.0.1

Storage Configuration:

3 x Sun Storage 6180 Array with
16 x 300 GB disks, 15K RPM (total of 48)
4 x Sun Storage CSM200 Expansion Trays, each with
16 x 73 GB disks, 15K RPM (total of 64)
Configured as RAID0, ASM external redundancy
Tests run with single instance DB (single node) and with ASM two nodes
ASM configuration identical on both 2 machines
Oracle Database 11g Release 2 ASM 11.2.0.3.0 64bit (19 TB)

Benchmark Description

The Oracle FLEXCUBE Universal Banking Release 12 benchmark models an actual customer bank with End of Cycle transaction batch jobs which typically execute during non-banking hours. This benchmark includes end of day accrual for savings and term deposit accounts, interest capitalization for saving accounts, and interest pay out for term deposit accounts. The results of the benchmark are certified by Oracle and a white paper is published.

End of cycle batch tests are conducted to measure the throughput capabilities of the system. It helps banks to decide the end of cycle processing window required to do the back office processing. The End of Day (EOD) batch test includes the following:

  • Mark End of Transaction Input
  • Value Dated Balance update
  • Interest and Charges (IC) Batch
  • Mark End of Financial Input
  • Mark End of Day
  • Date Change
  • Mark Transaction Input
The End of Month (EOM) batch test includes additional tests. These batches typically execute during non-banking hours. The goal is to ensure that the system is able to complete the batch operations for the planned volumes End of Day (EOD) within 60 minutes and End of Month (EOM) including interest and charges liquidation within 180 minutes.

 

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 26 March 2013.

SPARC T5-2 Performance Running Oracle Fusion Middleware SOA

Oracle's SPARC T5-2 server running Oracle Fusion Middleware SOA Suite 11g on Oracle Solaris 11 demonstrated 2.1x to 2.4x throughput improvement with 2x concurrency over a similarly configured SPARC T4-2 server for Fusion Order Demo and Oracle Service Bus (OSB) benchmark workloads using 5 KB message size.

  • Oracle Fusion Middleware SOA was deployed on virtualized environments using Oracle VM for SPARC to demonstrate consolidation of multiple SOA services onto a single system.

  • The benchmark demonstrates SPARC hardware crypto performance within an OSB service using 100-byte element encrypted with AES and signed with RSA128.

Performance Landscape

OSB Tests
System ch/co/th OS  Concurrency 
T5/T4 Test
SPARC T5-2
SPARC T5-2 (db)
1/8/64
2/32/256
Oracle
Solaris 11
144 2.1x http_passthrough
96 2.4x dyn_transform
64 2.3x body encryption

ch/co/th – chips, cores, threads


BPEL Test
System ch/co/th OS Users T5/T4 Test
SPARC T5-2
SPARC T5-2 (db)
1.5/24/192
2/32/256
S11 400 2.2x Fusion order demo

ch/co/th – chips, cores, threads

Configuration Summary

Application Server:

SPARC T5-2
2 x SPARC T5 processors, 3.6 GHz
256 GB memory
2 x 300 GB internal disks
Oracle Solaris 11.1
Oracle WebLogic 10.3.6
Oracle SOA 11.1.1.6 (PS5)
Oracle OSB 11.1.1.6 (PS5)
Oracle JDK 7

Database Server:

SPARC T5-2
2 x SPARC T5 processors, 3.6 GHz
256 GB memory
2 x 300 GB internal disks
1 x Sun Storage 6180, 16 x 146 GB SAS disks
Oracle Solaris 11.1
Oracle Database 11g Release 2 (11.2.0.3)

Benchmark Description

Three tests were performed as part of the Oracle SOA Suite profiling:

HTTP Passthrough (http_passthrough)

The client sends a 5 KB message to a HTTP Web Services Description Language (WSDL)-based proxy service on an Oracle Service Bus server. The proxy routes (using route action) the message to the backend servlet in a WLS domain. Oracle Service Bus monitoring is enabled as the message goes through the bus. The proxy's operation selection algorithm is SOAP Action Header. This workload involves more networking load than any of the other Oracle Service Bus microbenchmarks described.

Dynamic Transformation (dyn_transformation)

In this benchmark the HTTP proxy receives a 5 KB XML document. The XML document has an Xquery resource name in one of its leaf nodes. The pipeline uses an Xpath to retrieve the Xquery resource name and executes transformation on the inbound XML. The majority of CPU is spent on XML processing.

Body Encryption (body_encryption)

This benchmark tests the crypto performance within an Oracle Service Bus service. The client sends a 5 KB message, within which a 100-byte element is encrypted, to the WSDL-based Oracle Service Bus proxy service over HTTP. The WSDL binding references an Oracle Web Services Manager policy. The business service is also WSDL-based. The element is encrypted with AES and signed with RSA128. The encrypted element is decrypted, and the message is routed to the backend service as a clear SOAP message.

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 26 March 2013.

SPARC T5-1B Performance Running Oracle Communications ASAP

Oracle's SPARC T5-1B server module delivered outstanding results on Oracle Communications ASAP. The SPARC T5-1B server module ran Oracle Solaris 11 with Oracle Database 11g Release 2, Oracle WebLogic Server 11g and Oracle Communications ASAP version 7.2.

  • Running Oracle Communications ASAP, the SPARC T5-1B server module achieved 1,722 ASDLs (atomic network activation actions) per second, the highest throughput that has been achieved in the 12NEP test for a single Oracle Communications ASAP instance across any SPARC architecture.

