Sun Fire / Sun SPARC Enterprise T2000 Server
- A single Sun Fire T2000 Server configured with just a single, 8 Core UltraSPARC T1 processor and 64GB of memory running the Sun Java Application Server 9.1 delivered 883.66 JOPS for the SPECjAppServer2004 Benchmark in the Standard category. The Sun Fire T2000 Server consumed only 315 watts of power
- The HP rx2660 system based on two dual core Itanium 2 processors and running the Oracle Application Server consumed 1.8x more power, despite delivering slightly lower performance.
- When compared to the IBM p5 505Q based on the latest Power 5+ Quad Core Module and running the WebSphere Application Server, the Sun Fire T2000 Server delivered over 1.4x higher performance and performance per watt.
- This result demonstrates the throughput and power efficiency of the Sun Fire T2000 Server with the Solaris 10 OS running open source application servers.
- Sun is the only vendor that has published results using an open source application server, demonstrating full commitment to these price-efficient JEE alternatives.
The Sun Fire T2000 / SPARC EnterpriseT2000 server running the open source Solaris 10 OS and Sun Java System Application Server 9.1 achieved the highest ever open source application server result on the SPECjAppServer2004 benchmark.
In 2006 the GlassFish community delivered the world's first Java EE 5 compatible application server. With these results the GlassFish community has delivered the world's fastest open source application server. In addition to record setting performance, GlassFish V2 is a robust and production quality Java EE 5 compatible application server ready for enterprise deployments. GlassFish V2 is available from Sun Microsystems as the Sun Java System Application Server 9.1.
The record-setting benchmark was comprised of two award-winning Sun Fire / SPARC Enterprise T2000 servers based on the open source UltraSPARC T1 processor. The Sun Fire T2000 server with 1.4 GHz 8-core UltraSPARC T1 processor was used to run a single instance of the Sun Java System Application Server and the second Sun Fire T2000 server with a single 1 GHz 6-core UltraSPARC T1 processor was used to run IBM DB2 9.0 database software.
These results beat out proprietary HP Itanium 2 and IBM Power 5+ systems with higher levels of performance at lower levels of power consumption.
The unparalleled efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 Server running Solaris 10 enables customers to breakthrough traditional power and space limitations in their data centers to process more transactions and serve more customers than ever before. At the same time, organizations reduce software licensing costs as a result of both the single socket design of the Sun Fire T2000 and the excellent throughput demonstrated on free and open source software. Operational costs are also reduced through lower power, cooling and space consumption, enabling organizations to reduce their carbon footprint.
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the volume of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions.
Comparison Table |
|
Benchmark Outcome (21)
(22) SPECjAppServer2004 Sun Fire T2000 (8 cores, 1 chip) 883.66 JOPS@Standard. SPECjAppServer2004 IBM p5 505Q (4 cores, 1chip) 613.38 JOPS@Standard. SPECjAppServer2004 HP rx2660 (4 cores, 2 chips) 874.17 JOPS@Standard. SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. All results from www.spec.org as of 07/17/07). HP rx2660 power calculated as 70% of max input power reported 07/17/07: http://h18004.www1.hp.com/products/quickspecs/12698_div/12698_div.HTML#Technical%20Specifications IBM p505 power specifications from 80% of maximum report power consumption published in "Facts and Features Report", 03/27/07, posted at ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF Sun Fire T2000 server power consumption taken from measurements made during the benchmark run.
- A single Sun Fire T2000 Server configured with just a single, 8 Core UltraSPARC T1 processor and 16GB of memory running PostgreSQL 8.2 was able to support two Sun Fire X4200 servers in delivering 813.73 JOPS for the SPECjAppServer2004 Benchmark in the Standard category. The Sun Fire T2000 Server consumed only 286 watts of power at peak and by using the open source Solaris OS and PostgreSQL database, delivered a database Price / Performance of just $26.42 per JOP.
- The HP rx2660 system based on two dual core Itanium 2 processors used Oracle Database 10g which increased the overall cost by over 4x when compared to the Sun Fire T2000 solution. The HP system consumed 50% more power, resulting in the Sun Fire T2000 delivering 1.4x higher performance per watt.
- This result demonstrates the throughput and power efficiency of the Sun Fire T2000 Server with the Solaris 10 OS running as an OLTP database server, with the added benefit of requiring no database, operating system or application server licenses to reduce costs and TCO.
- The benchmark was run entirely on Open Source software: GlassFish Application Server, PostgreSQL and Solaris.
- Sun is the only vendor that has published SPECjAppServer2004 results using Open Source databases (MySQL and PostgreSQL), demonstrating full commitment to these price-efficient database alternatives.
One Sun Fire T2000 Server (based on OpenSPARC) running the open source Solaris OS, and PostgreSQL database was able to support a dual node Sun Fire X4200 server running the open source Solaris OS, and Glassfish application server configuration in delivering 813.73 SPECjAppServer2004 JOPS. This represents the highest performance of any open source database solution on this benchmark, and delivers over 4x better Price / Performance than an equivalent result achieved with an HP Itanium 2 system, at 50% lower power consumption.
The unparalleled efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 Server running Solaris 10 enables customers to breakthrough traditional power and space limitations in their data centers to process more transactions and serve more customers than ever before. At the same time, organizations reduce software licensing costs as a result of both the single socket design of the Sun Fire T2000 and the excellent throughput demonstrated on Free and Open Source Software. At the same time, organizations reduce software licensing costs as a result of both the single socket design of the Sun Fire T2000 and the excellent throughput demonstrated on an entire system comprised of Free and Open Source Software. Operational costs are also reduced through lower power, cooling and space consumption, enabling organizations to reduce their carbon footprint.
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the volume of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions. In addition to testing application server performance, it also tests the database performance of servers deployed to support the application tier via a high concurrency OLTP workload.
