Posted on 01/10/2002 6:12:44 PM PST by Bush2000
Store the database on a Symmetrix and replicate it with ECC-ERM and it doesn't matter what platform you use. The data is safe.
Crop dusters outmanuever stealth fighter bombers... doesn't mean I want to use one in a night time bombing raid." Bushman wont get it. Doesn't want to. Does he think any of us are going to consider for one SECOND, his favorite system?
Crop dusters outmanuever stealth fighter bombers... doesn't mean I want to use one in a night time bombing raid." Bushman wont get it. Doesn't want to. Does he think any of us are going to consider for one SECOND, his favorite system?
This may be good for the M$ PR machine and for the blind followers of Uncle Bill, but DO YOUR RESEARCH! M$ is not what it always the best choice. I run Linux and ME at home. Solaris 8, Win98SE, Win2K Pro, Win2K Advanced Server, NT 4.0 Server and XP at work. You got to look at what you need and make the wise choice.
Semper Fi,
DataDink
Rank | Company | System | tpmC | Price/tpmC | System Availability | Database | Operating System | TP Monitor | Date Submitted |
1 | PRIMEPOWER 2000 c/s w 66 Front-Ends | 455,818 | 28.58 US $ | 02/28/02 | SymfoWARE Server Enterp. Ed. VLM 3.0 | Sun Solaris 8 | BEA Tuxedo 6.5 CFS | 08/28/01 | |
2 | HP 9000 Superdome Enterprise Server | 389,434 | 21.24 US $ | 05/15/02 | Oracle 9i Database Enterprise Edition | HP UX 11.i 64-bit | BEA Tuxedo 6.4 | 12/21/01 | |
3 | Compaq AlphaServer GS320 | 230,533 | 44.62 US $ | 07/30/01 | Oracle 9i Database Enterprise Edition | Compaq Tru64 UNIX V5.1 | Compaq DB Web Connector V1.1 | 06/18/01 | |
4 | PRIMEPOWER 2000 c/s w /32 Front Ends | 222,772 | 43.42 US $ | 06/30/01 | SymfoWARE Server Enterp. Ed. VLM 3.0 | Sun Solaris 8 | BEA Tuxedo 6.4 CFS | 04/13/01 | |
5 | IBM eServer pSeries 680 Model 7017-S85 | 220,807 | 34.18 US $ | 04/13/01 | Oracle8 Enterprise Edition v8.1.7.1 | IBM AIX 4.3.3 | Webshpere App. Server Ent. Edition V.3.0 | 03/20/01 | |
*** | Bull Escala EPC2450 c/s | 220,807 | 34.67 US $ | 05/28/01 | Oracle 8i Enterprise Edition v. 8.1.7 | IBM AIX 4.3.3 | Webshpere App. Server Ent. Edition V.3.0 | 05/28/01 | |
6 | HP 9000 Superdome Enterprise Server | 197,024 | 32.13 US $ | 05/01/01 | Oracle8 Enterprise Edition v8.1.7.1 | HP UX 11.i 64-bit | BEA Tuxedo 6.4 | 12/21/01 | |
7 | Unisys e-@ction Enterprise Server ES7000 | 165,218 | 21.33 US $ | 03/10/02 | Microsoft SQL Server 2000 Enterprise Edition | Microsoft Datacenter Server Limited Edition | Microsoft COM+ | 11/11/01 | |
8 | IBM eServer iSeries 400 Model 840-2420-001 | 163,775 | 51.58 US $ | 12/15/00 | IBM DB2 for AS/400 V4 R5 | IBM OS/400 V4 R5 | BEA Tuxedo 6.4 | 03/19/01 | |
9 | Compaq AlphaServer GS320 Model 6/731 | 155,179 | 52.88 US $ | 02/02/01 | Oracle 8i Enterprise Edition v. 8.1.7 | Compaq Tru64 UNIX V5.1 | Compaq DB Web Connector | 04/03/01 | |
10 | IBM eServer iSeries 400 Model 840-2420 | 152,346 | 44.52 US $ | 09/15/00 | IBM DB2 for AS/400 V4 R5 | IBM OS/400 V4 R5 | BEA Tuxedo 6.4 | 03/19/01 |
*** - Duplicate results are shown with an asterisk (*) in the Rank column. Click here for more information about duplicates.
©2001 TPC. All Rights Reserved.