  • The SPARC T5-1B server module running a single instance of the Oracle Communications ASAP application, with both the application and database tiers consolidated onto a single machine, easily supported the service activation volumes of 1,722 ASDLs/sec which is representative of a typical mobile operator with more than 100 million subscribers.

  • Oracle Communications ASAP v7.2 delivered 48% higher throughput on a the SPARC T5-1B server module when compared to the SPARC T4-2 server.

  • The SPARC T5 processor delivered over 2 times the throughput compared to the previous generation SPARC T4 processor.

Performance Landscape

ASAP 7.2.0 12NEP Test Results
System ASDLs/sec CPU Usage
SPARC T5-1B 1,722.2 44.8%
SPARC T4-2 1,114.3 42.7%

Configuration Summary

Hardware Configuration:

SPARC T5-1B server module
1 x SPARC T5 processor at 3.6 GHz
256 GB memory

SPARC T4-2 server
2 x SPARC T4 processors at 2.85 GHz
256 GB memory

Storage Configuration:

Pillar Axiom

Software Configuration:

Oracle Solaris 11.1
Oracle Database 11g Release 2 (11.2.0.3.0)
Oracle WebLogic Server 10.3.6.0
Oracle Communications ASAP 7.2.0 (SR2B23)
Oracle JDK 7 update 7

Benchmark Description

Oracle Communications ASAP is used to activate a variety of services including data, video, voice and content services across mobile, fixed and satellite networks. Typical activities performed include activating new subscribers and services, moving / adding / changing / deleting services of existing subscribers and deleting existing subscribers and services.

The throughput of ASAP is measured in atomic actions per second (or ASDLs/sec). An atomic action is a single command or operation that can be executed on a network element. Atomic actions are grouped together to form a common service action, where each service action typically relates to an orderable item, such as "GSM voice" or "voice mail" or "GSM data". One or more service actions are invoked by an order management system via an activation work order request.

The workload resembles a typical mobile order to activate a GSM subscriber. A single service action to add a subscriber consists of seven atomic actions where each atomic action executes a command on a network element. Each network element was serviced by a dedicated Network Element Processor (NEP). The ASAP benchmark can vary the number of NEPs, which correlate to the complexity of a Telco operator's environment.

See Also

Disclosure Statement

Copyright 2013, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 26 March 2013.

Thursday Nov 08, 2012

SPARC T4-4 Delivers World Record Performance on Oracle OLAP Perf Version 2 Benchmark

Oracle's SPARC T4-4 server delivered world record performance with subsecond response time on the Oracle OLAP Perf Version 2 benchmark using Oracle Database 11g Release 2 running on Oracle Solaris 11.

  • The SPARC T4-4 server achieved throughput of 430,000 cube-queries/hour with an average response time of 0.85 seconds and the median response time of 0.43 seconds. This was achieved by using only 60% of the available CPU resources leaving plenty of headroom for future growth.

Performance Landscape

Oracle OLAP Perf Version 2 Benchmark
4 Billion Fact Table Rows
System Queries/
hour
Users* Response Time (sec)
Average Median
SPARC T4-4 430,000 7,300 0.85 0.43

* Users - the supported number of users with a given think time of 60 seconds

Configuration Summary and Results

Hardware Configuration:

SPARC T4-4 server with
4 x SPARC T4 processors, 3.0 GHz
1 TB memory
Data Storage
1 x Sun Fire X4275 (using COMSTAR)
2 x Sun Storage F5100 Flash Array (each with 80 FMODs)
Redo Storage
1 x Sun Fire X4275 (using COMSTAR with 8 HDD)

Software Configuration:

Oracle Solaris 11 11/11
Oracle Database 11g Release 2 (11.2.0.3) with Oracle OLAP option

Benchmark Description

The Oracle OLAP Perf Version 2 benchmark is a workload designed to demonstrate and stress the Oracle OLAP product's core features of fast query, fast update, and rich calculations on a multi-dimensional model to support enhanced Data Warehousing.

The bulk of the benchmark entails running a number of concurrent users, each issuing typical multidimensional queries against an Oracle OLAP cube. The cube has four dimensions: time, product, customer, and channel. Each query user issues approximately 150 different queries. One query chain may ask for total sales in a particular region (e.g South America) for a particular time period (e.g. Q4 of 2010) followed by additional queries which drill down into sales for individual countries (e.g. Chile, Peru, etc.) with further queries drilling down into individual stores, etc. Another query chain may ask for yearly comparisons of total sales for some product category (e.g. major household appliances) and then issue further queries drilling down into particular products (e.g. refrigerators, stoves. etc.), particular regions, particular customers, etc.

Results from version 2 of the benchmark are not comparable with version 1. The primary difference is the type of queries along with the query mix.

Key Points and Best Practices

  • Since typical BI users are often likely to issue similar queries, with different constants in the where clauses, setting the init.ora prameter "cursor_sharing" to "force" will provide for additional query throughput and a larger number of potential users. Except for this setting, together with making full use of available memory, out of the box performance for the OLAP Perf workload should provide results similar to what is reported here.

  • For a given number of query users with zero think time, the main measured metrics are the average query response time, the median query response time, and the query throughput. A derived metric is the maximum number of users the system can support achieving the measured response time assuming some non-zero think time. The calculation of the maximum number of users follows from the well-known response-time law

      N = (rt + tt) * tp

    where rt is the average response time, tt is the think time and tp is the measured throughput.

    Setting tt to 60 seconds, rt to 0.85 seconds and tp to 119.44 queries/sec (430,000 queries/hour), the above formula shows that the T4-4 server will support 7,300 concurrent users with a think time of 60 seconds and an average response time of 0.85 seconds.

