X

Everything you want and need to know about Oracle SPARC systems performance

SPECjEnterprise2010: SPARC T7-1 World Record with Single Application Server Using 1 to 4 Chips

Brian Whitney
Principal Software Engineer

Oracle's SPARC T7-1 servers have set a world record for the SPECjEnterprise2010 benchmark for solutions using a single application server with one to four chips.  The result of 25,818.85 SPECjEnterprise2010 EjOPS used two SPARC T7-1 servers, one server for the application tier and the other server for the database tier.

  • The SPARC T7-1 servers obtained a result of 25,093.06 SPECjEnterprise2010 EjOPS using encrypted data. This secured result used Oracle Advanced Security Transparent Data Encryption (TDE) for the application database tablespaces with the AES-256-CFB cipher. The network connection between the application server and the database server was also encrypted using the secure JDBC.

  • The SPARC T7-1 server solution delivered 34% more performance compared to the two-chip IBM x3650 M5 server result of 19,282.14 SPECjEnterprise2010 EjOPS. 

  • The SPARC T7-1 server solution delivered 14% more performance compared to the four-chip IBM Power System S824 server result of 22,543.34 SPECjEnterprise2010 EjOPS.

  • The SPARC T7-1 server based results demonstrated 20% more performance compared to the Oracle Server X5-2 system result of 21,504.30 SPECjEnterprise2010 EjOPS. Oracle holds the top x86 two-chip application server SPECjEnterprise2010 result.

  • The application server used Oracle Fusion Middleware components including the Oracle WebLogic 12.1 application server and Java HotSpot(TM) 64-Bit Server VM on Solaris, version 1.8.0_60. The database server was configured with Oracle Database 12c Release 1.

  • For the secure result, the application data was encrypted in the Oracle database using the Oracle Advanced Security Transparent Data Encryption (TDE) feature. Hardware accelerated cryptography support in the SPARC M7 processor for the AES-256-CFB cipher was used to provide data security.

  • The benchmark performance using the secure SPARC T7-1 server configuration with encryption was less than 3% when compared to the peak result.

  • This result demonstrated less than 1 second average response times for all SPECjEnterprise2010 transactions and represents Jave EE 5.0 transactions generated by over 210,000 users.

Performance Landscape

Select single application server results. Complete benchmark results are at the SPEC website, SPECjEnterprise2010 Results.

SPECjEnterprise2010 Performance Chart
10/25/2015
Submitter EjOPS* Java EE Server DB Server Notes
Oracle 25,818.85 1 x SPARC T7-1
1 x 4.13 GHz SPARC M7
Oracle WebLogic 12c (12.1.3)
1 x SPARC T7-1
1 x 4.13 GHz SPARC M7
Oracle Database 12c (12.1.0.2)
-
Oracle 25,093.06 1 x SPARC T7-1
1 x 4.13 GHz SPARC M7
Oracle WebLogic 12c (12.1.3)
Network Data Encryption for JDBC
1 x SPARC T7-1
1 x 4.13 GHz SPARC M7
Oracle Database 12c (12.1.0.2)
Transparent Data Encryption
Secure
IBM 22,543.34 1 x IBM Power S824
4 x 3.5 GHz POWER 8
WebSphere Application Server V8.5
1 x IBM Power S824
4 x 3.5 GHz POWER 8
IBM DB2 10.5 FP3
-
Oracle 21,504.30 1 x Oracle Server X5-2
2 x 2.3 GHz Intel Xeon E5-2699 v3
Oracle WebLogic 12c (12.1.3)
1 x Oracle Server X5-2
2 x 2.3 GHz Intel Xeon E5-2699 v3
Oracle Database 12c (12.1.0.2)
COD
IBM 19,282.14 1 x System x3650 M5
2 x 2.6 GHz Intel Xeon E5-2697 v3
WebSphere Application Server V8.5
1 x System x3850 X6
4 x 2.8 GHz Intel Xeon E7-4890 v2
IBM DB2 10.5 FP5
-

* SPECjEnterprise2010 EjOPS (bigger is better)

The Cluster on Die (COD) mode is a BIOS setting that effectively splits the chip in half, making the operating system think it has twice as many chips as it does (in this case, four, 9 core chips).  Intel has stated that COD is appropriate only for highly NUMA optimized workloads.  Dell has shown that there is a 3.7x slower bandwidth to the other half of the chip split by COD.

