 |
|
Lonely at the Top with SAP's BI Data Mart
When I hike, I love to get to the top of the mountain. Then it's time to break out the peanut butter and jelly sandwiches and enjoy the view. But I do admit that I feel a bit lonely when there's nobody else up there. No one to say hi to. No one to talk about where you are from. No one to ask to take your picture.
And that's how it's been with SAP's BI Data Mart Benchmark. Earlier this week, IBM published an IBM System i 520 SAP BI Data Mart Benchmark result that demonstrates excellent performance and scalability.(1) This result complements the other outstanding System i POWER5+ and POWER6 results on this benchmark.
It would be nice to see others on the mountain, even if it's only to take our picture.
******************************************************
(1) IBM System i 520 SAP BI Data Mart standard NetWeaver 7.0 (2004s) application benchmark result on the 2-core (2 processor, 2 cores, 4 threads) 1.9 GHz POWER5+ IBM System i 520 running SAP NetWeaver 7.0 (2004s), DB2 for i5/OS V5R4 on i5/OS V5R4. With 97% utilization, the i5-520 achieved a throughput/hour of 26,224 query navigation steps (cert # 2007061).
Source: http://www.sap.com/solutions/benchmark/; Results as of 10/26/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
SAP, mySAP and other SAP product and service names mentioned herein as well as their respective
logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all
over the world.
technorati tags: IBM, systems, performance, SAP,DataMart, NetWeaver,i5, POWER6,benchmarking
Categories
: [ BI | SAP | i5 ]
Oct 26 2007, 09:57:35 AM EDT
Permalink
|
|
Who Really Are Those Bloggers Out There ?
Something's really been bothering me lately. And no, it's not that the Indians lost.
There are bloggers out there who claim to be experts. They talk as if they truly know everything about everything, including systems and benchmarks.
The only problem is that we don't really know who they are.
The BM Seer has long claimed to know the score on benchmarks. But he's an Anonymous Sun Source.
If he were a real expert on benchmarking, he would know that you need to analyze everything in a benchmark comparison. He would also know that people still like TPC-C.
There's a blogger named Murphy. Paul Murphy. An IT Consultant speciliazing in Unix. The only problem is that's not his real name. He does say he has some Sun systems at home and at work. Maybe his real name is Scott. Or Jonathan. If he were a real expert on systems he would certainly know that it just doesn't make sense to use prices from benchmark reports for a pricing comparison of real systems.
Boy have these guys got a lot to learn. Including using their given names.
************************************************
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
TPC Benchmarks are registered trademarks of the Transaction Processing Performance Council.
technorati tags: IBM, systems, performance, TPC-C, benchmark,Sun,BMSeer,Murphy,blog,benchmarking
Categories
: [ BMSeer | Murphy ]
Oct 24 2007, 10:21:47 AM EDT
Permalink
|
|
System x Marks the Spot
IBM System x servers are downright superhuman these days with IBM's exclusive fourth-generation X-Architecture.
Last week the IBM System x3850 M2 achieved more than half a million tpmC in the TPC-C benchmark, surpassing HP by over 26%. (1)
Yesterday the new System x3950 M2 achieved the highest ever TPC-C result on the x86 platform. (2)
The x3850 M2 also just delivered outstanding performance on TPC-E, the next-generation OLTP benchmark.(3)
Like the X-Men, System x demonstrated this week why it's part of the IBM superhero benchmark team.
************************************************
(1) IBM System x3850 M2 with the Quad-Core Intel Xeon Processor X7350 2.93GHz (4 processors/16 cores/16 threads), 516,752 tpmC, $2.59 USD / tpmC, availability of March 14, 2008 vs. HP ProLiant DL580G5 with the Quad-Core Intel Xeon Processor X7350 2.93GHz (4 processors/16 cores/16 threads), 407,079 tpmC, $1.71 USD / tpmC, availability of September 5, 2007.