![]() |
||
Top Ten TPC-C by Price/Performance
|
![]() |
![]() |
All Results | Clustered Results | Non-Clustered Results |
Rank | Company | System | tpmC | Price/tpmC | System Availability | Database | Operating System | TP Monitor | Date Submitted | Cluster |
1 | ![]() |
PowerEdge 2500/1.13/1P | 11,314 | 4.38 US $ | 12/14/01 | Microsoft SQL Server 2000 Standard Edt. | Microsoft Windows 2000 Server | Microsoft COM+ | 12/14/01 | N |
2 | IBM e(logo) xSeries 250 c/s | 15,533 | 4.67 US $ | 11/05/01 | Microsoft SQL Server 2000 Standard Edt. | Microsoft Windows 2000 Server | Microsoft COM+ | 11/05/01 | N | |
3 | ![]() |
PowerEdge 2500/1.13/1P | 11,320 | 4.70 US $ | 10/31/01 | Microsoft SQL Server 2000 Standard Edt. | Microsoft Windows 2000 Server | Microsoft COM+ | 10/31/01 | N |
4 | IBM eServer xSeries 220 c/s | 9,112 | 4.76 US $ | 10/16/01 | Microsoft SQL Server 2000 Standard Edt. | Microsoft Windows 2000 Server | Microsoft COM+ | 10/16/01 | N | |
5 | ProLiant ML530-X1000-1P | 9,347 | 4.77 US $ | 09/25/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 Server | Microsoft COM+ | 09/25/01 | N | |
6 | ![]() |
IBM e(logo) xSeries 350 c/s | 20,422 | 5.39 US $ | 10/01/01 | Microsoft SQL 2000 | Microsoft Windows 2000 Server | Microsoft COM+ | 10/01/01 | N |
7 | ![]() |
ProLiant ML570-700 3P | 20,207 | 5.64 US $ | 09/26/00 | Microsoft SQL 2000 | Microsoft Windows 2000 Server | Microsoft COM+ | 07/27/01 | N |
8 | ![]() |
ProLiant ML570 6/900-4P | 37,100 | 6.36 US $ | 11/12/01 | Microsoft SQL Server 2000 Enterprise Edition | Microsoft Windows 2000 Advanced Server | Microsoft COM+ | 11/12/01 | N |
9 | ![]() |
PowerEdge 6400 | 20,331 | 6.46 US $ | 02/02/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 | Microsoft COM+ | 03/19/01 | N |
10 | ![]() |
PowerEdge 6450 | 20,320 | 6.62 US $ | 02/02/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 | Microsoft COM+ | 03/19/01 | N |
![]() |
||
Top Ten Clustered TPC-C by Performance
|
![]() |
![]() |
All Results | Clustered Results | Non-Clustered Results |
Rank | Company | System | tpmC | Price/tpmC | System Availability | Database | Operating System | TP Monitor | Date Submitted |
1 | ProLiant DL760-900-256P | 709,220 | 14.96 US $ | 10/15/01 | Microsoft SQL Server 2000 Enterprise Edition | Microsoft Windows 2000 Advanced Server | Microsoft COM+ | 09/19/01 | |
2 | IBM e(logo) xSeries 370 c/s | 688,220 | 22.58 US $ | 05/31/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 Datacenter Server | Microsoft COM+ | 04/10/01 | |
3 | ProLiant DL760-900-192P | 567,882 | 14.04 US $ | 10/15/01 | Microsoft SQL Server 2000 Enterprise Edition | Microsoft Windows 2000 Advanced Server | Microsoft COM+ | 09/19/01 | |
4 | IBM e(logo) xSeries 370 c/s | 440,879 | 19.35 US $ | 12/07/00 | IBM DB2 UDB 7.1 | Microsoft Windows 2000 Advanced Server | Microsoft COM+ | 04/11/01 | |
5 | ProLiant DL760-900-128P | 410,769 | 13.02 US $ | 10/15/01 | Microsoft SQL Server 2000 Enterprise Edition | Microsoft Windows 2000 Advanced Server | Microsoft COM+ | 09/19/01 | |
6 | IBM e(logo) xSeries 370 c/s | 363,129 | 21.80 US $ | 05/31/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 Datacenter Server | Microsoft COM+ | 04/10/01 | |
7 | IBM eServer xSeries 370 | 136,766 | 16.93 US $ | 09/20/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 Datacenter Server | Microsoft COM+ | 04/24/01 | |
8 | IBM e(logo) xSeries 370 c/s | 121,319 | 18.97 US $ | 05/31/01 | Microsoft SQL Server 2000 | Microsoft Windows 2000 Datacenter Server | Microsoft COM+ | 04/10/01 |
You know, I don't see why I should bother to answer your posts. I've said a number of times that I am not a computer bigot. Many of these other folks are trying to talk to you in friendly tones, too. From what I can tell, all you want to do is make the "other guy" look bad. I'm sick of your attitude and snotty tone. For example, in post number 14 you said:
Can you read? Go over to tpc.org and read the results, if you can. Windows and SQL Server kicked the crap out of every competitor.
