By Brian-Oracle on Jan 12, 2012
Oracle's Netra SPARC T4-2 server equipped with two SPARC T4 processors running at 2.85 GHz delivered a World Record result of 454.52 SPECjvm2008 Peak ops/m on the SPECjvm2008 benchmark. This result just eclipsed the previous record which was run on a similar product, Oracle's SPARC T4-2 server, which is also a two SPARC T4 processor based system.
The Netra SPARC T4-2 server demonstrates 41% better performance than the SPARC T3-2 server and similar performance to Oracle's SPARC T4-2 server.
The Netra SPARC T4-2 server running the SPECjvm2008 benchmark achieved a score of 454.52 SPECjvm2008 Peak ops/m while the Sun Blade X6270 server module achieved 317.13 SPECjvm2008 Base ops/m.
The Netra SPARC T4-2 server with hardware cryptography acceleration greatly increases performance with subtests using AES and RSA encryption ciphers.
This result was produced using Oracle Solaris 11 and Oracle JDK 7 Update 2.
There are no SPECjvm2008 results published by IBM on POWER7 based systems.
The Netra SPARC T4-2 server demonstrates Oracle's position of leadership in Java-based computing by publishing world record results for the SPECjvm2008 benchmark.
Complete benchmark results are at the SPECjvm2008 website.
SPECjvm2008 Performance Chart
(ordered by performance)
|Netra SPARC T4-2||2 x 2.85 GHz SPARC T4||-||454.52|
|SPARC T4-2||2 x 2.85 GHz SPARC T4||-||454.25|
|SPARC T3-2||2 x 1.65 GHz SPARC T3||-||320.52|
|Sun Blade X6270||2 x 2.93 GHz Intel X5570||317.13||-|
base: SPECjvm2008 Base ops/m (bigger is better)
peak: SPECjvm2008 Peak ops/m (bigger is better)
SPEC allows base and peak to be submitted separately. The base metric does not allow any optimization of the JVM, the peak metric allows optimization.
256 GB memory
Java Platform, Standard Edition, JDK 7 Update 2
SPECjvm2008 (Java Virtual Machine Benchmark) is a benchmark suite for measuring the performance of a Java Runtime Environment (JRE), containing several real life applications and benchmarks focusing on core Java functionality. The suite focuses on the performance of the JRE executing a single application; it reflects the performance of the hardware processor and memory subsystem, but has low dependence on file I/O and includes no network I/O across machines.
The SPECjvm2008 workload mimics a variety of common general purpose application computations. These characteristics reflect the intent that this benchmark will be applicable to measuring basic Java performance on a wide variety of both client and server systems.
SPECjvm2008 benchmark highlights:
- Leverages real life applications (like derby, sunflow, and javac) and area-focused benchmarks (like xml, serialization, crypto, and scimark).
- Also measures the performance of the operating system and hardware in the context of executing the JRE.
The current rules for the benchmark allow either base or peak to be run. The base run is done without any tuning of the JVM to improve the out of the box performance. The peak run allows tuning of the JVM.
Key Points and Best Practices
- Enhancements to the JVM had a major impact on performance, especially for the security tests.
- SPEC website
- SPECjvm2008 website
- Netra SPARC T4-2 Server
SPEC and SPECjvm are registered trademarks of Standard Performance Evaluation Corporation. Results from www.spec.org and this report as of 1/9/2012. Netra SPARC T4-2 454.52 SPECjvm2008 Peak ops/m submitted for review, SPARC T4-2 454.25 SPECjvm2008 Peak ops/m, SPARC T3-2 320.52 SPECjvm2008 Peak ops/m, Sun Blade X6270 317.13 SPECjvm2008 Base ops/m.