(2) IBM System x3950 M2 with the Quad-Core Intel Xeon Processor X7350 2.93GHz (8 processors/32 cores/32 threads), 841,809 tpmC, $3.46 USD / tpmC, total solution availability of April 1, 2008.
(3) IBM System x3850 M2 with the Quad-Core Intel Xeon Processor X7350 2.93GHz (4 processors/16 cores/16 threads), 419.80 tpsE, $1,527.25 USD / tpsE, total solution availability of December 7, 2007.
Source: http://www.tpc.org; Results current as of 10/23/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
TPC Benchmarks are registered trademarks of the Transaction Processing Performance Council.
technorati tags: IBM, systems, performance, TPC-C, benchmark,HP,x3950,x3850,TPC-E,benchmarking
Categories
: [ HP | TPC-C | TPC-E | x3850 | x3950 ]
Oct 23 2007, 10:17:09 AM EDT
Permalink
|
|
Thumper Does TPC-H
"Thumper" is the cute rabbit in "Bambi," a character in a James Bond film, and Hopper's pet grasshopper in "A Bug's Life." I also understand that it is a name for a motorcycle and a popular drinking game. What it does not sound like is the name of a serious data server.
A few days ago Sun published X4500 (named, yes it's really true, Thumper) results on the TPC-H Business Intelligence benchmark. In the 3TB benchmark, the Sun result comes in at a stunning #10.(1) The 1TB result on a very small configuration was far, far behind the leader. (2)
Maybe a good codename next time should be something more like Simba.
************************************************
(1) Sun TPC-H 3000GB result on the Sun Fire X4500 of 38,672.4 QphH@3000GB ($29.39/QphH, avail. 10/12/2007). The result was obtained on a cluster of 10 Sun Fire X4500 servers for a total of 20 dual-core 2.6 GHz AMD Opteron 285 processors (20 processor chips, 40 cores, 40 threads) with 100 GB of memory and 218 TB of SATA disk drives running Solaris 10.
(2)Sun TPC-H 1 TB scale factor result of 5604.9 QphH@1000GB ($8.11/QphH, Avail. 10/15/2007) on the Sun Fire X4500 (2 Opteron 285 2.8 GHz chips, 4 cores, 4 threads) running Solaris 10.
Source: http://www.tpc.org; Results current as of 10/18/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
TPC Benchmarks are registered trademarks of the Transaction Processing Performance Council.
technorati tags: IBM, systems, performance, Thumper, benchmark,Sun,X4500,BI,TPC-H,benchmarking
Categories
: [ Sun | TPC-H | X4500 ]
Oct 18 2007, 01:32:32 PM EDT
Permalink
|
|
A POWER6 DB2 Home Run
I'm not even a baseball fan. Where I grew up, football was everything and the Washington Redskins were my team. Maybe because we had no baseball team.
But it's always nice to embrace a winner and since I happen to live in Cleveland now, it's easy to love the winning Indians.
Today, IBM published a POWER6 home run on the Business Intelligence benchmark, TPC-H, with #1 performance and #1 price/performance.(1)
This result is an outstanding winning combination for POWER6 technology and the new IBM Balanced Warehouse, the complete data warehousing solution comprised of pre-tested, scalable and fully-integrated system components of DB2 Warehouse, Server and Storage.
Winning may not be everything, but it usually is in benchmarks.
************************************************
(1) IBM TPC-H 10000GB result on the IBM System p 570 of 343,551.2 QphH@10000GB ($32.89/QphH, avail. 4/15/2008) on a 32 node cluster of p570 (64 POWER6 4.7 GHz processor chips, 128 cores, 256 threads) with 24 MB L3 cache per processor chip and 32 GB of memory per node running DB2 Warehouse 9.5 on AIX 5L V5.3. Total disk capacity was 110,489.27 GB in a IBM Totalstorage DS4800 storage subsystem and the cluster was networked via one Integrated 10Gigabit Ethernet.
