IDS 11.5 vs SQL SERVER 2008
The poster asked whether IDS 11.5 outperforms SQL Server 2008 and wanted white papers/percentage figures to argue against his company switching to SQL Server. Replies noted no published Informix benchmark data existed (IBM said 'soon'), and that platform choices usually hinge on cost/skills, not raw speed. Art Kagel suggested running the open-source BenchmarkSQL (TPC-C-like) tool, already ported to Informix by AGS, as a quick comparison; Obnoxio argued benchmarking your own application workload on identical hardware is far more meaningful. No actual performance numbers or firm conclusion were given.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Is IDS 11.5 faster then SQL SERVER 2008 ? Are there any white papers that show this ?
Yes and no. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Sun, May 23, 2010 at 11:32 PM, BRETT SIMPSON <brett.simpson@actrol.com.au > wrote: > Is IDS 11.5 faster then SQL SERVER 2008 ? > > Are there any white papers that show this ? > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --000e0cd2e198e7c85204874f6375
hmmm maybe a bit more example data this is very much quess work... how about we have IDS 11.5 x64 on a linux x64 (maybe red-hat/suse) machine with a raw database chunks database. vs sql server 2008 on windows 2008 x64 both servers optimized properly with the same database data. how much faster would the IDS be against SQL server ? % wise.... its hard to find any info reguarding this via google.... my company is considering jumping to sql server so im trying to build a case to stay with informix....
BRETT SIMPSON wrote: > Is IDS 11.5 faster then SQL SERVER 2008 ? In what sense? -- Cheers, Obnoxio The Clown http://obotheclown.blogspot.com I will now proceed to pleasure myself with this fish. -- This message has been scanned for viruses and dangerous content by OpenProtect(http://www.openprotect.com), and is believed to be clean.
Typically a decision like that is not based on performance. Are you having problems with Informix? Which version are you using? Are your company considering maintenance costs or acquisition costs? Do you prefer Windows or other platforms? What is the driving force behind the will of change? Regards. On Mon, May 24, 2010 at 5:59 AM, BRETT SIMPSON <brett.simpson@actrol.com.au>wrote: > hmmm maybe a bit more example data > > this is very much quess work... > > how about we have IDS 11.5 x64 on a linux x64 (maybe red-hat/suse) machine > with a raw database chunks database. > > vs sql server 2008 on windows 2008 x64 > > both servers optimized properly > > with the same database data. > > how much faster would the IDS be against SQL server ? > > % wise.... > > its hard to find any info reguarding this via google.... > > my company is considering jumping to sql server so im trying to build a > case > to stay with informix.... > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --0016e6567d16bfe2940487535032
There are as of now no benchmark data published on Informix performance. IBM promises to have it out 'soon' but no firm dates and no hint about what platforms will be published first. You should just perform you own benchmarks. There is an open source benchmark project that produces a TPC/C-like benchmark that has been ported to Informix. Advanced DataTools used it for the Fastest DBA contest at the recent IIUG Conference. You can set up both databases on your own hardware, even resize the test databases and transaction mix and configure the workload to better match your environment and run the thing in-house to show to your management. Note that engine tuning is important for these benchmarks. In the contest we saw transaction rates from 80TPS to over 900TPS for the simple system we set up and the contest restricted the contestants from some important optimizations like isolating high access tables from each other on disk. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Mon, May 24, 2010 at 12:59 AM, BRETT SIMPSON <brett.simpson@actrol.com.au > wrote: > hmmm maybe a bit more example data > > this is very much quess work... > > how about we have IDS 11.5 x64 on a linux x64 (maybe red-hat/suse) machine > with a raw database chunks database. > > vs sql server 2008 on windows 2008 x64 > > both servers optimized properly > > with the same database data. > > how much faster would the IDS be against SQL server ? > > % wise.... > > its hard to find any info reguarding this via google.... > > my company is considering jumping to sql server so im trying to build a > case > to stay with informix.... > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --00504502ad23770065048754ad9d
Art Kagel wrote: > There are as of now no benchmark data published on Informix performance. > IBM promises to have it out 'soon' but no firm dates and no hint about what > platforms will be published first. You should just perform you own > benchmarks. There is an open source benchmark project that produces a > TPC/C-like benchmark that has been ported to Informix. Advanced DataTools > used it for the Fastest DBA contest at the recent IIUG Conference. This is a very bad idea. Sorry Art, but it is. What you should do is benchmark key parts of *your own* application on SQL Server and on IDS on the same kit. A third-party benchmark in this situation could give you entirely invalid information for your environment. If you're going to run your own benchmark, make it a useful one. -- Cheers, Obnoxio The Clown http://obotheclown.blogspot.com I will now proceed to pleasure myself with this fish. -- This message has been scanned for viruses and dangerous content by OpenProtect(http://www.openprotect.com), and is believed to be clean.