Comparison Table |
|
Benchmark Outcome (20)
(20) SPECjAppServer2004 2 x Sun Fire X4200 (8 cores, 4 chips) and 1 Sun Fire T2000 (8 cores, 1 chip ) 813.73 SPECjAppServer2004 JOPS@Standard". Sun Fire T2000 Server peak power consumption from the Sun Fire T2000 power calculator: http://www.sun.com/servers/coolthreads/t2000/calc/. 1 HP rx2660 (4 cores, 2 chip) and 1 HP rx2660 (4 cores, 2 chips) 874.17 JOPS@Standard". SPECjAppServer2004. HP pricing and power consumption from: http://h20341.www2.hp.com/integrity/downloads/SPECjApp_rx2660-874JOPS-v2.pdf SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 07/18/07 Oracle database Licensing, 07/17/07: http://www.oracle.com/corporate/pricing/multicore_faq.pdf
- A single Sun Fire T2000 Server configured with just a single, 8 Core UltraSPARC T1 processor and 32GB of memory running IBM DB2 V8.2 was able to support a Sun Fire E2900 Server in delivering 1,781.37 JOPS for the SPECjAppServer2004 Benchmark in the Standard category. The Sun Fire T2000 Server consumed only 321 watts of power at peak and required just two database licenses, resulting in a database Price / Performance of $46.03 per JOP.
- The Sun Fire T2000 Server required 7x fewer database licenses and 8x lower database licensing costs than the HP rx8620 Itanium 2 configuration, while consuming 8.5x less data center space and consuming 13x less power, resulting in nearly 18x higher performance per watt. Based on software licensing costs, the Sun Fire T2000 delivered 11x better price / performance.
- Compared to the HP rx2660 Itanium 2 server, the Sun Fire T2000 Server requires fewer database licenses, adding up to 3x better price/performance, while delivering over 3.5x higher performance / watt. Compared to the IBM p5+ 550 equipped with the latest 2.1GHz processors, the Sun Fire T2000 delivered nearly 3X higher performance while consuming nearly 3x less power and delivering 7x higher performance per watt in half the space. The Sun Fire T2000 server delivered nearly 5x better database price / performance than the IBM system, when running IBM's own DB2 software.
- This result demonstrates the throughput and power efficiency of the Sun Fire T2000 Server with the Solaris 10 OS running as an OLTP database server, with the added benefit of requiring fewer database licenses to reduce costs and TCO.
One Sun Fire T2000 Server running IBM DB2 database was able to support a single node Sun Fire E2900 server delivering 1781.37 SPECjAppServer2004 JOPS when tested under the SPECjAppServer2004 benchmark. This represents the highest performance of any single processor database server, along with the best performance per watt and leading database Price / Performance in class.
Compared to competitive servers, the Sun Fire T2000 Server requires up to 7x fewer database licenses with 8x lower licensing costs, while consuming 13x less power and delivering 18x higher performance per watt. Space consumption was reduced by up to 8x, and price/performance was improved by up to 11x.
The unparalleled level of efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 Server running Solaris 10 enables customers to breakthrough traditional power and space limitations in their data centers to process more transactions and serve more customers than ever before. At the same time, organizations reduce both software licensing costs as a result of the single socket design of the Sun Fire T2000, and reduce operational costs through lower power, cooling and space consumption. The lower power consumption of the Sun Fire T2000 will also enables organizations to reduce their environmental pollution.
The database tier of the SPECjAppServer2004 benchmark is representative of a high concurrency OLTP database instance, and as such demonstrates the the Sun Fire T2000 Server delivers leading database performance and price / performance with the lowest power consumption when compared to HP Itanium 2 and IBM p5-based systems.
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the volume of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions. In addition to testing application server performance, it also tests the database performance of servers deployed to support the application tier.
Comparison Table |
|
Benchmark Outcome (19)
(19)
SPECjAppServer2004 1 Sun Fire E2900 (24 cores, 12 chip) and 1 Sun Fire T2000 (8 cores, 1 chip ) 1781.37 SPECjAppServer2004 JOPS@Standard". Sun Fire T2000 Server peak power consumption from the Sun Fire T2000 power calculator: http://www.sun.com/servers/coolthreads/t2000/calc/. 1 HP rx6600 (8 cores, 4 chips) and HP rx8620 (16 cores, 16 chips) 1266.42 JOPS@Standard.. HP rx8620 power consumption estimated by applying 70% to the Maximum Theoretical Power reported 11/23/06: http://h18000.www1.hp.com/products/quickspecs/12471_div/12471_div.HTML#Technical%20Specifications.
1 HP rx2660 (4 cores, 2 chip) and 1 HP rx2660 (4 cores, 2 chips) 802.61 JOPS@Standard". SPECjAppServer2004. HP rx2620 power consumption calculated by applying 70% to the maximum input power, reported 04/10/07:
http://h18004.www1.hp.com/products/quickspecs/12698_div/12698_div.HTML#Technical%20Specifications
1 IBM p5 505Q (4 cores, 2chip) and 1 p5+ 550 (4 cores, 2chips) 613.38 JOPS@Standard. IBM 550 power specifications calculated by applying 70% of the Maximum Watts published in "Facts and Features Report", 11/14/06, posted at: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF.
SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 11/24/06
Oracle database Licensing, 04/10/07: http://www.oracle.com/corporate/pricing/multicore_faq.pdf
IBM DB2 licensing from: http://www-142.ibm.com/software/sw-lotus/services/cwepassport.nsf/wdocs/pvu_table_for_customers
IBM p505 requires 200 PVUs, and therefore 2 licenses
IBM DB2 licensing, 11/24/06 from a href="https://www112.ibm.com/software/howtobuy/buyingtools/paexpress/Express?P0=E1&part_number=D55IULL,D55IYLL,D55LVLL,D56CILL,D58N3LL,D585VLL,D5874LL,D5876LL,D58U9LL,D58V3LL,D58V1LL,D58NALL,D55J0LL&catalogLocale=en_US&Locale=en_US&country=USA&PT=html&S_TACT=none&S_CMP=none">
https://www112.ibm.com/software/howtobuy/buyingtools/paexpress/Express?P0=E1&part_number=
D55IULL,D55IYLL,D55LVLL,D56CILL,D58N3LL,D585VLL,D5874LL,D5876LL,D58U9LL,D58V3LL,D58V1LL,D58NALL,D55J0LL
&catalogLocale=en_US&Locale=en_US&country=USA&PT=html&S_TACT=none&S_CMP=none
Sun and Oracle have published the first ever dual node Oracle Application Server results on the industry standard benchmark for Java application server performance. The benchmark was conducted using an end-to-end Sun / Oracle infrastructure, comprising Sun Fire T2000 Servers in the application and database tiers running Solaris 10 OS and the Oracle Application Server 10g and Oracle Database Server 10g. As such, the result demonstrates the leading performance and power efficiency of Sun Fire CoolThreads servers, running Oracle application and database environments. (Click here for more information on Oracle Database Server performance on Sun Fire T2000)
Two Sun Fire T2000 Servers, with one running Oracle Application Server 10g and the other running Oracle Database Server 10g were able to deliver 733.22 SPECjAppServer2004 JOPS (Java Operations Per Second) when tested under the SPECjAppServer2004 benchmark. The Sun Fire servers consumed just over 300 watts of power each, demonstrating great levels of power efficiency when running demanding workloads, enabling customers to reduce operational costs through lower energy consumption and environmental pollution.