Source: http://www.tpc.org; Results current as of 10/15/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
TPC Benchmarks are registered trademarks of the Transaction Processing Performance Council.
technorati tags: IBM, systems, performance, POWER6, benchmark,DB2,Balanced Warehouse,p6,TPC-H,benchmarking
Categories
: [ DB2 | POWER6 | TPC-H ]
Oct 15 2007, 12:52:25 PM EDT
Permalink
|
|
Comparing Apples to Kumquats with the Sun T2
Today Sun announced the SPARC Enterprise T5120 and T5220 servers - the first servers using the new UltraSPARC T2 processor.
Sun compares an SAP SD T5120 benchmark to an IBM System p 570 POWER6 result - but, amazingly, the footnote about this claim is all about the HP rx6600. What's going on here? Maybe the footnote is about the HP system because in reality the POWER6 p570 achieves 1.84 times more performance than the T5120.(1)
Sun compares a T5220 Notesbench result to the IBM System x3650, a configuration that is half the number of cores and more than 2.6 times less memory.(2)
Sun compares other vendors' power configurator values and maximum power values to Sun's actual power measurements from their benchmarks. Not exactly apples to apples is it ?
Not even Cortlands vs. guavas.
******************************************************
(1) Sun SPARC Enterprise Model T5120 (8 cores, 64 threads, 1 UltraSPARC T2 1.4 GHz chip) with 64 GB main memory SAP SD 2-tier result of 2,175 SD Users ( 1.91 seconds avg. response time, cert #2007059) vs. IBM System p 570 (4.7 GHz) best 8-core two-tier SAP SD Standard Application Benchmark result (4010 benchmark users, 1.96 second average response time, cert # 2007038) running Oracle 10g, AIX 5L V5.3, SAP ECC Release 6.0. (8 processor cores/4 chips/16 threads) with 64 GB memory.
(2) Sun SPARC Enterprise 5220 (1 1.4 GHz UltraSPARC T2, 8 cores, 64GB RAM) result of 43,000 users, $2.89 per user, 36,240 NotesMark tpm, 584 ms avg on NotesBench R6iNotes vs. IBM x3650 4-core(2 Intel Xeon 5160 3.0GHz chips, 24GB RAM) result of 22,000 Users, 17,777 NotesMark, $3.47/User, $4.30/NotesMark
Sources: http://www.sap.com/solutions/benchmark/; http://www.notesbench.org; Results as of 10/09/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
SAP, mySAP and other SAP product and service names mentioned herein as well as their respective
logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all
over the world.
technorati tags: IBM, systems, performance, SAP,POWER6, Notesbench,Sun, T2,benchmarking
Categories
: [ Notesbench | POWER6 | SAP | Sun | T2 | power ]
Oct 09 2007, 02:01:56 PM EDT
Permalink
|
|
The Day Chicago Died
Yesterday, Chicago had to stop their marathon right in the middle of it; you see, runners were falling all over the place, many had to be taken to hospitals, and one runner died. All because of the heat.
Last week, it was announced that sea ice is disappearing at an even faster rate than anyone thought.
Just when you think it might be safe to forget about the possibility of something some call global warming, it stares you in the face.
IBM has led the technology industry in energy-smart innovation for over 40 years and is committed to climate protection. It is IBM's goal to sustain leadership in energy conservation and management by continuing to deliver power-management and cooling technologies.
And make sure not to miss the new white paper, A Power Benchmarking Study on the IBM System z9: Applying Energy Efficiency Metrics to Performance, highlighting a Java power benchmarking study and analyzing client server consolidation as an energy management strategy.
I walk everywhere I can. I have not turned on the air conditioner all sweltering week. But I may move the hiking trip to Glacier from 2014 to 2010 just in case.
************************************************
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
technorati tags: IBM, systems, performance, energy efficiency, benchmark,power benchmarking,z9,benchmarking
Categories
: [ energy | z9 ]
Oct 08 2007, 02:18:14 PM EDT
Permalink
|
|
A Billion Served
Yesterday I noticed that my eyes were getting really tired. In 10 hours, I had probably scanned several hundred pages of notes, articles, and technical papers. So for a busy day, my average pages per hour may approach 50 or so.
