NEW 1.4GHz Sun Fire T2000 World Record 1-socket SPECweb2005 & World Record Perf/Watt
Tuesday Jan 09, 2007
The Sun Fire T2000 obtained a world record single-socket 16,407 SPECweb2005 with one UltraSPARC T1 and 64 GB memory running Solaris 10 11/06 with Sun Java[TM] System Web Server 6.1 SP5 64-bit web server.
The Sun Fire T2000 had an average power consumption of 340 watts for all benchmark workloads at steady-state.
Secured web server is critical for most of today's datacenters, unfortunately other vendors have a big performance and cost penalty for security. The Sun Fire T2000 has a unique hardware design and coupled with Solaris 10's software features secured web services only adds 10%. Now users can use SSL-enabled web service by default. This is a huge advantage for Sun customers.
Various Comparisons
- The Sun Fire T2000 delivers 24% faster on SPECweb2005 than the 4-core HP DL380G5 with 3.0GHz Intel Xeon 5160 (WoodCrest), while consuming 2.5X less power, and up to 3x higher SWaP.
- The Sun Fire T2000 is 13% faster on SPECweb2005 performance than the 4-core Dell PowerEdge 2950 with 3.0GHz Intel Xeon 5160 (WoodCrest).
- The Sun Fire T2000 is 10% faster on SPECweb2005 than the 8-core Fujitsu Siemens PRIMERGY RX600 S3 with Intel Xeon 7140M.
- The Sun Fire T2000 is 10% faster on SPECweb2005 than the 8-core HP ProLiant DL585 G2 with 2.0 GHz AMD Opteron 8212.
- The Sun Fire T2000 is 4% faster on SPECweb2005 than the 8-core HP Proliant DL585 with 2.6GHz AMD Opteron 885.
- The Dell 2950 equipped with 2 of the latest Quad-core Intel Xeon 5300 Series “Clovertown” running Red Hat Enterprise Linux delivers only 3% higher throughput than the single socket Sun Fire T2000, but consumes nearly 55% more power. As a result the T2000 delivers 1.5X better Performance per Watt and SWaP
- The Sun Fire T2000 is 208% faster on SPECweb2005 than the 4-core IBM p5 550 with 1.9GHz POWER5+.
The Sun Java[TM] System Web Server 6.1 Service Pack 5 for 64-bit Solaris 10 on SPARC hardware is optimized to leverage the 64-bit processing environments.
This world record benchmark result clearly demonstrates that the Sun Fire T2000 running the Solaris 10 OS and Java System Webserver 6.1 SP5 can support thousands of concurrent web server sessions while allowing larger and more complex Java applications to be run.
SPECweb2005 Benchmark Performance as of 01/09/2007. all results at http://www.spec.org website.
| System | CPU MHz |
OS | Web Server |
SPEC web2005 |
Bank/Ecom/Supp | Watts |
|---|---|---|---|---|---|---|
| HP/Proliant
DL585 G2 |
AMD Opteron
8220 @ 2.8GHz (4 chips, 2 cores/chip) |
RH Enterprise Linux 4 Update 4 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 20235 |
36400/28000/18016 | |
| Dell/PowerEdge
2950 |
Intel Xeon
X5355 (Clovertown) @ 2.66GHz (2 Chips, 4 cores/chip) |
RH Enterprise Linux 4 AS Update 3 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 16830 |
33250/20500/15500 | 525 |
| Sun/Sun Fire T2000 | UltraSPARC
T1 @ 1.4 GHz (1 Chip, 8 Cores/Chip, 4 threads/core) |
Solaris 10 11/06 | Sun
Java [TM] System Web Server 6.1 SP5 64-bit |
16407 |
25812/24048/15768 | 340 |
| HP/Proliant DL585 | AMD Opteron
885 @ 2.6GHz (4 chips, 2 cores/chip) |
RH Enterprise Linux 4 Update 4 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 15850 |
32992/25024/10688 | |
| HP/Proliant DL585 G2 | AMD Opteron
8212 @ 2.0GHz (4 chips, 2 cores/chip) |
RH Enterprise Linux 4 Update 4 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 14964 |
27200/20960/13024 | 812 |
| Fujitsu/PRIMERGY RX600 S3 | Intel Xeon
7140M @ 3.4GHz (4 chips, 2 cores/chip) |
RH Enterprise Linux 4 Update 3 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 14896 |
25260/21480/13500 | |
| Dell/PowerEdge 2950 | Intel Xeon
5160 (Woodcrest) @ 3.0GHz (2 Chips, 2 cores/chip) |