This result complements other submissions on Sun Fire CoolThreads servers from leading Java application server vendors, including Sun (Java Enterprise System Application Server), BEA WebLogic and IBM WebSphere.
The Sun Fire T2000 Server has consistently delivered world record levels of performance in class, along with the best levels of energy efficiency and SWaP (data center efficiency), demonstrating that Sun Fire CoolThreads servers are the leading platforms for deployment of secure web services.
Read Disclosure: (18)
(18) SPECjAppServer2004 1 Sun Fire T2000's (8 cores, 1 chip) and 1 Sun Fire T2000 (8 cores, 1 chip ) 733.22 SPECjAppServer2004 JOPS@Standard. Sun Fire T2000 Server power consumption from Sun Fire T2000 Power Calculator: http://www.sun.com/servers/coolthreads/t2000/calc/. SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 01/17/07
- A single Sun Fire T2000 Server configured with just a single, 8 Core UltraSPARC T1 processor and 32GB of memory running Oracle Database Server 10g was able to support a Sun Fire T2000 Server in delivering 733.22 JOPS for the SPECjAppServer2004 Benchmark in the Standard category. The Sun Fire T2000 Server consumed only 310 watts of power and required just two database licenses, resulting in a database Price / Performance of $109.11 per JOP.
- The Sun Fire T2000 Server required 8x fewer database licenses and 8x lower database licensing costs than the HP rx8620 Itanium 2 configuration, while consuming 8.5x less data center space and consuming over 13x lower power, resulting in nearly 8x higher performance per watt. The Sun Fire T2000 Server delivered nearly 5x better price / performance.
- Compared to the HP rx4640 Itanium II server, the Sun Fire T2000 Server required half the Oracle licenses and cost, while consuming 2x less data center space and over 4x less power, resulting in 5x higher performance per watt and 2.5x better price/performance.
- Compared to the IBM p550 equipped with the latest 2.1GHz p5+ processors, the Sun Fire T2000 delivered nearly 80% higher performance while consuming 2.5x less power and delivering 4.5x higher performance per watt in half of the space. The Sun Fire T2000 server delivered nearly 3.5x better database price / performance than the IBM system.
- This result demonstrates the throughput and power efficiency of the Sun Fire T2000 Server with the Solaris 10 OS running as both Oracle application and database servers, with the added benefit of requiring fewer database licenses to reduce costs and TCO.
The benchmark was conducted using an end-to-end Sun / Oracle infrastructure, comprising Sun Fire T2000 Servers in the application and database tiers running Solaris 10 OS and the Oracle Application Server 10g and Oracle Database Server 10g. As such, the result demonstrates the leading performance, price / performance and efficiency of Sun infrastructure running Oracle application and database environments.
One Sun Fire T2000 Server running Oracle Database 10g was able to support a single node T2000 running Oracle Application Server 10g in delivering 733.22 SPECjAppServer2004 JOPS (Java Operations Per Second) when tested under the SPECjAppServer2004 benchmark. This represents the highest performance of any single processor database server, along with the best performance per watt and leading database Price / Performance in class
Compared to competitive servers, the Sun Fire T2000 Server requires up to 8x fewer database licenses with 8x lower licensing costs, while consuming 13x less power and delivering 8x higher performance per watt.
The database tier of the benchmark is representative of a high concurrency OLTP database instance, and therefore further demonstrates the the Sun Fire T2000 Server delivers leading database performance and price / performance with the lowest power consumption when compared to HP Itanium 2 and IBM p5-based systems
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the amount of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions. In addition to testing application server performance, it also tests the database performance of servers deployed to support the application tier.
Comparison Table |
|
Benchmark Outcome (17)
(17)
SPECjAppServer2004 1 Sun Fire T2000's (8 cores, 1 chip) and 1 Sun Fire T2000 (8 cores, 1 chip ) 733.22
SPECjAppServer2004 JOPS@Standard". Sun Fire T2000 Server power consumption from Sun Fire T2000 Power
SPECjAppServer2004 1 Sun Fire T2000's (8 cores, 1 chip) and 1 Sun Fire T2000 (8 cores, 1 chip ) 733.22 SPECjAppServer2004 JOPS@Standard". Sun Fire T2000 Server power consumption from Sun Fire T2000 Power
Calculator: http://www.sun.com/servers/coolthreads/t2000/calc/. 1 HP rx6600 (8 cores, 4 chips) and HP rx8620 (16 cores, 16 chips) 1266.42 JOPS@Standard.. HP rx8620 power consumption estimated by applying 70% to the Maximum Theoretical Power reported 11/23/06: http://h18000.www1.hp.com/products/quickspecs/12471_div/12471_div.HTML#Technical%20Specifications.
1 HP rx3600 (4 cores, 2 chip) and 1 HP rx4640 (4 cores, 4 chips) 618.22 JOPS@Standard". SPECjAppServer2004. HP rx4640 power consumption taken from HP Enterprise Configurator 10/19/05 from http://h30099.www3.hp.com/configurator/catalog-hpintegrity.asp. System configured with Redundant Power, 4 x 1.6GHz Itanium processors, 8 x 2GB DIMMs, 0 x PCI cards and 2 x 73GB HDDs
1 IBM p5 505 (2 cores, 1chip) and 1 p5 550 (4 cores, 2chips) 404.881 JOPS@Standard. IBM 550 power specifications calculated by applying 70% of the Maximum Watts published in "Facts and Features Report", 11/14/06, posted at: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF.
SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 01/17/07
Oracle database Licensing, 10/04/06: http://www.oracle.com/corporate/pricing/multicore_faq.pdf
IBM DB2 licensing from: http://www-142.ibm.com/software/sw-lotus/services/cwepassport.nsf/wdocs/pvu_table_for_customers
IBM p505 requires 200 PVUs, and therefore 2 licenses
IBM DB2 licensing, 11/24/06 from https://www112.ibm.com/software/howtobuy/buyingtools/paexpress/Express?P0=E1
&part_number=D55IULL,D55IYLL,D55LVLL,D56CILL,D58N3LL,D585VLL,D5874LL,D5876LL,D58U9LL,D58V3LL,D58V1LL,D58NALL,D55J0LL&catalog
Locale=en_US&Locale=en_US&country=USA&PT=html&S_TACT=none&S_CMP=none
\
- A single Sun Fire T2000 Server configured with just a single, 6 Core UltraSPARC T1 processor and 8GB of memory running IBM DB2 V8.2 was able to support a Sun Fire T2000 Server in delivering 801.70 JOPS for the SPECjAppServer2004 Benchmark in the Standard category. The Sun Fire T2000 Server consumed only 260 watts of power and required just two database licenses, resulting in a database Price / Performance of $90.81 per JOP.
- The Sun Fire T2000 Server required 8x fewer database licenses and 8x lower database licensing costs than the HP rx8620 Itanium II configuration, while consuming 8.5x less data center space and consuming over 16x lower power, resulting in nearly 10x higher performance per watt.
- Compared to the HP rx4640 Itanium II server, the Sun Fire T2000 Server required half the Oracle licenses and cost, while consuming 2x less data center space and over 5x lower power consumption, resulting in nearly 6.5x higher performance per watt.
- Compared to the IBM p550 equipped with the latest 2.1GHz p5+ processors, the Sun Fire T2000 delivered nearly 2X higher performance while consuming nearly 3x less power and delivering 6x higher performance per watt. The Sun Fire T2000 server delivered 4x better database price / performance than the IBM system.
- This result demonstrates the throughput and power efficiency of the Sun Fire T2000 Server with the Solaris 10 OS running as both application and database servers, with the added benefit of requiring fewer database licenses to reduce costs and TCO.
One Sun Fire T2000 Server running IBM DB2 database was able to support a single node T2000 server delivering 801.70 SPECjAppServer2004 JOPS when tested under the SPECjAppServer2004 benchmark. This represents the highest performance of any single processor database server, along with the best performance per watt and leading database Price / Performance in class.
Compared to competitive servers, the Sun Fire T2000 Server requires up to 8x fewer database licenses with 8x lower licensing costs, while consuming 16x less power and delivering 10x higher performance per watt.
The unparalleled level of efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 Server running Solaris 10 enables customers to breakthrough traditional power and space limitations in their data centers to process more transactions and serve more customers than ever before. At the same time, organizations reduce both software licensing costs as a result of the single socket design of the Sun Fire T2000, and reduce operational costs through lower power, cooling and space consumption. The lower power consumption of the Sun Fire T2000 will also enables organizations to reduce their environmental pollution.
The database tier of the SPECjAppServer2004 benchmark is representative of a high concurrency OLTP database instance, and as such demonstrates the the Sun Fire T2000 Server delivers leading database performance and price / performance with the lowest power consumption when compared to HP Itanium 2 and IBM p5-based systems.
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the amount of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions. In addition to testing application server performance, it also tests the database performance of servers deployed to support the application tier.
Comparison Table |
|
Benchmark Outcome (16)
(16)
SPECjAppServer2004 1 Sun Fire T2000's (8 cores, 1 chip) and 1 Sun Fire T2000 (6 cores, 1 chip ) 801.70 SPECjAppServer2004 JOPS@Standard". Sun Fire T2000 Server power consumption measured over the course of the benchmark run.
1 HP rx6600 (8 cores, 4 chips) and HP rx8620 (16 cores, 16 chips) 1266.42 JOPS@Standard.. HP rx8620 power consumption estimated by applying 70% to the Maximum Theoretical Power reported 11/23/06:
http://h18000.www1.hp.com/products/quickspecs/12471_div/12471_div.HTML#Technical%20Specifications.
1 HP rx3600 (4 cores, 2 chip) and 1 HP rx4640 (4 cores, 4 chips) 618.22 JOPS@Standard". SPECjAppServer2004. HP rx4640 power consumption taken from HP Enterprise Configurator 10/19/05 from http://h30099.www3.hp.com/configurator/catalog-hpintegrity.asp. System configured with Redundant Power, 4 x 1.6GHz Itanium processors, 8 x 2GB DIMMs, 0 x PCI cards and 2 x 73GB HDDs
1 IBM p5 505 (2 cores, 1chip) and 1 p5 550 (4 cores, 2chips) 404.881
JOPS@Standard. IBM 550 power specifications calculated by applying 70%
of the Maximum Watts published in "Facts and Features Report", 11/14/06,
posted at: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF.
SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 11/24/06
Oracle database Licensing, 10/04/06: http://www.oracle.com/corporate/pricing/multicore_faq.pdf
IBM DB2 licensing from: http://www-142.ibm.com/software/sw-lotus/services/cwepassport.nsf/wdocs/pvu_table_for_customers
IBM p505 requires 200 PVUs, and therefore 2 licenses
IBM DB2 licensing, 11/24/06 from
https://www112.ibm.com/software/howtobuy/buyingtools/paexpress/Express?P0=E1
&part_number=D55IULL,D55IYLL,D55LVLL,D56CILL,D58N3LL,D585VLL,D5874LL,D5876LL,D58U9LL,D58V3LL,D58V1LL,D58NALL,D55J0LL&catalog
Locale=en_US&Locale=en_US&country=USA&PT=html&S_TACT=none&S_CMP=none
- Running BEA WebLogic 9, one Sun Fire T2000 / Sun SPARC Enterprise T2000 Application Server, equipped with an 8 Core UltraSPARC T1 processor and one Sun Fire T2000 / Sun SPARC Enterprise T2000 database server equipped with 1 x 6 Core UltraSPARC T1 processor at 1.0 GHz delivered a dual-node result of 801.70 JOPS for the SPECjAppServer2004 Benchmark in the Standard category.
- When compared to HP's rx4640 equipped with Itanium 2 processors and running RHEL, the T2000 delivers nearly 1.5X higher performance in 4x less power and half the space, delivering over 6X higher performance per watt and 12x SWaP.
- The Sun Fire T2000 / Sun SPARC Enterprise T2000 delivers nearly 1.3X higher performance in nearly 3x less power and half the space, resulting in nearly 3.5X higher performance per watt and 7x SWaP when compared to HP's rx3600 equipped with two of the latest dual core Itanium II processors and running HP-UX11i.
- Compared to IBM's p5+ p505 running AIX, the T2000 delivers nearly 2X higher performance, resulting in 2.5X higher performance per watt. and 30% higher SWaP.