Yesterday, IBM introduced the latest version of General Parallel File System. GPFS identifies and migrates files between different storage pools and feeds high-speed business intelligence and scientific computers. Highlighting the announcement, IBM revealed that it was able to scan one billion files in less than three hours using GPFS in an internal performance benchmark.
One billion files in less than 3 hours - and these numbers are only going to improve in the future. Whereas I will be looking at reading glasses.
************************************************
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
technorati tags: IBM, systems, performance, GPFS, benchmark,press release,business intelligence,HPC,files,benchmarking
Categories
: [ GPFS ]
Oct 03 2007, 10:51:48 AM EDT
Permalink
|
|
And They Thought We Wouldn't Notice
HP, Sun, and Hitachi all published SPC-1 storage benchmarks yesterday. The results were certainly not embarrassing, though they were 36% behind the #1 IBM SAN Volume Controller result. (1) Note that the IBM price/performance was also at least 30% better.
What was most interesting about these results is that the Host System used for this benchmark with the storage was not an HP server, not a Sun server, and not even a server with a name you don't recognize. The Host System used for all three of these benchmarks was the IBM p5-595.
Yes, you heard it right. An IBM System p5-595.
This one's for the "And They Thought We Wouldn't Notice the p5-595 in the Corner" file.
******************************************************
(1)The top SPC-1 result is the IBM TotalStorage SAN Volume Controller result of 272,505 IOPS (22,433 GB, $12.05/IOP, Mirroring).
Hitachi, Sun and HP published the same SPC-1 storage benchmark result of 200,245.73 IOPS over 26,000 GB ASU Capacity (mirrored) on the Hitachi Universal Storage Platform V ($17.61 $/IOPS), the Sun StorageTek® 9990 System ($17.31 $/IOPS) and the HP StorageWorks XP24000 Disk Array ($17.96 $/IOPS), respectively.
Source: http://www.storageperformance.org/results; Results as of 10/02/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
SPC Benchmark-1 and SPC Benchmark-2 are trademarks of the Storage Performance Council.
technorati tags: IBM, systems, performance, Storage,SPC-1, Sun,HP,p5-595,SVC,benchmarking
Categories
: [ HP | SPC-1 | Sun | p5-595 ]
Oct 02 2007, 01:58:34 PM EDT
Permalink
|
|
What's Wrong with this Picture, Sun ?
Remember when you were a kid and couldn't wait to get Highlights magazine. Some of us got it in the mail and some of us had to wait until the next visit to the dentist - but it was always a treat. And one of my favorite parts of the magazine was the picture game that had all sorts of things wrong with it - a penguin in a swimming pool, a dog with rabbit ears, a Christmas tree in July.
Yesterday, Sun published a press release announcing the new Sun Fire X4450 and X4150 Xeon systems with "industry-leading performance" claims. The only problem is that there was no performance data, no benchmarks.
Sort of like the horse with no head in the magazine.
************************************************
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
technorati tags: IBM, systems, performance, Sun, benchmark,press release,X4450,X4150,Highlights,benchmarking
Categories
: [ Sun ]
Sep 26 2007, 04:23:48 PM EDT
Permalink
|
|
A Bad Hair Day for Unisys
Sometimes, we don't know why, but we just have a bad hair day. Whatever we do, whether it involves special shampoos, conditioners, or even gels, it just seems that we can't get it right.
That must have been the feeling Unisys had yesterday. First, they had to dispute an allegation made in The Washington Post that they "did not properly install essential security systems" for the Department of Homeland Security.
Then, they publish an SAP SD Benchmark result that is 35% behind even Sun.(1) And the IBM POWER6 result that uses just HALF the cores and HALF the memory is 47% better.(2)
The good news here for Unisys is that tomorrow is another day.