    For more information see chapter 3 from the book "Quantitative System Performance" cited below.

See Also

Disclosure Statement

Copyright 2012, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 11/2/2012.

Improved Performance on PeopleSoft Combined Benchmark using SPARC T4-4

Oracle's SPARC T4-4 server running Oracle's PeopleSoft HCM 9.1 combined online and batch benchmark achieved a world record 18,000 concurrent users experiencing subsecond response time while executing a PeopleSoft Payroll batch job of 500,000 employees in 32.4 minutes.

  • This result was obtained with a SPARC T4-4 server running Oracle Database 11g Release 2, a SPARC T4-4 server running PeopleSoft HCM 9.1 application server and a SPARC T4-2 server running Oracle WebLogic Server in the web tier.

  • The SPARC T4-4 server running the application tier used Oracle Solaris Zones which provide a flexible, scalable and manageable virtualization environment.

  • The average CPU utilization on the SPARC T4-2 server in the web tier was 17%, on the SPARC T4-4 server in the application tier it was 59%, and on the SPARC T4-4 server in the database tier was 47% (online and batch) leaving significant headroom for additional processing across the three tiers.

  • The SPARC T4-4 server used for the database tier hosted Oracle Database 11g Release 2 using Oracle Automatic Storage Management (ASM) for database files management with I/O performance equivalent to raw devices.

Performance Landscape

Results are presented for the PeopleSoft HRMS Self-Service and Payroll combined benchmark. The new result with 128 streams shows significant improvement in the payroll batch processing time with little impact on the self-service component response time.

PeopleSoft HRMS Self-Service and Payroll Benchmark
Systems Users Ave Response
Search (sec)
Ave Response
Save (sec)
Batch
Time (min)
Streams
SPARC T4-2 (web)
SPARC T4-4 (app)
SPARC T4-4 (db)
18,000 0.988 0.539 32.4 128
SPARC T4-2 (web)
SPARC T4-4 (app)
SPARC T4-4 (db)
18,000 0.944 0.503 43.3 64

The following results are for the PeopleSoft HRMS Self-Service benchmark that was previous run. The results are not directly comparable with the combined results because they do not include the payroll component.

PeopleSoft HRMS Self-Service 9.1 Benchmark
Systems Users Ave Response
Search (sec)
Ave Response
Save (sec)
Batch
Time (min)
Streams
SPARC T4-2 (web)
SPARC T4-4 (app)
2x SPARC T4-2 (db)
18,000 1.048 0.742 N/A N/A

The following results are for the PeopleSoft Payroll benchmark that was previous run. The results are not directly comparable with the combined results because they do not include the self-service component.

PeopleSoft Payroll (N.A.) 9.1 - 500K Employees (7 Million SQL PayCalc, Unicode)
Systems Users Ave Response
Search (sec)
Ave Response
Save (sec)
Batch
Time (min)
Streams
SPARC T4-4 (db)
N/A N/A N/A 30.84 96

Configuration Summary

Application Configuration:

1 x SPARC T4-4 server with
4 x SPARC T4 processors, 3.0 GHz
512 GB memory
Oracle Solaris 11 11/11
PeopleTools 8.52
PeopleSoft HCM 9.1
Oracle Tuxedo, Version 10.3.0.0, 64-bit, Patch Level 031
Java Platform, Standard Edition Development Kit 6 Update 32

Database Configuration:

1 x SPARC T4-4 server with
4 x SPARC T4 processors, 3.0 GHz
256 GB memory
Oracle Solaris 11 11/11
Oracle Database 11g Release 2
PeopleTools 8.52
Oracle Tuxedo, Version 10.3.0.0, 64-bit, Patch Level 031
Micro Focus Server Express (COBOL v 5.1.00)

Web Tier Configuration:

1 x SPARC T4-2 server with
2 x SPARC T4 processors, 2.85 GHz
256 GB memory
Oracle Solaris 11 11/11
PeopleTools 8.52
Oracle WebLogic Server 10.3.4
Java Platform, Standard Edition Development Kit 6 Update 32

Storage Configuration:

1 x Sun Server X2-4 as a COMSTAR head for data
4 x Intel Xeon X7550, 2.0 GHz
128 GB memory
1 x Sun Storage F5100 Flash Array (80 flash modules)
1 x Sun Storage F5100 Flash Array (40 flash modules)

1 x Sun Fire X4275 as a COMSTAR head for redo logs
12 x 2 TB SAS disks with Niwot Raid controller

Benchmark Description

This benchmark combines PeopleSoft HCM 9.1 HR Self Service online and PeopleSoft Payroll batch workloads to run on a unified database deployed on Oracle Database 11g Release 2.

The PeopleSoft HRSS benchmark kit is a Oracle standard benchmark kit run by all platform vendors to measure the performance. It's an OLTP benchmark where DB SQLs are moderately complex. The results are certified by Oracle and a white paper is published.

PeopleSoft HR SS defines a business transaction as a series of HTML pages that guide a user through a particular scenario. Users are defined as corporate Employees, Managers and HR administrators. The benchmark consist of 14 scenarios which emulate users performing typical HCM transactions such as viewing paycheck, promoting and hiring employees, updating employee profile and other typical HCM application transactions.

All these transactions are well-defined in the PeopleSoft HR Self-Service 9.1 benchmark kit. This benchmark metric is the weighted average response search/save time for all the transactions.