- When compared to the dual node Inspur submission, which unlike all of the above results used two servers rather than one in the application tier, the Sun Fire T2000 delivers 1.6X higher performance per watt in 5x less space and 8x higher SWaP. Each of the Inspur systems were equipped with two of the latest dual core Intel 5100 Series "WoodCrest" processors and running RHEL4.
- This latest SPECjAppServer2004 result was achieved with the 1.4GHz UltraSPARC T1 processor and represents a 30% performance gain when compared with the existing results achieved with the 1.2GHz processor – demonstrating the high levels of scalability of the UltraSPARC T1 "Niagara" processor.
The Sun Fire T2000 / Sun SPARC Enterprise T2000 server running the latest Solaris 10 OS 11/06 release outperformed all other competitive servers in delivering the highest overall Performance per Watt and the best per-socket performance in Java application serving. The Sun Fire T2000 / Sun SPARC Enterprise T2000 also delivers the best level of data center efficiency as measured by the SWaP metric.
The Sun Fire T2000 / Sun SPARC Enterprise T2000 results were obtained with the BEA WebLogic application server, while the database layer was also run on a Sun Fire T2000 / Sun SPARC Enterprise T2000 Server, all running the Solaris 10 OS. These results demonstrate the end to end superiority of Sun Fire CoolThreads servers running the Solaris 10 OS for application and database serving.
The unparalleled level of efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 / Sun SPARC Enterprise T2000 Server running Solaris 10 enables customers to breakthrough traditional power and space limitations in their data centers to process more transactions and serve more customers than ever before. At the same time, organizations reduce both software licensing costs as a result of the single socket design of the Sun Fire T2000 / Sun SPARC Enterprise T2000, and reduce operational costs through lower power, cooling and space consumption. The lower power consumption of the Sun Fire T2000 / Sun SPARC Enterprise T2000 will also enables organizations to reduce their environmental pollution.
These results confirm the Sun Fire T2000 / Sun SPARC Enterprise T2000 server as the most efficient and cost-effective platform on the planet for deployment of secure application and OLTP database services, with higher performance and efficiency than competitive systems running Linux, HP-UX or AIX.
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the amount of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions.
SPECjAppServer2004 Comparison Table |
|
Benchmark Outcome (14)
(14) SPECjAppServer2004 Sun Fire T2000 / Sun SPARC Enterprise T2000 (8 cores, 1 chip) 801.70 JOPS@Standard. SPECjAppServer2004 HP rx4600 (4 cores, 4 chip) 542.18 JOPS@Standard. SPECjAppServer2004 IBM p5 505 (2 cores, 1chip) 404.88 JOPS@Standard. SPECjAppServer2004 HP rx3600 (4 cores, 2 chips) 618.22 JOPS@Standard. SPEC. SPECjAppServer2004 Inspur NF380D, 2 application tier nodes (each 4 cores, 2 chips) 1466.28 or JOPS@Standard. SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. All results from www.spec.org as of 04/17/07. HP rx4640 server specifications 10/19/05 from http://www.hp.com/products1/servers/integrity/entry_level/rx4640/ HP rx4640 power rating of 1,303 watts taken from HP Enterprise Configurator 10/19/05 from http://h30099.www3.hp.com/configurator/catalog-hpintegrity.asp. System configured with Redundant Power, 4 x 1.6GHz Itanium processors, 8 x 2GB DIMMs, 0 x PCI cards and 2 x 73GB HDDs. IBM specifications from Fact and Features report, 1/9/06: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF. IBM power is based on the reported maximum power consumption. HP rx3600 power consumption estimated by taking 70% of the maximum output power supply rating reported here on 11/14/06: http://h20341.www2.hp.com/integrity/cache/387834-0-0-225-121.html Inspur specifications from http://www.langchao.com/products/channel_server/nl_3596_2.shtml Sun Fire T2000 / Sun SPARC Enterprise T2000 server power consumption taken from measurements made during the benchmark run.
- The Sun Fire T2000 server, equipped with the UltraSPARC T1 processor with 32 concurrent threads of execution achieved best per-socket performance and performance/watt in its class using the Lotus Domino 7 Messaging server running the HTTP based R6iNotes benchmark.
- This world record benchmark result clearly demonstrates that the Sun Fire T2000 based on the unique 8-core architecture of the UltraSPARC T1 processor running the Solaris 10 OS and Lotus Domino can support thousands of mail server users with the best levels of energy and data center efficiency.
- The Sun Fire T2000 delivers 22% more performance with 25% faster response times than the latest HP DL380G5 based on 2 x 3GHz dual core Intel 5160 (WoodCrest) processors running Microsoft Windows Server 2003. This translates to 2.4x higher performance per socket. The Sun Fire T2000 Server was able to deliver this performance while consuming nearly 3X less power, adding up to 3.4x higher Performance per Watt and SWaP. While the HP system achieved lower price/performance, for many customers, this would be offset by the higher energy consumption of the system.
- When compared to IBM's latest p5+ 550Q server, based on 2 x Quad Core p5+ processors and the AIX operating system, the Sun Fire T2000 was able to deliver comparable performance with 35% faster response times, and 2x higher Performance per Socket. Also, the Sun Fire T2000 Server consumed 2.5x less power, and only half the space of the IBM system, adding up to 5X higher SWaP. In addition to the power, cooling and space savings, the Sun Fire T2000 also delivered 1/3rd superior price / performance
The Sun Fire T2000 server running the latest Solaris 10 OS 11/06 release outperformed all other comparable competitive servers and processor architectures in delivering the highest overall Performance per Watt and the best per-socket performance. The Sun Fire T2000 also delivers the best level of data center efficiency as measured by the SWaP metric.
The unparalleled level of efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 Server running Solaris 10 and Lotus Domino enables organizations to breakthrough traditional power and space limitations in their data centers in delivering eMail services to more customers than ever before. At the same time, organizations reduce both software licensing costs as a result of the single socket design of the Sun Fire T2000, and reduce operational costs through lower power, cooling and space consumption. The lower power consumption of the Sun Fire T2000 will also enables organizations to reduce their environmental pollution.
These results confirm the Sun Fire T2000 server as the most efficient platform on the planet for deployment of secure mail servers, with higher per socket performance and efficiency than competitive systems running Windows on Intel Xeon or AIX on p5+ platforms
Benchmark Description
This benchmark simulates active users accessing their Domino R6iNotes mailfiles via a standard Web browser. The R6iNotes workload is HTTP-based and uses the Lotus Domino Mail server.