RH Enterprise Linux 4 Update 3 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 14495 |
23800/20400/13900 | |
| Sun/Sun
Fire T2000 |
UltraSPARC
T1 @ 1.2 GHz (1 Chip, 8 Cores/Chip, 4 threads/core) |
Solaris
10 3/05 HW2 plus patches |
Sun
Java [TM] System Web Server 6.1 SP5 64-bit |
14001 | 21500/21500/13160 | 339 |
| HP/Proliant DL380 G5 | Intel Xeon
5160 (Woodcrest) @ 3.0GHz (2 Chips, 2 cores/chip) |
RH Enterprise Linux 4 Update 4 | Accoria Rock Web Server v1.4.0 (x86_64), Accoria Rock JSP/Servlet Container v1.2.0 (x86_64) | 13257 |
23808/20400/10632 | 835 |
| Sun/Sun Fire T1000 | UltraSPARC
T1 @ 1.0 GHz (1 Chip, 8 Cores/Chip, 4 threads/core) |
Solaris
10 6/06 |
Sun
Java [TM] System Web Server 6.1 SP5 64-bit |
10466 | 20000/16500/7700 | 178 |
| IBM/IBM System p5 550 |
POWER5+
@
1.9 GHz (2 Chips, 2 Cores/Chip w SMT) |
SLES 9 SP2 | Zeus
4.3r1 (64 bit), Apache Tomcat 5.5.9 plus Compat 5.5.9 |
7881 |
12240/11820/7500 | 770 |
The Sun Fire T2000 1.4 results currently under SPEC review.
Sun Fire T2000/T1000 server power consumption taken from measurements made during the benchmark run. Power is average measured watts during benchmark run.
IBM 550 power specifications calculated by applying 70% of the Maximum Wattss published in “Facts and Features Report”, 11/14/06, posted at ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF.
Dell 2950 power rating estimated by calculating 70% of the power supply data reported in the product brochure,http://www.dell.com/downloads/global/products/pedge/en/2950_specs.pdf
HP DL585 power consumption estimated by taking 70% of the maximum input wattage rating of 1160W reported here on 11/14/06: http://h18004.www1.hp.com/products/quickspecs/11902_na/11902_na.html#Power%20Specifications
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
Benchmark Description
SPECweb2005, is the latest
industry standard benchmark for evaluating Web Server performance
developed by SPEC. The benchmark simulates multiple user sessions
accessing a Web Server and generating static and dynamic HTTP
requests. The major features of SPECweb2005 are:
- Measures simultaneous user sessions
- Dynamic content: currently PHP and JSP implementations
- Page images requested using 2 parallel HTTP connections
- Multiple, standardized workloads: Banking (HTTPS), E-commerce (HTTP and HTTPS), and Support (HTTP)
- Simulates browser caching effects
- File accesses more accurately simulate today's disk access
patterns
Disclosure Statement:
SPEC, SPECweb reg tm of Standard Performance Evaluation Corporation. Sun Fire T2000 (8 cores/1 chip) 16407 SPECweb2005, submitted to SPEC for review. HP Proliant DL585 G2 (2 cores/4 chips) 20235 SPECweb2005. Dell PowerEdge 2950 (4 cores/2 chips) 16830 SPECweb2005. HP Proliant DL585 (2 cores/4 chips) 15850 SPECweb2005. HP Proliant DL585 G2 (2 cores/4 chips) 14964 SPECweb2005. Fujitsu PRIMERGY RX600 S3 (2 cores/4 chip) 14896 SPECweb2005. Dell PowerEdge 2950 (2 cores/2 chip) 14495 SPECweb2005. Sun Fire T2000 (8 cores/1 chip) 14001 SPECweb2005. HP Proliant DL380 G5 (2 cores/2 chips) 13257 SPECweb2005. Sun Fire T1000 (8 cores/1 chip) 10466 SPECweb2005. IBM System p5 550 (2 cores/2 chips 7881 SPECweb2005. Results from www.spec.org as of January 9, 2007.
| Certified Results | 16,407 SPECweb2005 |
||
| Reference Date: | January 9, 2007 | ||
| Systems: | 1 x Sun Fire T2000 |
||
| Total # Processors: | 1 chip / 8 cores/chip (4 threads/core) |
||
| GHz Processor: | Sun UltraSPARC T1 1.4 GHz | ||
| Operating System: | Solaris 10 11/06 |
||
| Software: | Sun Java[TM] System Web Server 6.1 SP5 64-bit | ||
| Storage/network: | 3 internal 73GB 10K SAS drives 1 PCI-X Sun PCI Dual Fibre Channel 2GB HBA, 1 Sun StorEdge 3510 RAID, 2 Sun StorEdge 3510 JBOD, 3 Sun PCI-E Dual GigE UTP adapters | ||
|
|
|
|