The PeopleSoft 9.1 Payroll (North America) benchmark demonstrates system performance for a range of processing volumes in a specific configuration. This workload represents large batch runs typical of a ERP environment during a mass update. The benchmark measures five application business process run times for a database representing large organization. They are Paysheet Creation, Payroll Calculation, Payroll Confirmation, Print Advice forms, and Create Direct Deposit File. The benchmark metric is the cumulative elapsed time taken to complete the Paysheet Creation, Payroll Calculation and Payroll Confirmation business application processes.

The benchmark metrics are taken for each respective benchmark while running simultaneously on the same database back-end. Specifically, the payroll batch processes are started when the online workload reaches steady state (the maximum number of online users) and overlap with online transactions for the duration of the steady state.

Key Points and Best Practices

  • Two PeopleSoft Domain sets with 200 application servers each on a SPARC T4-4 server were hosted in 2 separate Oracle Solaris Zones to demonstrate consolidation of multiple application servers, ease of administration and performance tuning.

  • Each Oracle Solaris Zone was bound to a separate processor set, each containing 15 cores (total 120 threads). The default set (1 core from first and third processor socket, total 16 threads) was used for network and disk interrupt handling. This was done to improve performance by reducing memory access latency by using the physical memory closest to the processors and offload I/O interrupt handling to default set threads, freeing up cpu resources for Application Servers threads and balancing application workload across 240 threads.

  • A total of 128 PeopleSoft streams server processes where used on the database node to complete payroll batch job of 500,000 employees in 32.4 minutes.

See Also

Disclosure Statement

Copyright 2012, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 8 November 2012.

Tuesday Oct 02, 2012

Performance of Oracle Business Intelligence Benchmark on SPARC T4-4

Oracle's SPARC T4-4 server configured with four SPARC T4 3.0 GHz processors delivered 25,000 concurrent users on Oracle Business Intelligence Enterprise Edition (BI EE) 11g benchmark using Oracle Database 11g Release 2 running on Oracle Solaris 10.

  • A SPARC T4-4 server running Oracle Business Intelligence Enterprise Edition 11g achieved 25,000 concurrent users with an average response time of 0.36 seconds with Oracle BI server cache set to ON.

  • The benchmark data clearly shows that the underlying hardware, SPARC T4 server, and the Oracle BI EE 11g (11.1.1.6.0 64-bit) platform scales within a single system supporting 25,000 concurrent users while executing 415 transactions/sec.

  • The benchmark demonstrated the scalability of Oracle Business Intelligence Enterprise Edition 11g 11.1.1.6.0, which was deployed in a vertical scale-out fashion on a single SPARC T4-4 server.

  • Oracle Internet Directory configured on SPARC T4 server provided authentication for the 25,000 Oracle BI EE users with sub-second response time.

  • A SPARC T4-4 with internal Solid State Drive (SSD) using the ZFS file system showed significant I/O performance improvement over traditional disk for the Web Catalog activity. In addition, ZFS helped get past the UFS limitation of 32767 sub-directories in a Web Catalog directory.

  • The multi-threaded 64-bit Oracle Business Intelligence Enterprise Edition 11g and SPARC T4-4 server proved to be a successful combination by providing sub-second response times for the end user transactions, consuming only half of the available CPU resources at 25,000 concurrent users, leaving plenty of head room for increased load.

  • The Oracle Business Intelligence on SPARC T4-4 server benchmark results demonstrate that comprehensive BI functionality built on a unified infrastructure with a unified business model yields best-in-class scalability, reliability and performance.

  • Oracle BI EE 11g is a newer version of Business Intelligence Suite with richer and superior functionality. Results produced with Oracle BI EE 11g benchmark are not comparable to results with Oracle BI EE 10g benchmark. Oracle BI EE 11g is a more difficult benchmark to run, exercising more features of Oracle BI.

Performance Landscape

Results for the Oracle BI EE 11g version of the benchmark. Results are not comparable to the Oracle BI EE 10g version of the benchmark.

Oracle BI EE 11g Benchmark
System Number of Users Response Time (sec)
1 x SPARC T4-4 (4 x SPARC T4 3.0 GHz) 25,000 0.36

Results for the Oracle BI EE 10g version of the benchmark. Results are not comparable to the Oracle BI EE 11g version of the benchmark.

Oracle BI EE 10g Benchmark
System Number of Users
2 x SPARC T5440 (4 x SPARC T2+ 1.6 GHz) 50,000
1 x SPARC T5440 (4 x SPARC T2+ 1.6 GHz) 28,000

Configuration Summary

Hardware Configuration:

SPARC T4-4 server
4 x SPARC T4-4 processors, 3.0 GHz
128 GB memory
4 x 300 GB internal SSD

Storage Configuration:

Sun ZFS Storage 7120
16 x 146 GB disks

Software Configuration:

Oracle Solaris 10 8/11
Oracle Solaris Studio 12.1
Oracle Business Intelligence Enterprise Edition 11g (11.1.1.6.0)
Oracle WebLogic Server 10.3.5
Oracle Internet Directory 11.1.1.6.0
Oracle Database 11g Release 2

Benchmark Description

Oracle Business Intelligence Enterprise Edition (Oracle BI EE) delivers a robust set of reporting, ad-hoc query and analysis, OLAP, dashboard, and scorecard functionality with a rich end-user experience that includes visualization, collaboration, and more.

The Oracle BI EE benchmark test used five different business user roles - Marketing Executive, Sales Representative, Sales Manager, Sales Vice-President, and Service Manager. These roles included a maximum of 5 different pre-built dashboards. Each dashboard page had an average of 5 reports in the form of a mix of charts, tables and pivot tables, returning anywhere from 50 rows to approximately 500 rows of aggregated data. The test scenario also included drill-down into multiple levels from a table or chart within a dashboard.