Lotus iNotes Comparison Table |
|
Benchmark Outcome (13)
(13)
NotesBench R6iNotes Sun Fire T2000 (1x1400 MHz UltraSPARC T1, 64GB), 4 partitions, Solaris [TM] 10, Lotus[R] Domino 7.0.2, 23,200 users, $4.48 per user, 19,518 NotesMark tpm, 692 ms avg NotesBench R6iNotes.
NotesBench R6iNotes HP DL380G5 (2 x 3GHz Intel Xeon 5160, 6GB), 3 partitions, Microsoft Windows Server 2003, Lotus[R] Domino 7.0.1, 19,000 users, $3.15 per user, 16,244 NotesMark tpm, 868 ms avg NotesBench R6iNotes.
NotesBench R6iNotes IBM p5 550Q (2 x 4 core 1.5GHz Power 5+, 6GB), 4 partitions, AIX 5L V5.3, Lotus[R] Domino 7.0, 24,000 users, $5.97 per user, 20,108 NotesMark tpm, 932 ms avg NotesBench R6iNotes.
More info www.notesbench.org
HP DL380G5 power consumption estimated by taking 70% of the maximum input wattage rating of 1193W reported here on 11/14/06:
http://h18004.www1.hp.com/products/quickspecs/12477_na/12477_na.html#Power%20Specifications
IBM p5 550Q power specifications calculated by applying 70% of the Maximum Watts published in "Facts and Features Report", 11/14/06, posted at ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF
Sun Fire T2000 server power consumption taken from measurements made during the benchmark run.
- light read-only query
- medium read-only query
- 'heavy' read and insert operation
- The Sun Fire T2000 Server was configured with the 1.2GHz UltraSPARC T1 processor and 32GB of memory, and was running the Solaris 10 OS.
- The Dell PowerEdge 6850 was configured with 4 x 3.16GHz Xeon EM64T processors and 32GB memory, and running the 64 bit SuSE Linux Enterprise Server 9 OS.
- The IBM p5+ 550 was configured with 4 x 1.9GHz p5+ processors processors and 32GB memory, and running AIX 5.3 OS
- The Sun Fire T2000 CoolThreads Server achieved a peak of 162,000 transactions per minute, with an average power consumption of 329 watts. The total configuration delivered a price/performance of $0.35 TPM.
- The Dell PowerEdge 6850 Server achieved a peak of 44,000 transactions per minute, with an average power consumption of 510 watts. The total configuration delivered a price/performance of $1.81 TPM.
- The IBM p5+ 550 Server achieved a peak of 117,533 transactions per minute with an average power consumption of 710 watts. The total configuration delivered a price/performance of $0.84 TPM.
- 3.7X more transaction per minute
- 1.6X lower power consumption
- 2X smaller footprint (RU)
- 11.4X higher server efficiency (SWaP)
- 5.2X better price / performance
- 1.4X more transaction per minute
- 2.2X lower power consumption
- 2X smaller footprint (RU)
- 5.9X higher server efficiency (SWaP)
- 2.4X better price / performance
The Sun Fire T2000 server running Solaris 10 OS beats both the 4-way Dell PowerEdge 6850 server running Linux and the IBM p5+ 550 server running AIX in a head-to-head test of OLTP database performance.
Benchmark Description
The iGEN-OLTP benchmark is a stress and performance test, measuring the throughput and number of simultaneous user connections of an OLTP database workload. The iGEN-OLTP workload is based on real-world customer applications and is constructed as a 2-tier "orders" database application where three transaction types are executed:
The transactions are comprised of various SQL statements: read-only selects, joins, update and insert operations.
iGen OLTP avoids problems that plague other OTLP benchmarks like TPC-C. TPC-C has problems with only using light-weight queries, allowing artificial data partitioning, and only testing a few database functions. The iGen transactions take almost twice the computation work compared to the TPC-C transactions.
These iGEN-OLTP tests were based on the 64 bit Oracle 10g R2 database and JDBC V1.5 client software
iGEN-OLTP, Sun Fire T2000 |
|
Benchmark Outcome (12)
Based on these results above, the Sun Fire T2000 Server delivered the following advantages when compared to the 4-socket Dell PowerEdge 6850 Xeon-based server
Solaris is well know for stable performance at high levels of utilization. In these tests Linux running on a variety of platforms often exhibits stability issues.
Compared to the IBM p5+ 550, the Sun Fire T2000 Server delivered the following advantages:
These results demonstrate the world-beating scalability, performance, price/performance and operational efficiency of the Sun Fire CoolThreads servers and the Solaris 10 OS when running intensive, multi-thousand user database workloads.
(12) iGEN-OLTP tests were based on the 64 bit Oracle 10g R2 database and JDBC V1.5 client software. The Sun Fire T2000 Server was configured with the 1.2GHz UltraSPARC T1 processor and 32GB of memory, and was running the Solaris 10 OS. The Dell PowerEdge 6850 was configured with 4 x 3.16GHz Xeon EM64T processors and 32GB memory, and running the 64 bit SuSE Linux Enterprise Server 9 OS (2.6.5-7.216-smp) The IBM p5+ 550 was configured with 4 x 1.9GHz p5+ processors processors and 32GB memory, and running AIX 5.3 OS The Sun Fire T2000 server configuration was priced at $56,995, comprised $26,995 for the T2000, $30,000 for Oracle licenses. The Dell PowerEdge 6850 server configuration was priced at $79,752, comprised $19,752 for the 6850, $60,000 for Oracle licenses. The IBM p5+ 550 server configuration was priced at $98,340, comprised $52,740 for the server, $600 for OS and $45,000 for Oracle licenses. Pricing from sun.com and dell.com on 3/8/06. IBM pricing from cost of test system acquired by Sun
- The Sun Fire T2000 server, equipped with the UltraSPARC T1 processor with 32 concurrent threads of execution achieved best per-socket performance and performance/watt in its class using the Lotus Domino 7 Messaging server running the HTTP based R6iNotes benchmark.
- This world record benchmark result clearly demonstrates that the Sun Fire T2000 based on the unique 8-core architecture of the UltraSPARC T1 processor running the Solaris 10 OS and Lotus Domino can support thousands of mail server users with the best levels of energy and data center efficiency.
