Dual CPU Database Server Comparison
by Johan De Gelas on December 2, 2004 12:11 AM EST- Posted in
- IT Computing
Benchmarks IBM DB2 8.1.3: Intel versus AMD
The first question that most people will ask is, of course, how the best AMD Opteron compares to the newest Intel Xeon "Nocona" CPU. Below is a quick table to refresh your memory and to enable you to compare price/performance:
| Intel Xeon CPUs | Core | L2 cache | L3 cache | x86-64 bit | In Test | Price |
| 3.60 GHz w/ 1M cache 800 MHz FSB (90nm) | Nocona = "Prescott server" | 1 MB | No | Yes | Yes | $851 |
| 3.40 GHz w/ 1M cache 800 MHz FSB (90nm) | Nocona = "Prescott server" | 1 MB | No | Yes | No | $690 |
| 3.20D GHz w/ 1M cache 800 MHz FSB (90nm) | Nocona = "Prescott server" | 1 MB | No | Yes | No | $455 |
| 3 GHz w/ 1M cache 800 MHz FSB (90nm) | Nocona = "Prescott server" | 1 MB | No | Yes | No | $316 |
| 3.20C GHz w/ 2M cache 533 MHz FSB (.13) | Galatin = "P4 EE Server" | 0,5 MB | 2 MB | No | Yes | $1,043 |
| 3.20 GHz w/ 1M cache 533 MHz FSB (.13) | Galatin = "P4 EE Server" | 0,5 MB | 1 MB | No | No | $690 |
| 3.06A GHz w/ 1M cache 533 MHz FSB (.13) | Galatin = "P4 EE Server" | 0,5 MB | 1 MB | No | Yes | $455 |
| 3.06 GHz w/ 512k cache 533 MHz FSB (.13) | Prestonia = "Northwood Server" | 0,5 MB | No | No | Yes | $316 |
| AMD Opteron CPU's | Core | L2 cache | L3 cache | x86-64 bit | In Test | Price |
| Model 250 (2.4 GHz) | Sledgehammer | 1 MB | No | Yes | Yes | $851 |
| Model 248 (2.2 GHz) | Sledgehammer | 1 MB | No | Yes | Yes | $690 |
| Model 246 (2.0 GHz) | Sledgehammer | 1 MB | No | Yes | No | $455 |
| Model 244 (1.8 GHz) | Sledgehammer | 1 MB | No | Yes | No | $316 |
We were also very curious about the Xeon Nocona, as the it brings higher clock speeds, a bigger L2-cache, no L3-cache and a pipeline 11 stages longer than the previous Xeon "Prestonia" and Xeon "Gallatin", which maxed out at 3.2 GHz. The first two features mentioned should boost the performance quite well, while the two last are disadvantages.
We should emphasize that, as we tested with SUSE SLES 8 (kernel 2.4.21), the Xeon Nocona was disadvantaged, since we could not test it in 64-bit mode. We assure you that we will update this report with 2.6 kernel. For now, we decided to give you a full report on SLES 8 and kernel 2.4. (All numbers are expressed in queries per second.)
| Concurrency | Xeon 3.6 GHz | Dual Xeon 3.2 L3 (2MB) | Dual Xeon 3.2 | Dual Xeon 3.06 L3 (1MB) | Dual Xeon 3.06 | Opteron 250 DDR400 32 bit | Dual Opteron 250 DDR 400 64 bit | Dual Opteron 248 DDR 400 64 bit |
| 1 | 55 | 46 | 44 | 43 | 42 | 57 | 61 | 57 |
| 2 | 87 | 74 | 61 | 72 | 61 | 105 | 118 | 107 |
| 5 | 128 | 104 | 100 | 98 | 98 | 123 | 137 | 129 |
| 10 | 136 | 112 | 107 | 105 | 102 | 129 | 145 | 132 |
| 20 | 136 | 113 | 106 | 106 | 104 | 131 | 147 | 132 |
| 35 | 138 | 113 | 106 | 104 | 99 | 133 | 150 | 129 |
| 50 | 138 | 110 | 106 | 102 | 100 | 130 | 145 | 128 |
All concurrency tests below 5 are not reliable enough to make any firm conclusion, especially for the Xeon. The margin of error is somewhat higher, but that is not all.