The benchmark test scenario uses a typical business user sequence of dashboard navigation, report viewing, and drill down. For example, a Service Manager logs into the system and navigates to his own set of dashboards using Service Manager. The BI user selects the Service Effectiveness dashboard, which shows him four distinct reports, Service Request Trend, First Time Fix Rate, Activity Problem Areas, and Cost Per Completed Service Call spanning 2002 to 2005. The user then proceeds to view the Customer Satisfaction dashboard, which also contains a set of 4 related reports, drills down on some of the reports to see the detail data. The BI user continues to view more dashboards – Customer Satisfaction and Service Request Overview, for example. After navigating through those dashboards, the user logs out of the application. The benchmark test is executed against a full production version of the Oracle Business Intelligence 11g Applications with a fully populated underlying database schema. The business processes in the test scenario closely represent a real world customer scenario.

See Also

Disclosure Statement

Copyright 2012, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 30 September 2012.

World Record Oracle E-Business Consolidated Workload on SPARC T4-2

Oracle set a World Record for the Oracle E-Business Suite Standard Medium multiple-online module benchmark using Oracle's SPARC T4-2 and SPARC T4-4 servers which ran the application and database.

  • Oracle's SPARC T4 servers demonstrate performance leadership and world-record results on Oracle E-Business Suite Applications R12 OLTP benchmark by publishing the first result using multiple concurrent online application modules with Oracle Database 11g Release 2 running Solaris.

  •  

  • This results shows that a multi-tier configuration of SPARC T4 servers running the Oracle E-Business Suite R12.1.2 application and Oracle Database 11g Release 2 is capable of supporting 4,100 online users with outstanding response-times, executing a mix of complex transactions consolidating 4 Oracle E-Business modules (iProcurement, Order Management, Customer Service and HR Self-Service).

  •  

  • The SPARC T4-2 server in the application tier utilized about 65% and the SPARC T4-4 server in the database tier utilized about 30%, providing significant headroom for additional Oracle E-Business Suite R12.1.2 processing modules, more online users, and future growth.

  •  

  • Oracle E-Business Suite Applications were run in Oracle Solaris Containers on SPARC T4 servers and provides a consolidation platform for multiple E-Business instances.

  •  

Performance Landscape

Multiple Online Modules (Self-Service, Order-Management, iProcurement, Customer-Service)
Medium Configuration
System Users Average
Response Time
90th Percentile
Response Time
SPARC T4-2 4,100 2.08 sec 2.52 sec

Configuration Summary

Application Tier Configuration:

1 x SPARC T4-2 server
2 x SPARC T4 processors, 2.85 GHz
256 GB memory
3 x 300 GB internal disks
Oracle Solaris 10
Oracle E-Business Suite 12.1.2

Database Tier Configuration:

1 x SPARC T4-4 server
4 x SPARC T4 processors, 3.0 GHz
256 GB memory
2 x 300 GB internal disks
Oracle Solaris 10
Oracle Solaris Containers
Oracle Database 11g Release 2

Storage Configuration:

1 x Sun Storage F5100 Flash Array (80 x 24 GB flash modules)

Benchmark Description

The Oracle R12 E-Business Suite Standard Benchmark combines online transaction execution by simulated users with multiple online concurrent modules to model a typical scenario for a global enterprise. The online component exercises the common UI flows which are most frequently used by a majority of our customers. This benchmark utilized four concurrent flows of OLTP transactions, for Order to Cash, iProcurement, Customer Service and HR Self-Service and measured the response times. The selected flows model simultaneous business activities inclusive of managing customers, services, products and employees.

See Also

Disclosure Statement

Oracle E-Business Suite R12 medium multiple-online module benchmark, SPARC T4-2, SPARC T4, 2.85 GHz, 2 chips, 16 cores, 128 threads, 256 GB memory, SPARC T4-4, SPARC T4, 3.0 GHz, 4 chips, 32 cores, 256 threads, 256 GB memory, average response time 2.08 sec, 90th percentile response time 2.52 sec, Oracle Solaris 10, Oracle Solaris Containers, Oracle E-Business Suite 12.1.2, Oracle Database 11g Release 2, Results as of 9/30/2012.

SPARC T4-4 Delivers World Record First Result on PeopleSoft Combined Benchmark

Oracle's SPARC T4-4 servers running Oracle's PeopleSoft HCM 9.1 combined online and batch benchmark achieved World Record 18,000 concurrent users while executing a PeopleSoft Payroll batch job of 500,000 employees in 43.32 minutes and maintaining online users response time at < 2 seconds.

  • This world record is the first to run online and batch workloads concurrently.

  • This result was obtained with a SPARC T4-4 server running Oracle Database 11g Release 2, a SPARC T4-4 server running PeopleSoft HCM 9.1 application server and a SPARC T4-2 server running Oracle WebLogic Server in the web tier.

  • The SPARC T4-4 server running the application tier used Oracle Solaris Zones which provide a flexible, scalable and manageable virtualization environment.

  • The average CPU utilization on the SPARC T4-2 server in the web tier was 17%, on the SPARC T4-4 server in the application tier it was 59%, and on the SPARC T4-4 server in the database tier was 35% (online and batch) leaving significant headroom for additional processing across the three tiers.

  • The SPARC T4-4 server used for the database tier hosted Oracle Database 11g Release 2 using Oracle Automatic Storage Management (ASM) for database files management with I/O performance equivalent to raw devices.