- The Sun Fire T2000 delivers 22% more performance with 25% faster response times than the latest HP DL380G5 based on 2 x 3GHz dual core Intel 5160 (WoodCrest) processors running Microsoft Windows Server 2003. This translates to 2.4x higher performance per socket. The Sun Fire T2000 Server was able to deliver this performance while consuming nearly 3X less power, adding up to 3.4x higher Performance per Watt and SWaP. While the HP system achieved lower price/performance, for many customers, this would be offset by the higher energy consumption of the system.
- When compared to IBM's latest p5+ 550Q server, based on 2 x Quad Core p5+ processors and the AIX operating system, the Sun Fire T2000 was able to deliver comparable performance with 35% faster response times, and 2x higher Performance per Socket. Also, the Sun Fire T2000 Server consumed 2.5x less power, and only half the space of the IBM system, adding up to 5X higher SWaP. In addition to the power, cooling and space savings, the Sun Fire T2000 also delivered 1/3rd superior price / performance
The Sun Fire T2000 server running the latest Solaris 10 OS 11/06 release outperformed all other comparable competitive servers and processor architectures in delivering the highest overall Performance per Watt and the best per-socket performance. The Sun Fire T2000 also delivers the best level of data center efficiency as measured by the SWaP metric.
The unparalleled level of efficiency of the UltraSPARC T1 "Niagara" processor and the Sun Fire T2000 Server running Solaris 10 and Lotus Domino enables organizations to breakthrough traditional power and space limitations in their data centers in delivering eMail services to more customers than ever before. At the same time, organizations reduce both software licensing costs as a result of the single socket design of the Sun Fire T2000, and reduce operational costs through lower power, cooling and space consumption. The lower power consumption of the Sun Fire T2000 will also enables organizations to reduce their environmental pollution.
These results confirm the Sun Fire T2000 server as the most efficient platform on the planet for deployment of secure mail servers, with higher per socket performance and efficiency than competitive systems running Windows on Intel Xeon or AIX on p5+ platforms
Benchmark Description
This benchmark simulates active users accessing their Domino R6iNotes mailfiles via a standard Web browser. The R6iNotes workload is HTTP-based and uses the Lotus Domino Mail server.
Lotus iNotes Comparison Table |
|
Benchmark Outcome (11)
(11)
NotesBench R6iNotes Sun Fire T2000 (1x1400 MHz UltraSPARC T1, 64GB), 4 partitions, Solaris [TM] 10, Lotus[R] Domino 7.0.2, 23,200 users, $4.48 per user, 19,518 NotesMark tpm, 692 ms avg NotesBench R6iNotes.
NotesBench R6iNotes HP DL380G5 (2 x 3GHz Intel Xeon 5160, 6GB), 3 partitions, Microsoft Windows Server 2003, Lotus[R] Domino 7.0.1, 19,000 users, $3.15 per user, 16,244 NotesMark tpm, 868 ms avg NotesBench R6iNotes.
NotesBench R6iNotes IBM p5 550Q (2 x 4 core 1.5GHz Power 5+, 6GB), 4 partitions, AIX 5L V5.3, Lotus[R] Domino 7.0, 24,000 users, $5.97 per user, 20,108 NotesMark tpm, 932 ms avg NotesBench R6iNotes.
More info www.notesbench.org
HP DL380G5 power consumption estimated by taking 70% of the maximum input wattage rating of 1193W reported here on 11/14/06:
http://h18004.www1.hp.com/products/quickspecs/12477_na/12477_na.html#Power%20Specifications
IBM p5 550Q power specifications calculated by applying 70% of the Maximum Watts published in "Facts and Features Report", 11/14/06, posted at ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF
Sun Fire T2000 server power consumption taken from measurements made during the benchmark run.
- Vertical scalability-The Siebel CRM Release 7.7 server showed excellent scalability within an application server.
- Horizontal scalability-The benchmark demonstrated scalability across multiple servers without degradation.
- Low network utilization-The Siebel CRM Release 7.7 Smart Web Architecture and Smart Network Architecture efficiently managed the network, consuming only 2.615 Kbps per user.
- Efficient use of the database server-Siebel CRM Release 7.7 Smart Database Connection Pooling and Multiplexing allowed the database to service 12,500 concurrent users and the supporting Siebel Release 7.7 server application services.
Two white papers from Oracle have been published to describe the performance and scalability capabilities of Oracle's Siebel Customer Relationship Management (CRM) Release 7.7. The benchmarks comprised 8,000 and 12,500 concurrent users running Siebel CRM Release 7.7 industry applications on Sun Fire T2000 and UltraSPARC IIIi & IV+ Servers and Oracle9i Database.
The test systems demonstrated that Siebel CRM Release 7.7 architecture on Sun servers easily scales to 12,500 concurrent users.
- Running BEA WebLogic 9, one Sun Fire T2000 Application Server, equipped with the UltraSPARC T1 processor and one Sun Fire V490 database server equipped with 4 UltraSPARC IV+ processors at 1.5 GHz delivered a dual-node result of 615.64 JOPS for the SPECjAppServer2004 Benchmark in the Standard category.
- This result beats the HP rx4640 dual node result of 471.28 JOPS@Standard by 31% using the same application server and database software while consuming over 4x less power and half the space, delivering over 5X better performance per watt. Based on the SWaP metric, the Sun Fire T2000 delivers nearly 11X higher server efficiency than the HP rx4640.
- Running IBM WebSphere 6.1, one Sun Fire T2000 Application Server, equipped with the UltraSPARC T1 processor and one Sun Fire X4200 database server equipped with 2 x Dual Core AMD Opteron processors at 2.2 GHz delivered a world record dual-node result of 616.22 JOPS for the SPECjAppServer2004 Benchmark in the Standard category.
- This result beats the IBM p505 dual node result of 349.11 JOPS@Standard by 1.8x using the same application server and database software while consuming 1.3x less power and delivering 2.2x higher performance per watt. Despite the IBM p5 505 being half the height of the Sun Fire T2000 (1RU vs 2RU), the Sun server still delivers 10% higher SWaP as result of higher performance and power efficiency
- This result is equivalent to that achieved by the HP rx3600 server, equipped with two of the the latest dual core Itanium II processors. The Sun Fire T2000 server achieved performance parity but only required half the space and less than half the power of the HP rx3600, therefore delivering over 2X higher performance per watt. Based on the SWaP metric, the Sun Fire T2000 delivers over 4X higher server efficiency than the HP rx3600.