******************************************************
(1) Unisys SAP SD result of 5,408 SD Users (1.93 avg. response time, cert #2007058) on the Unisys Enterprise Server Model ES7000/one with 16 processors / 32 cores / 64 threads, DC Xeon 7140M, 3.4 GHz, 1 MB L2 cache per core, 16 MB L3 cache per processor and 256 GB main memory running SQL Server 2005 on Windows Server 2003 Datacenter Edition with SAP ECC Release 6.0 vs. Sun SAP SD result of 7300 SD Users (1.98 avg. response time, cert #2007026) on the Sun/Fujitsu SPARC Enterprise Server Model M8000 (16 2.4 GHzSPARC64 VI chips, 64 threads) running Oracle 10g on Solaris 10.
(2) IBM SAP SD 16-core IBM System p 570 (8 4.7 GHz POWER6 chips, 32 threads) result of 8000 users (1.98 second average response time, cert # 2007039) 128 GB main memory, running DB2 9, AIX 5L V5.3, SAP ECC Release 6.0
Source: http://www.sap.com/solutions/benchmark/; Results as of 9/25/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
SAP, mySAP and other SAP product and service names mentioned herein as well as their respective
logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all
over the world.
technorati tags: IBM, systems, performance, SAP,POWER6, Unisys,Sun, benchmarking
Categories
: [ POWER6 | SAP | Sun | Unisys ]
Sep 25 2007, 05:28:34 PM EDT
Permalink
|
|
Play it Again, Sam
Sometimes, you can never hear your favorite song too often.
Last week, IBM once again published a leadership POWER6 benchmark, this time on SPECjAppServer2004. IBM's benchmark result was the #1 4-core result, 37% greater than HP's. (1)
SPECjAppServer2004 (Java Application Server) is a multi-tier end-to-end benchmark for measuring the performance of Java 2 Enterprise Edition (J2EE) technology-based application servers.
As time goes by, the POWER6 benchmark song just doesn't get old.
************************************************
Source: www.spec.org . Results current as of 09/18/07.
(1) IBM SPECjAppServer2004 result of 1,198 JOPS on an IBM System p 570 (4 cores, 2 4.7 GHz POWER6 chips, 2 cores/chip, 16 GB memory) running WebSphere 6.1 Application Server on IBM AIX 5L v5.3 (database server: p5-550 (4 cores, 2.1 GHz POWER5+ chips, 2 cores/chip) running DB2 V9.1 on AIX 5L v5.3) vs. HP SPECjAppServer2004 result of 874 JOPS on an HP rx2600 (4 cores, 2 1.6 GHz Itanium 2 chips, 2 cores/chip, 16 GB memory) running Oracle Application Server 10g Release 10.1.3.2 - Java Edition on HP-UX (database server: HP rx2660 (4 cores, 2 1.6 GHz Itanium 2 chips, 2 cores/chip, 24 GB memory) running Oracle Database 10g Enterprise Edition Release 10.2.0.2 on HP-UX 11i Version 3)
SPEC and the SPEC benchmark names are registered trademarks of the Standard Performance Evaluation Corporation.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
technorati tags: IBM, systems, performance, SPECjAppServer2004, benchmark,HP,POWER6,benchmarking
Categories
: [ HP | POWER6 | SPECjAppServer2004 ]
Sep 18 2007, 01:20:26 PM EDT
Permalink
|
|
On Top of the World and On World Records
On May 29,1953 Sir Edmund Hillary and Tenzing Norgay were the first to reach the top of Mount Everest. Since that time, there have been numerous other "world records": first winter ascent, first ascent by a woman, fastest descent, first legally blind person, longest stay on top, first person to hike from sea level to summit with no oxygen . . . and the list goes on.
Sun published a press release yesterday announcing the Sun Blade X8440 Server Module, a blade server designed for Quad-Core AMD Opteron processors. The press release claimed "an 8-thread World Record on the industry-standard SPECompM2001 benchmark." (1)
Note that this result is actually on the dual-core model. Also note that this result is claimed as an "8-thread" record. IBM holds the #1 results for SPECompM2001 at 4-core, 8-core, 16-core, plus the overall #1 "world record." (2)
I could probably get the world record for the first female IBMer over 39 from Ohio to make it to Base Camp. (Note that I haven't actually verified this claim.)