  • This is the first three tier mixed workload (online and batch) PeopleSoft benchmark also processing PeopleSoft payroll batch workload.

Performance Landscape

PeopleSoft HR Self-Service and Payroll Benchmark
Systems Users Ave Response
Search (sec)
Ave Response
Save (sec)
Batch
Time (min)
Streams
SPARC T4-2 (web)
SPARC T4-4 (app)
SPARC T4-4 (db)
18,000 0.944 0.503 43.32 64

Configuration Summary

Application Configuration:

1 x SPARC T4-4 server with
4 x SPARC T4 processors, 3.0 GHz
512 GB memory
1 x 600 GB SAS internal disks
4 x 300 GB SAS internal disks
1 x 100 GB and 2 x 300 GB internal SSDs
2 x 10 Gbe HBA
Oracle Solaris 11 11/11
PeopleTools 8.52
PeopleSoft HCM 9.1
Oracle Tuxedo, Version 10.3.0.0, 64-bit, Patch Level 031
Java Platform, Standard Edition Development Kit 6 Update 32

Database Configuration:

1 x SPARC T4-4 server with
4 x SPARC T4 processors, 3.0 GHz
256 GB memory
1 x 600 GB SAS internal disks
2 x 300 GB SAS internal disks
Oracle Solaris 11 11/11
Oracle Database 11g Release 2
PeopleTools 8.52
Oracle Tuxedo, Version 10.3.0.0, 64-bit, Patch Level 031

Web Tier Configuration:

1 x SPARC T4-2 server with
2 x SPARC T4 processors, 2.85 GHz
256 GB memory
2 x 300 GB SAS internal disks
1 x 300 GB internal SSD
1 x 100 GB internal SSD
Oracle Solaris 11 11/11
PeopleTools 8.52
Oracle WebLogic Server 10.3.4
Java Platform, Standard Edition Development Kit 6 Update 32

Storage Configuration:

1 x Sun Server X2-4 as a COMSTAR head for data
4 x Intel Xeon X7550, 2.0 GHz
128 GB memory
1 x Sun Storage F5100 Flash Array (80 flash modules)
1 x Sun Storage F5100 Flash Array (40 flash modules)

1 x Sun Fire X4275 as a COMSTAR head for redo logs
12 x 2 TB SAS disks with Niwot Raid controller

Benchmark Description

This benchmark combines PeopleSoft HCM 9.1 HR Self Service online and PeopleSoft Payroll batch workloads to run on a unified database deployed on Oracle Database 11g Release 2.

The PeopleSoft HRSS benchmark kit is a Oracle standard benchmark kit run by all platform vendors to measure the performance. It's an OLTP benchmark where DB SQLs are moderately complex. The results are certified by Oracle and a white paper is published.

PeopleSoft HR SS defines a business transaction as a series of HTML pages that guide a user through a particular scenario. Users are defined as corporate Employees, Managers and HR administrators. The benchmark consist of 14 scenarios which emulate users performing typical HCM transactions such as viewing paycheck, promoting and hiring employees, updating employee profile and other typical HCM application transactions.

All these transactions are well-defined in the PeopleSoft HR Self-Service 9.1 benchmark kit. This benchmark metric is the weighted average response search/save time for all the transactions.

The PeopleSoft 9.1 Payroll (North America) benchmark demonstrates system performance for a range of processing volumes in a specific configuration. This workload represents large batch runs typical of a ERP environment during a mass update. The benchmark measures five application business process run times for a database representing large organization. They are Paysheet Creation, Payroll Calculation, Payroll Confirmation, Print Advice forms, and Create Direct Deposit File. The benchmark metric is the cumulative elapsed time taken to complete the Paysheet Creation, Payroll Calculation and Payroll Confirmation business application processes.

The benchmark metrics are taken for each respective benchmark while running simultaneously on the same database back-end. Specifically, the payroll batch processes are started when the online workload reaches steady state (the maximum number of online users) and overlap with online transactions for the duration of the steady state.

Key Points and Best Practices

  • Two Oracle PeopleSoft Domain sets with 200 application servers each on a SPARC T4-4 server were hosted in 2 separate Oracle Solaris Zones to demonstrate consolidation of multiple application servers, ease of administration and performance tuning.

  • Each Oracle Solaris Zone was bound to a separate processor set, each containing 15 cores (total 120 threads). The default set (1 core from first and third processor socket, total 16 threads) was used for network and disk interrupt handling. This was done to improve performance by reducing memory access latency by using the physical memory closest to the processors and offload I/O interrupt handling to default set threads, freeing up cpu resources for Application Servers threads and balancing application workload across 240 threads.

See Also

Disclosure Statement

Oracle's PeopleSoft HR and Payroll combined benchmark, www.oracle.com/us/solutions/benchmark/apps-benchmark/peoplesoft-167486.html, results 09/30/2012.

Monday Oct 01, 2012

World Record Batch Rate on Oracle JD Edwards Consolidated Workload with SPARC T4-2

Oracle produced a World Record batch throughput for single system results on Oracle's JD Edwards EnterpriseOne Day-in-the-Life benchmark using Oracle's SPARC T4-2 server running Oracle Solaris Containers and consolidating JD Edwards EnterpriseOne, Oracle WebLogic servers and the Oracle Database 11g Release 2. The workload includes both online and batch workload.

  • The SPARC T4-2 server delivered a result of 8,000 online users while concurrently executing a mix of JD Edwards EnterpriseOne Long and Short batch processes at 95.5 UBEs/min (Universal Batch Engines per minute).