As the Dual Xeon with Hyperthreading spawns 4 logical CPUs, with a concurrency of 2, it is possible that only one physical CPU is doing all the work. Looking at the numbers and the linux tool top, we feel pretty sure that this is exactly what happens most of the time. Compare Row "5" with "2", and "2" with "1" to see what I mean. Note that the results of rows 10 to 50 do not vary a lot; so, we look at these numbers for our conclusions. In the table below, you can see an overview of how the different CPUs compare in percentages.
| 3.6 vs 3.2 | 2 MB L3-cache vs none | 1 MB L3-cache vs none | Xeon 3.2 vs 3.06 | Xeon 3.2 vs 3.06 (both with L3) | Xeon 3.6 vs Opteron 250 | Opteron 64 bit vs 32 bit |
| 20% | 3% | 1% | 7% | 7% | -4% | 6% |
| 17% | 22% | 18% | 3% | 3% | -17% | 12% |
| 24% | 4% | 1% | 5% | 5% | 5% | 12% |
| 21% | 5% | 3% | 6% | 6% | 6% | 13% |
| 21% | 6% | 2% | 6% | 6% | 3% | 12% |
| 22% | 7% | 5% | 8% | 8% | 3% | 12% |
| 26% | 4% | 2% | 8% | 8% | 7% | 12% |
If we had published a similar report back in August, the Opteron would enjoyed a landslide victory. Luckily for Intel, Nocona is very competitive and is about 5% faster than the Opteron 250.
The gigantic - for x86 - L3-cache can not help the Xeon much. We measured only a 2% to 5% performance boost from the 1 MB L2-cache (at 3.06 GHz), and a 4% to 7% performance boost from the 2 MB L3-cache (at 3.2 GHz). The L3-cache seems to boost performance as much as 5% to 6% clock speed increase - nothing to write home about. So a Xeon "Galatin" 3.2 GHz 2 MB L3-cache performs more or less like a Xeon "Galatin" 3.4 GHz, if such a beast should exist.
A comparison between the 3.2 GHz and 3.06 GHz shows that CPU clockscaling - given equal cache sizes - is almost perfect, a testimony to how CPU intensive this benchmark is. Clearly, the generalisation, "databases are all about I/O" is not accurate for a number of database applications. Read-heavy databases seem to be "all about the CPU".
Using a 64 bit database (DB2 8.1.3) on a 64 bit operating system delivers about 12% to 13% better performance. Since we didn't use more than 2 GB, the most likely explanation is the fact that the software can make use of 16 registers instead of 8. We also tested with a twice as large database and 4 GB of RAM, and the results were very similar.
The performance of the Nocona Xeon compared to the older Xeons is also remarkable. The database doesn't mind the longer pipeline and absence of the L3-cache. On the contrary, it performs better than its clock speed indicates, leaving the older 3.2 GHz Xeon (with 2 MB L3 cache!) behind with 21% to 22%, while the Nocona has only a 13% clock speed advantage over the latter. To be honest, we expected Nocona, with its huge branch misprediction penalty, a result of its extremely long pipeline, to scale much worse.
Facebook
Twitter
RSS
46 Comments
View All Comments
markwrob - Friday, February 4, 2005 - link
Great article. One thing I'm looking for onthis and similar comparisons is how the rigs
or their components do on power consumption.
Are there any places to find that out easily?
markwrob - Friday, February 4, 2005 - link
vaxinator - Monday, December 13, 2004 - link
Very interesting review. I think you are wise to narrow your scope to basically in-memory queries, since this pretty much ensures you are testing just the CPU and memory. Disk I/O is another topic.I would be interested to know a few more details about the database (the data model) and the actual queries used.
But the key element you need to watch for is ensuring your client box is not overloaded. I have seen lots of "performance reviews" where the client box was smaller than the device under test...wrong! It should be under-utilized and the cpu consumption/memory use etc should be reported as well. If necessary use lots of client boxes, but watch the network load.
rogerjlarsson - Monday, December 6, 2004 - link
MySQL v. IBM DB2 Single v. Dual"MySQL doesn't seem to scale as well as DB2."
Yea, right... It all depends on what level you start from!
MySQL _Single_ Xeon Concurrency 50: 157
DB2 _Dual_ Xeon Concurrency 50: 102
MySQL scales a little to _Dual_ Conc. 50: 222
DB2 is not even near!!! MySQL might be network
limited rather than Memory or CPU...
davesbeer - Sunday, December 5, 2004 - link
DecoderOracle's ETL tool is $5k also comes with
JDeveloper
Forms Designer
Reports Designer
etc.
etc
I've worked with some of the biggest financial institutions in the world including ABN Amro... etc...
Oracle's DB also is less costly then MS.. Not thier EE version but then again MS has nothing to compete with that one..