These results were achieved using the same application server and database software - This benchmark result clearly demonstrates that the Sun Fire T2000 running the Solaris 10 Operating system can support thousands of concurrent users accessing Web Services applications.
- This result highlights the performance benefits of the BEA Weblogic Server and IBM WebSphere Application Servers on Sun Fire servers with CoolThreads technology.
- This result also demonstrates the superior performance of Oracle 10g and IBM DB2 on the Sun Fire servers equipped with SPARC or Opteron processors and the Solaris 10 OS.
Benchmark Description
SPECjAppServer2004 is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the amount of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions.
The Sun Fire T2000 server running Solaris 10 OS beat both Itanium II based servers from HP and p5 based servers from IBM in a head-to-head test of application server performance, while delivering significantly higher levels of efficiency as demonstrated by superior performance per watt and SWaP ratings.
The Sun Fire T2000 results were obtained with both BEA WebLogic and IBM WebSphere application servers, while the database layer was run on a mixture of UltraSPARC IV+ and Opteron based Sun Fire Servers, all running the Solaris OS. These results demonstrate the end to end superiority of Sun servers running the Solaris 10 OS.
SPECjappServer 2004 2-Node Results |
|
Benchmark Outcome (8)
(8) SPECjAppServer2004 Sun Fire T2000 (8 cores, 1 chip) 615.64 JOPS@Standard and 616.22 JOPS@Standard. SPECjAppServer2004 HP rx4600 (4 cores, 4 chip) 471.28 JOPS@Standard. SPECjAppServer2004 IBM p5 505 (2 cores, 1chip) 349.11 JOPS@Standard. SPECjAppServer2004 HP rx3600 (4 cores, 2 chips) 618.22 JOPS@Standard. SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. All results from www.spec.org as of 08/31/06. HP rx4640 server specifications 10/19/05 from http://www.hp.com/products1/servers/integrity/entry_level/rx4640/ HP rx4640 power rating of 1,303 watts taken from HP Enterprise Configurator 10/19/05 from http://h30099.www3.hp.com/configurator/catalog-hpintegrity.asp. System configured with Redundant Power, 4 x 1.6GHz Itanium processors, 8 x 2GB DIMMs, 0 x PCI cards and 2 x 73GB HDDs. IBM specifications from Fact and Features report, 1/9/06: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF. IBM power is based on the reported maximum power consumption. HP rx3600 specifications taken 09/26/06 from http://h18000.www1.hp.com/products/quickspecs/12595_div/12595_div.HTML#Overview HP rx3600 power based on reported Typical Power Dissipation Sun Fire T2000 server power consumption taken from measurements made during the benchmark run.
- Java[tm] Platform, Standard Edition 5.0_08 with Java HotSpot[tm] Virtual Machine technology extends Sun's best in class Java server performance with new SPECjbb2005 World records running the open-source Ultra SPARC T1 processor, demonstrating performance gains exceeding 17% over previous Sun World Record publications using Java SE 5.0_06.
- A Sun Fire T2000 Server set a new world record by delivering 74,365 SPECjbb2005 bops for the SPECjbb2005 Benchmark, beating all other 2U servers.
- The Sun Fire T2000 server beat IBMs 4-way x346 server configured with 2 x 2.8GHz Intel Xeon dual core processors by nearly 2X, while consuming nearly 1.5X less power and delivering 3X higher performance per watt.
- The Sun Fire T2000 server beat IBMs new 4-way, 4 core p510Q server configured with 1.5GHz Power 5+ processors by over 1.3X, while consuming nearly 1.5X less power and delivering 2X better performance per watt.
- The Sun Fire T2000 server beat IBMs 4-way, 4 core p550 server configured with 1.9GHz Power 5+ processors by 1.2X, while consuming 2.5X less power in half the space and delivering 3X higher performance per watt.
- Based on the SWaP metric, the Sun Fire T2000 delivers nearly 3X higher efficiency than the IBM x346 and 6X higher server efficiency than the IBM p550.
The Sun Fire T2000 server running Solaris 10 OS extended its world record leadership on the SPECjbb2005 Java benchmark by beating 2 processor, 4-Core Intel Xeon / Windows servers and P5+ servers from IBM in a head-to-head test of Java Virtual Machine performance. In addition to performance, the Sun Fire T2000 server also excelled in performance per watt and SWaP.
These results confirm the Sun Fire CoolThreads servers as the most efficient and cost-effective platforms on the planet for deployment of JVM components, providing the foundation for the delivery of highly scalable and secure web services.
Benchmark Description
SPECjbb2005 (Java Business Benchmark) measures the performance of a Java implemented application tier (server-side Java). The benchmark is based on the order processing in a wholesale supplier application. The metrics given are number of SPECjbb2005 bops (Business Operations per Second) and SPECjbb2005 bops/JVM (bops per JVM instance).
Comparison Table |
|
Benchmark Outcome (7)
(7) Sun Fire T2000 Server (1 chip, 8 cores, 1-way) 74,365 SPECjbb2005 bops, 18,591 SPECjbb2005 bops/JVM
IBM x346 (2 chip, 4 cores, 4-way) 39,585 SPECjbb2005 bops, 39,585 SPECjbb2005 bops/JVM IBM p510Q (2 chips, 4 cores) 54,785 SPECjbb2005 bops, 54,785 SPECjbb2005 bops/JVM IBM p550 (2 chip, 4 cores, 4-way) 61,789 SPECjbb2005 bops, 61,789 SPECjbb2005 bops/JVM
IBM p510Q specifications, 19/04/06, from http://www-03.ibm.com/systems/p/hardware/entry/510q/index.html
IBM p550 specifications, 10/06/05, from http://www-03.ibm.com/servers/eserver/pseries/hardware/entry/550.html
IBM p5 power specifications calculated by applying 70% of the power numbers published in "Facts and Features Report", 3/10/06, posted at http://www-03.ibm.com/servers/eserver/pseries/hardware/factsfeatures.html
Sun Fire T2000 server power consumption taken from measurements made during the benchmark run.
Benchmark Description
SPECjAppServer2004 Multi-Node provides a great demonstration of the performance a cluster of systems can produce in Java environment powering critical applications and web based services.
It is the only industry-standard benchmark used for Java Enterprise Edition application servers. It measures the amount of transactions a particular system configuration (hardware and software) can deliver while maintaining acceptable response times for 90% of the transactions.
SPECjAppServer2004, Sun Fire T2000 |
|