  • In order to obtain this record benchmark result, the JD Edwards EnterpriseOne, Oracle WebLogic and Oracle Database 11g Release 2 servers were executed each in separate Oracle Solaris Containers which enabled optimal system resources distribution and performance together with scalable and manageable virtualization.

  • One SPARC T4-2 server running Oracle Solaris Containers and consolidating JD Edwards EnterpriseOne, Oracle WebLogic servers and the Oracle Database 11g Release 2 utilized only 55% of the available CPU power.

  • The Oracle DB server in a Shared Server configuration allows for optimized CPU resource utilization and significant memory savings on the SPARC T4-2 server without sacrificing performance.

  • This configuration with SPARC T4-2 server has achieved 33% more Users/core, 47% more UBEs/min and 78% more Users/rack unit than the IBM Power 770 server.

  • The SPARC T4-2 server with 2 processors ran the JD Edwards "Day-in-the-Life" benchmark and supported 8,000 concurrent online users while concurrently executing mixed batch workloads at 95.5 UBEs per minute. The IBM Power 770 server with twice as many processors supported only 12,000 concurrent online users while concurrently executing mixed batch workloads at only 65 UBEs per minute.

  • This benchmark demonstrates more than 2x cost savings by consolidating the complete solution in a single SPARC T4-2 server compared to earlier published results of 10,000 users and 67 UBEs per minute on two SPARC T4-2 and SPARC T4-1.

  • The Oracle DB server used mirrored (RAID 1) volumes for the database providing high availability for the data without impacting performance.

Performance Landscape

JD Edwards EnterpriseOne Day in the Life (DIL) Benchmark
Consolidated Online with Batch Workload

System Rack
Units
(U)
Batch
Rate
(UBEs/m)
Online
Users
Users
/ U
Users
/ Core
Version
SPARC T4-2 (2 x SPARC T4, 2.85 GHz) 3 95.5 8,000 2,667 500 9.0.2
IBM Power 770 (4 x POWER7, 3.3 GHz, 32 cores) 8 65 12,000 1,500 375 9.0.2

Batch Rate (UBEs/m) — Batch transaction rate in UBEs per minute

Configuration Summary

Hardware Configuration:

1 x SPARC T4-2 server with
2 x SPARC T4 processors, 2.85 GHz
256 GB memory
4 x 300 GB 10K RPM SAS internal disk
2 x 300 GB internal SSD
2 x Sun Storage F5100 Flash Arrays

Software Configuration:

Oracle Solaris 10
Oracle Solaris Containers
JD Edwards EnterpriseOne 9.0.2
JD Edwards EnterpriseOne Tools (8.98.4.2)
Oracle WebLogic Server 11g (10.3.4)
Oracle HTTP Server 11g
Oracle Database 11g Release 2 (11.2.0.1)

Benchmark Description

JD Edwards EnterpriseOne is an integrated applications suite of Enterprise Resource Planning (ERP) software. Oracle offers 70 JD Edwards EnterpriseOne application modules to support a diverse set of business operations.

Oracle's Day in the Life (DIL) kit is a suite of scripts that exercises most common transactions of JD Edwards EnterpriseOne applications, including business processes such as payroll, sales order, purchase order, work order, and manufacturing processes, such as ship confirmation. These are labeled by industry acronyms such as SCM, CRM, HCM, SRM and FMS. The kit's scripts execute transactions typical of a mid-sized manufacturing company.

  • The workload consists of online transactions and the UBE – Universal Business Engine workload of 61 short and 4 long UBEs.

  • LoadRunner runs the DIL workload, collects the user’s transactions response times and reports the key metric of Combined Weighted Average Transaction Response time.

  • The UBE processes workload runs from the JD Enterprise Application server.

    • Oracle's UBE processes come as three flavors:

      • Short UBEs < 1 minute engage in Business Report and Summary Analysis,

      • Mid UBEs > 1 minute create a large report of Account, Balance, and Full Address,

      • Long UBEs > 2 minutes simulate Payroll, Sales Order, night only jobs.

    • The UBE workload generates large numbers of PDF files reports and log files.

    • The UBE Queues are categorized as the QBATCHD, a single threaded queue for large and medium UBEs, and the QPROCESS queue for short UBEs run concurrently.

Oracle's UBE process performance metric is Number of Maximum Concurrent UBE processes at transaction rate, UBEs/minute.

Key Points and Best Practices

Two JD Edwards EnterpriseOne Application Servers, two Oracle WebLogic Servers 11g Release 1 coupled with two Oracle Web Tier HTTP server instances and one Oracle Database 11g Release 2 database on a single SPARC T4-2 server were hosted in separate Oracle Solaris Containers bound to four processor sets to demonstrate consolidation of multiple applications, web servers and the database with best resource utilizations.

  • Interrupt fencing was configured on all Oracle Solaris Containers to channel the interrupts to processors other than the processor sets used for the JD Edwards Application server, Oracle WebLogic servers and the database server.

  • A Oracle WebLogic vertical cluster was configured on each WebServer Container with twelve managed instances each to load balance users' requests and to provide the infrastructure that enables scaling to high number of users with ease of deployment and high availability.

  • The database log writer was run in the real time RT class and bound to a processor set.

  • The database redo logs were configured on the raw disk partitions.

  • The Oracle Solaris Container running the Enterprise Application server completed 61 Short UBEs, 4 Long UBEs concurrently as the mixed size batch workload.

  • The mixed size UBEs ran concurrently from the Enterprise Application server with the 8,000 online users driven by the LoadRunner.