Also you mentioned MS tools are the best.. not likely.. they are the easiest to use but far from the best.. The worlds biggest datawarehouses, datamarts etc.. are not on MS and they don't use MS's tools... except for maybe MS's themselves...
jensend - Sunday, December 5, 2004 - link
Comparing these systems in 32-bit mode with a 2.4-based distro makes very little sense. Among a lot of other reasons, the 2.4 scheduler is AFAIK rather poorly suited for use with combined SMT & SMP.marcuri - Saturday, December 4, 2004 - link
as #5 methods are far from being anywhere remotely 'mature' as the article states. Actually, after talking to people who run databases, its close to useless. The benchmarks used test cpu speed, not memory or memory management. Opterons are a better solution because they are 64 bit and thus handel the management of 4+gigs of memory much better (it starts in excess of 1gig that 64 helps out i believe). By using a data set of 1 gig the benchmarks are completely missing whats important, its niave at best.Decoder - Saturday, December 4, 2004 - link
davesbeer:I've worked with Oracle 8i, 9, IBM UDB v7, SQL Server 6.5,7, 2K and 2005. I also do J2EE, ETL (Informatica, Datastage, DTS), .NET etc. $ for $, MS SQL Server 2005 is a much better value.
Have you ever built a data mart using MS SQL Server DTS? It's an ETL tool that comes with MS SQL server for free. Try building a data mart on Oracle and you will find yourself spending 100K US for an ETL tool. Also from a developers perspective, SQL Server tools are the best (Read up on SQL Profiler, Query Analyzer etc).
.NET not scalable? Scalable compared to what? J2EE EJB's?? Please. Sure there's a limit to scalability on 32-bit boxes, but with x86-64 and 64bit OS's, that limit will go away.
davesbeer: "SQL Server is for people who can't program real databases"
Program databases? Please elaborate. You code to a database, design a database, write procs for a database but what does program databases mean?
Also, tell that to the Finanical Services companies in New York.
davesbeer: "SQL Server will be delayed again... (you can remember I said that) and the technology is already outdated."
SQL Server 2005 will be extremely popular. Also SQL Server 2005 DTS (a complete rewrite) will penetrate the ETL market.
davesbeer - Saturday, December 4, 2004 - link
Decoder.I work with companies that have over 6000 processors actively running multiple databases in thier environments. MS just does'nt compete.. .Net is interesting but again you are stuck with M$ and scalabilty is just not there for real apps. Most financial industries run Sybase... its thier last bastion of full fledged support, I worked with 15 of the largest American Financial institutes, I know what I am taking about. I currently work with some of the largest IT departments in the world. Thousands of developers in a single company alone.. I know IT.
Oracle, DB2 etc's 64 bit knowledge is far superior to M$ and will remain that way for a long time.. they'll been writing 64bit code for years. SQL Server will be delayed again... (you can remember I said that) and the technology is already outdated. SQL Server is for people who can't program real databases (can feel the flames now) sorry Anandtech but your databases are lightweight... once they grow up you will switch to Oracle or DB2 etc..
at80eighty - Saturday, December 4, 2004 - link
First off: Welcome Johan : )next...im a frickin n00b, so pardon my ignorance : )
my questions may not be entirely on topic, but i hope someone...ANYONE can help...
im lookin to *build* an entry level server, so i've got the following doubts (based on the following factors) :
Company: startup...less than a year old
Expected # of initial users: 8 (2 people will be responsible for writing data, all employees will read db) requirements should go up to 15-20 people (2-4 write: all read) by year end (should business expand as planned)
Server: Mail + DB + File server in one
Cost: would be nice to keep things economical , but im willing to stretch for a middle ground : )
1) What CPU is best bang for buck?
2) OS will be Windows SBS Premium...was going for this since i gathered MS SQL is the best way to go (more so coz Anandtech uses SQL too :p )...after reading some of the comments im a little confused viz. DB2 v/s Oracle v/s MS SQL...plus Exchange looks kinda sweet too : )
3) Should i wait for SQL 2005? What benefits if any (with reference to costs) ?
4) Which is the best front-end in your opinion? i was looking to use Infopath as i found the GUI VERY easy to use and frankly the look is quite easy on the eyes too : p. Also, its important to note that I am *supposedly* the most computer literate in my company...so im looking to create a DB with an easy interface for my employees : p
5) Is Crystal Reports easy to integrate into MS SQL (or any of the above for that matter)? OR im lookin for a back-end that can export data to Excel, so that i can analyse & crunch numbers
6) Will i *need* an in-house IT guy to manage the DB? ...or even the server as a whole? (looking to keep costs to a minimum)
7)
i've got TONS of further queries (and i dont want to attention-whore on the thread :), so, should anyone have any spare time (and any inclination in the first place : ) to help a fellow AT'er, my mail add is: ucanmailme AT gmail DOT com
Thanks in advance
Cheers
'80'