I actually agree with you OTC. And if the OP's company were making an actual reasonable and reasoned evaluation of different database platforms for their systems it would be exactly what I would suggest. However, I suspect that what's actually happening is that the OP is trying to head off a MS bigot trying to force the company to switch for no better reason than "Everyone uses SQL Server!". For that purpose, a quick and dirty benchmark that can be implemented and run in a few hours without anyone knowing until it's completed and the results can be presented is a better solution than a more comprehensive and usable benchmark that would require many days of implementing and porting some portion of the existing database and application code. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Mon, May 24, 2010 at 6:47 AM, Obnoxio The Clown <obnoxio@serendipita.com>wrote: > Art Kagel wrote: > > There are as of now no benchmark data published on Informix performance. > > IBM promises to have it out 'soon' but no firm dates and no hint about > what > > platforms will be published first. You should just perform you own > > benchmarks. There is an open source benchmark project that produces a > > TPC/C-like benchmark that has been ported to Informix. Advanced DataTools > > used it for the Fastest DBA contest at the recent IIUG Conference. > > This is a very bad idea. Sorry Art, but it is. > > What you should do is benchmark key parts of *your own* application on > SQL Server and on IDS on the same kit. A third-party benchmark in this > situation could give you entirely invalid information for your environment. > > If you're going to run your own benchmark, make it a useful one. > > -- > Cheers, > Obnoxio The Clown > > http://obotheclown.blogspot.com > I will now proceed to pleasure myself with this fish. > > -- > This message has been scanned for viruses and > dangerous content by OpenProtect(http://www.openprotect.com), and is > believed to be clean. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001636e0a7f5e0fefb048757beb7
Art Kagel wrote: > I actually agree with you OTC. And if the OP's company were making an > actual reasonable and reasoned evaluation of different database platforms > for their systems it would be exactly what I would suggest. However, I > suspect that what's actually happening is that the OP is trying to head off > a MS bigot trying to force the company to switch for no better reason than > "Everyone uses SQL Server!". For that purpose, a quick and dirty benchmark > that can be implemented and run in a few hours without anyone knowing until > it's completed and the results can be presented is a better solution than a > more comprehensive and usable benchmark that would require many days of > implementing and porting some portion of the existing database and > application code. Fair comment. However, I'm pretty sure the benchmark that you're referring to won't run on Informix without some modification... -- Cheers, Obnoxio The Clown http://obotheclown.blogspot.com I will now proceed to pleasure myself with this fish. -- This message has been scanned for viruses and dangerous content by OpenProtect(http://www.openprotect.com), and is believed to be clean.
The nice folk at AGS modified it for Informix already. I don't remember if the source forge version has been been updated ( http://sourceforge.net/projects/benchmarksql) but anyone who wants the updated version can contact me directly. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Mon, May 24, 2010 at 10:28 AM, Obnoxio The Clown <obnoxio@serendipita.com > wrote: > Art Kagel wrote: > > I actually agree with you OTC. And if the OP's company were making an > > actual reasonable and reasoned evaluation of different database platforms > > for their systems it would be exactly what I would suggest. However, I > > suspect that what's actually happening is that the OP is trying to head > off > > a MS bigot trying to force the company to switch for no better reason > than > > "Everyone uses SQL Server!". For that purpose, a quick and dirty > benchmark > > that can be implemented and run in a few hours without anyone knowing > until > > it's completed and the results can be presented is a better solution than > a > > more comprehensive and usable benchmark that would require many days of > > implementing and porting some portion of the existing database and > > application code. > > Fair comment. However, I'm pretty sure the benchmark that you're > referring to won't run on Informix without some modification... > > -- > Cheers, > Obnoxio The Clown > > http://obotheclown.blogspot.com > I will now proceed to pleasure myself with this fish. > > -- > This message has been scanned for viruses and > dangerous content by OpenProtect(http://www.openprotect.com), and is > believed to be clean. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --00504502c5f5abea1c0487580486