See Also

Disclosure Statement

Copyright 2012, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 09/30/2012.

Oracle TimesTen In-Memory Database Performance on SPARC T4-2

The Oracle TimesTen In-Memory Database is optimized to run on Oracle's SPARC T4 processor platforms running Oracle Solaris 11 providing unsurpassed scalability, performance, upgradability, protection of investment and return on investment. The following demonstrate the value of combining Oracle TimesTen In-Memory Database with SPARC T4 servers and Oracle Solaris 11:

On a Mobile Call Processing test, the 2-socket SPARC T4-2 server outperforms:

  • Oracle's SPARC Enterprise M4000 server (4 x 2.66 GHz SPARC64 VII+) by 34%.

  • Oracle's SPARC T3-4 (4 x 1.65 GHz SPARC T3) by 2.7x, or 5.4x per processor.

Utilizing the TimesTen Performance Throughput Benchmark (TPTBM), the SPARC T4-2 server protects investments with:

  • 2.1x the overall performance of a 4-socket SPARC Enterprise M4000 server in read-only mode and 1.5x the performance in update-only testing. This is 4.2x more performance per processor than the SPARC64 VII+ 2.66 GHz based system.

  • 10x more performance per processor than the SPARC T2+ 1.4 GHz server.

  • 1.6x better performance per processor than the SPARC T3 1.65 GHz based server.

In replication testing, the two socket SPARC T4-2 server is over 3x faster than the performance of a four socket SPARC Enterprise T5440 server in both asynchronous replication environment and the highly available 2-Safe replication. This testing emphasizes parallel replication between systems.

Performance Landscape

Mobile Call Processing Test Performance

System Processor Sockets/Cores Tps Tps/
Socket
SPARC T4-2 SPARC T4, 2.85 GHz 2 16 218,400 109,200
M4000 SPARC64 VII+, 2.66 GHz 4 16 162,900 40,725
SPARC T3-4 SPARC T3, 1.65 GHz 4 64 80,400 20,100

TimesTen Performance Throughput Benchmark (TPTBM) Read-Only

System Processor Sockets/Cores Tps Tps/
Socket
SPARC T4-2 SPARC T4, 2.85 GHz 2 16 6.5M 3.3M
SPARC T3-4 SPARC T3, 1.65 GHz 4 64 7.9M 2.0M
M4000 SPARC64 VII+, 2.66 GHz 4 16 3.1M 0.8M
T5440 SPARC T2+, 1.4 GHz 4 32 3.1M 0.8M

TimesTen Performance Throughput Benchmark (TPTBM) Update-Only

System Processor Sockets/Cores Tps Tps/
Socket
SPARC T4-2 SPARC T4, 2.85 GHz 2 16 547,800 273,900
M4000 SPARC64 VII+, 2.66 GHz 4 16 363,800 90,950
SPARC T3-4 SPARC T3, 1.65 GHz 4 64 240,250 60,125

TimesTen Replication Tests

System Processor Sockets/Cores Asynchronous 2-Safe
SPARC T4-2 SPARC T4, 2.85 GHz 2 16 38,024 13,701
SPARC T5440 SPARC T2+, 1.4 GHz 4 32 11,621 4,615

Configuration Summary

Hardware Configurations:

SPARC T4-2 server
2 x SPARC T4 processors, 2.85 GHz
256 GB memory
1 x 8 Gbs FC Qlogic HBA
1 x 6 Gbs SAS HBA
4 x 300 GB internal disks
Sun Storage F5100 Flash Array (40 x 24 GB flash modules)
1 x Sun Fire X4275 server configured as COMSTAR head

SPARC T3-4 server
4 x SPARC T3 processors, 1.6 GHz
512 GB memory
1 x 8 Gbs FC Qlogic HBA
8 x 146 GB internal disks
1 x Sun Fire X4275 server configured as COMSTAR head

SPARC Enterprise M4000 server
4 x SPARC64 VII+ processors, 2.66 GHz
128 GB memory
1 x 8 Gbs FC Qlogic HBA
1 x 6 Gbs SAS HBA
2 x 146 GB internal disks
Sun Storage F5100 Flash Array (40 x 24 GB flash modules)
1 x Sun Fire X4275 server configured as COMSTAR head

Software Configuration:

Oracle Solaris 11 11/11
Oracle TimesTen 11.2.2.4

Benchmark Descriptions

TimesTen Performance Throughput BenchMark (TPTBM) is shipped with TimesTen and measures the total throughput of the system. The workload can test read-only, update-only, delete and insert operations as required.

Mobile Call Processing is a customer-based workload for processing calls made by mobile phone subscribers. The workload has a mixture of read-only, update, and insert-only transactions. The peak throughput performance is measured from multiple concurrent processes executing the transactions until a peak performance is reached via saturation of the available resources.

Parallel Replication tests using both asynchronous and 2-Safe replication methods. For asynchronous replication, transactions are processed in batches to maximize the throughput capabilities of the replication server and network. In 2-Safe replication, also known as no data-loss or high availability, transactions are replicated between servers immediately emphasizing low latency. For both environments, performance is measured in the number of parallel replication servers and the maximum transactions-per-second for all concurrent processes.

See Also

Disclosure Statement

Copyright 2012, Oracle and/or its affiliates. All rights reserved. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Results as of 1 October 2012.

About

BestPerf is the source of Oracle performance expertise. In this blog, Oracle's Strategic Applications Engineering group explores Oracle's performance results and shares best practices learned from working on Enterprise-wide Applications.

Index Pages
Search

Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today