The point here is that world records are great but some only really matter to the person who holds the record and maybe their two best friends.
************************************************
Source: www.spec.org . Results current as of 09/11/07.
(1) Sun Blade X8440 (4xAMD Opteron model 8222, 8 cores, 4 chips, 2 cores/chip, 8 threads, 3.0 GHz): 24,246 SPECompM2001.
(2) http://www.ibm.com/systems/p/benchmarks/hpc.html
SPEC and the SPEC benchmark names are registered trademarks of the Standard Performance Evaluation Corporation.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
technorati tags: IBM, systems, performance, SPECompM, benchmark,Sun,threads,benchmarking
Sep 11 2007, 11:23:58 AM EDT
Permalink
|
|
The Art and Science of Press Releases
On the last day of August, IBM published a press release highlighting leadership benchmarks. There were several performance claims, each one backed up by a fabulously detailed footnote.
Yesterday HP published a press release announcing their new quad-core systems. The questionable performance claims required much imagination and research to get the details, as evidenced by the scanty footnotes.
This week Sun published a press release announcing new quad-core systems. There were no performance claims, no benchmarks, no footnotes.
The devil is in the details.
************************************************
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
technorati tags: IBM, systems, performance, POWER6, benchmark,press release,HP,Sun,footnotes,benchmarking
Categories
: [ HP | POWER6 | Sun ]
Sep 07 2007, 05:42:31 PM EDT
Permalink
|
|
With a Cherry on Top
Last Sunday, I had the greatest sundae . . . chocolate ice cream filled with nuts and caramel, marshmallow topping, a ton of whipped cream, and a cherry on top. Of course, it was all the better coming from an outside stand in a small village by the river, after a 26 mile bike ride.
The sundae was sweet, which reminded me of all the great benchmark results announced with the POWER6 launch, all the great results that have been added to those as topping, and now another cherry on top.
Last week, IBM once again announced a POWER6 #1 benchmark result, this time on the SAP EP-ESS (Employee Self-Service) standard application benchmark. This result is the top result for this benchmark, surpassing Sun by 80%. (1)
The EP-ESS Benchmark uses the Business Package for Employee Self-Service. EP-ESS provides a number of iViews that allow employees to create, display, and change their own data, such as department, name, address. This benchmark focuses on concurrent users and tests the performance of the portal platform (running on the SAP NetWeaver Java stack), while launching business transactions.
And now you just have to wonder. What other amazing POWER6 benchmarks are going to be added to the portfolio, like yummy sprinkles on top ?
******************************************************
(1) IBM System p 570 #1 result of 2,600 EP-ESS users with average online response time of 1.81 seconds (cert.# 2007052) on the dual-core 4.7 GHz POWER6 p570 (4 cores, 2 chips, 8 threads) running AIX 5L Version 5.3, IBM Java 1.4.2, NetWeaver Portal 7.0 SP06 and NetWeaver 7.0 (2004s) vs. Sun SAP EP-ESS result of 1,440 EP-ESS users with average online response time of 1.94 seconds (cert.# 2006012) on the dual-core 2.2 GHz Opteron Sun Fire Model V20z (4 cores, 2 chips, 2 cores/chip) running 64-bit versions of SuSE Linux Enterprise Server 9, IBM JSE 1.4.2, NetWeaver Portal 6.0 and NetWeaver 2004.
Source: http://www.sap.com/solutions/benchmark/; Results as of 9/06/07.
The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.
SAP, mySAP and other SAP product and service names mentioned herein as well as their respective
logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all
over the world.
technorati tags: IBM, systems, performance, SAP,POWER6, NetWeaver,Sun, benchmarking
Categories
: [ POWER6 | SAP | Sun ]
Sep 06 2007, 11:10:14 AM EDT
Permalink
|
|
|
|
Benchmarking and systems performance - hosted by Elisabeth Stahl