This isn't exactly true. You always pay a premium for top of the line performance. I don't have a problem with Window's price point, or having a good price/performance ratio. This has always been the selling point for IBM/Intel and Microsoft/Intel. But performance costs, whether it's a mid or large computer.
The Unisys ES7000 that came in a the good pricepoint was a single 32-processor Pentium III machine with a main memory of 64GB. For $3.5 million you can get a Sun or IBM that performs just as well. But it wasn't the top performer. The Fujitsu PrimePower 2000 that had the good performance was a 128-processor computer with 256GB main memory. It's certianly expensive, but gives you a great deal of computing power in one package.
The absolute top-performing configration from the Intel/Microsoft side was a group of 32 Compaq Proliant DL-760s each with 8 processors. Note that this configuration was nowhere in the price/performance category. All of the configurations in the price/performance categort are an order of magnitude less in terms of price and transactions/per unit time. The configurations in the pure performance category are all state-of-the-art computers in which money is no object.
So if you are going to debate, please try to compare apples to apples and oranges to oranges. First you were saying that Microsoft beat Unix totally, but then you back-pedaled and restricted yourself to the lower end computers.
The Dell PowerEdge 2500 is a dual 1Gz and 6GB memory machine. The price for the test machine came to about $50,000 total. These Dells are good machines, and I actually recommend them in my job. However, for other things I recommend the Sun 280 series which is similar, with Sparc CPUs (900Mhz) and up to 8GB RAM. This computer comes in more than $50,000, but not that much more. The real factor is what are the needs of the customer.
Clustered configurations don't always "kick the crap" as you so glibly put it, out of non-clustered ones. There are other issues with this set up, like more points of failure, even floor space.
I get the feeling you are looking at these charts and saying one is better than the other, but I do not get the impression you have no experience actually having to choose one over the other. Especially saying things like this:
This is the reason that Sun and Oracle guys are having a tough time explaining to their customers why they should pay twice as much and get half the performance.
This shows me you don't know what you're talking about. I'll say that again, because it bears repeating: I don't have a hard time justifying an expensive computer, my customers know why they are buying. You are just blowing off steam. Bravado. Insulting bravado. The cost of these machines does not end with the purchase price. There are many other factors involved with a major purchase in the multi millions of dollars. Have you actually signed a deal for a multimillion dollar computer, then had to set it up and operate it? I have, being in the oil industry. I also spec out the midrange computers, too. The cost does not end with the purchase price. It never has. There are support contracts, maintenence, what is the expected life of the equipment, who at your company has the skills to make the beast worth the money, has your company committed to a particular platform already, etc.
So in the end, it doesn't come down to that number on the TPC webpage. I don't have a hard time justifying unix to my customers. And you, sir, are full of yourself and full of it. Shape up or ship out.
That is the only spot in the entire article that mentions price/performance. The rest of the article deals with the numbers that I posted.
The point I was making was the fact that the numbers mentioned in the article do not jive with the website any longer.
Maybe you should: 1) re-read the article and 2) not skip the big words that are apparently hurting your brain. MS does NOT compare to the high end systems as well as Uncle Bill wants everyone to beleive and your comprehension of the article is skewed as well. Another thing you may want to check out is the date of the article itself - November 11,2001. The HP server I highlighted that beats the Windows server was tested on December 21, 2001, showing how the Unix community has begun to responded to the challenge.
Like I said - I deal with at lot of different systems on a daily basis. Each has its strong and weak spots. You have to look at all the data to make the best decision. If you need a "Datacenter server" for a large operation, Unix (not Microsoft) is the way to go.
For something else interesting - but along the same lines - have a look at the Top 500 Supercompters then look at this site. Take note on the second site how many of the world's fastest 500 Supercomputers are clusters - a lot of them from the Unix houses (Sun, Compaq and IBM)
Enjoy,
DataDink
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.