Configuration Summary

Application Server:

1 x SPARC T7-1 server, with
1 x SPARC M7 processor (4.13 GHz)
256 GB memory (16 x 16 GB)
2 x 600 GB SAS HDD
2 x 400 GB SAS SSD
3 x Sun Dual Port 10 GbE PCIe 2.0 Networking card with Intel 82599 10 GbE Controller
Oracle Solaris 11.3 (11.3.0.0.30)
Oracle WebLogic Server 12c (12.1.3)
Java HotSpot(TM) 64-Bit Server VM on Solaris, version 1.8.0_60

Database Server:

1 x SPARC T7-1 server, with
1 x SPARC M7 processor (4.13 GHz)
512 GB memory (16 x 32 GB)
2 x 600 GB SAS HDD
1 x Sun Storage 16 Gb Fibre Channel Universal HBA
Oracle Solaris 11.3 (11.3.0.0.30)
Oracle Database 12c (12.1.0.2)

Storage Servers:

1 x Oracle Server X5-2L (8-Drive), with
2 x Intel Xeon Processor E5-2699 v3 (2.3 GHz)
32 GB memory
1 x Sun Storage 16 Gb Fibre Channel Universal HBA
4 x 1.6 TB NVMe SSD
2 x 600 GB SAS HDD
Oracle Solaris 11.3 (11.3.0.0.30)

1 x Oracle Server X5-2L (24-Drive), with
2 x Intel Xeon Processor E5-2699 v3 (2.3 GHz)
32 GB memory
1 x Sun Storage 16 Gb Fibre Channel Universal HBA
14 x 600 GB SAS HDD
Oracle Solaris 11.3 (11.3.0.0.30)

1 x Brocade 6510 16 Gb FC switch

Benchmark Description

SPECjEnterprise2010 is the third generation of the SPEC organization's J2EE end-to-end industry standard benchmark application. The SPECjEnterprise2010 benchmark has been designed and developed to cover the Java EE 5 specification's significantly expanded and simplified programming model, highlighting the major features used by developers in the industry today. This provides a real world workload driving the Application Server's implementation of the Java EE specification to its maximum potential and allowing maximum stressing of the underlying hardware and software systems,

  • The web zone, servlets, and web services
  • The EJB zone
  • JPA 1.0 Persistence Model
  • JMS and Message Driven Beans
  • Transaction management
  • Database connectivity

Moreover, SPECjEnterprise2010 also heavily exercises all parts of the underlying infrastructure that make up the application environment, including hardware, JVM software, database software, JDBC drivers, and the system network.

The primary metric of the SPECjEnterprise2010 benchmark is jEnterprise Operations Per Second (SPECjEnterprise2010 EjOPS). The primary metric for the SPECjEnterprise2010 benchmark is calculated by adding the metrics of the Dealership Management Application in the Dealer Domain and the Manufacturing Application in the Manufacturing Domain. There is NO price/performance metric in this benchmark.

Key Points and Best Practices

  • Four Oracle WebLogic server instances on the SPARC T7-1 server were hosted in 4 separate Oracle Solaris Zones.
  • The Oracle WebLogic application servers were executed in the FX scheduling class to improve performance by reducing the frequency of context switches.
  • The Oracle log writer process was run in the RT scheduling class.

See Also

Disclosure Statement

SPEC and the benchmark name SPECjEnterprise are registered trademarks of the Standard Performance Evaluation Corporation. Results from www.spec.org as of 10/25/2015.  SPARC T7-1, 25,818.85 SPECjEnterprise2010 EjOPS (unsecure); SPARC T7-1, 25,093.06 SPECjEnterprise2010 EjOPS (secure); Oracle Server X5-2, 21,504.30 SPECjEnterprise2010 EjOPS (unsecure); IBM Power S824, 22,543.34 SPECjEnterprise2010 EjOPS (unsecure); IBM x3650 M5, 19,282.14 SPECjEnterprise2010 EjOPS (unsecure).

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.Captcha
Oracle

Integrated Cloud Applications & Platform Services