Apache Today [Your Apache News Source]
Your Daily Source for Apache News and Information  
Breaking News Preferences Contribute Triggers Link Us Search About
To internet.com

Webopedia

Apache HTTPD Links
The Java Apache Project
The Jakarta Project
Apache Project
PHP Server Side Scripting
Apache Module Registry
The Apache FAQ
Apache-Perl Integration Project
The Apache Software Foundation
Apache-Related Projects
Apache XML Project
ApacheCon
The Linux Channel at internet.com
Linux Programming
BSD Today
PHPBuilder
Just Linux
Linuxnewbie.org
Enterprise Linux Today
Linux Apps
Linux Planet
BSD Central
Linux Central
Linux Start
All Linux Devices
Apache Today
Linux Today
SITE DESCRIPTIONS
PR: Postgres Routs Competition in New Benchmark Tests
Aug 14, 2000, 15 :46 UTC (53 Talkback[s]) (52831 reads)

PRESS RELEASE: "NORFOLK, Va, August 14, 2000--PostgreSQL, a leading open source database, routed the competition in recent benchmark testing, topping the proprietary database leaders in industry-standard transaction-processing tests. PostgreSQL, also known as 'Postgres,' is an object-relational database management system (DBMS) that newly formed Great Bridge LLC will professionally market, service and support. Postgres also consistently outperformed open source competitors, including MySQL and Interbase, in the benchmark tests. Great Bridge will market Postgres-based open source solutions as a highly reliable and lower cost option for businesses seeking an alternative to proprietary databases."

"On the ANSI SQL Standard Scalable And Portable (AS3AP) benchmark, a rudimentary information retrieval test that measures raw speed and scalability, Postgres performed an average of four to five times faster than every other database tested, including two major proprietary DBMS packages, the MySQL open source database, and Interbase, a formerly proprietary product which was recently made open source by Inprise/Borland."

"In the Transaction Processing Council's TPC-C test, which simulates a real-world online transaction processing (OLTP) environment, Postgres consistently matched the performance of the two leading proprietary database applications. The two industry leaders cannot be mentioned by name because their restrictive licensing agreements prohibit anyone who buys their closed source products from publishing their company names in benchmark testing results without the companies' prior approval."

"'The test results show that Postgres is a robust, well-built product that must be considered in the same category as enterprise-level competition,' said Robert Gilbert, Great Bridge President and CEO. 'Look at the trendlines in the AS3AP test: Postgres, like the proprietary leaders, kept a relatively consistent output level all the way up to 100 concurrent users--and that output was four to five times faster than the proprietary products. Interbase and MySQL fell apart under heavy usage. That's a strong affirmation that Postgres today is a viable alternative to the market-leading proprietary databases in terms of performance and scalability--and the clear leader among open source databases.'"

"The tests were conducted by Xperts Inc. of Richmond, Virginia, an independent technology solutions company, using Quest Software's Benchmark Factory application. Both the AS3AP and the TPC-C benchmarks simulated transactions by one to 100 simultaneous users in a client-server environment. One hundred concurrent users approximates the middle range of a traditional database user pool; many applications never see more than a few users on the system at any given time, while other more sophisticated enterprise platforms number concurrent users in the thousands. In a Web-based application, where the connection to the database is measured in milliseconds, 100 simultaneous users would represent a substantial load--the equivalent of 100 customers hitting the "submit" button on an order form at exactly the same time."

"The AS3AP test measures raw database data retrieval power, showing an application's scalability, portability and ease of use and interpretation through the use of simple ANSI standard SQL queries. The TPC-C test simulates a warehouse distribution system, including order creation, customer payments, order status checking, delivery, and inventory management."

"'What stood out for us was the consistent performance of Postgres, which stayed the same or tested better than those of the leading proprietary applications. Postgres performed consistently whether it was being used by one or 100 people,' said Richard Brosnahan, senior software developer at Xperts."

"Postgres is a standards-based object-relational SQL database designed for e-business and enterprise applications. The software is open source and freely owned, continuously augmented by a global collaborative community of elite programmers who volunteer their time and expertise to improve the product. In the last two years, with the introduction of versions 6.5 and 7.0 of the software, Postgres has seen rapid enhancement through a series of high-level refinements."

"'Postgres' performance is a powerful affirmation of the open source method of development,' said Gilbert of Great Bridge. 'Hundreds, even thousands, of open source developers work on this software, demonstrating a rate of innovation and improvement that the proprietary competition simply can't match. And it's only going to get better.'" "Xperts ran the benchmark tests on Compaq Proliant ML350 servers with 512 mb of RAM and two 18.2 Gb hard disks, equipped with Intel Pentium III processors and Red Hat Linux 6.1 and Windows NT operating systems. The company ensured the tests' consistency by using the same computers for each test, with each product connecting to the tests through its own preferred ODBC driver. While Benchmark Factory does provide native drivers for some commercial databases, using each product's own ODBC ensured the most valid 'apples to apples' comparison."

"In the AS3AP tests, PostgreSQL 7.0 significantly outperformed both the leading commercial and open source applications in speed and scalability. In the tested configuration, Postgres peaked at 1127.8 transactions per second with five users, and still processed at a steady rate of 1070.4 with 100 users. The proprietary leader also performed consistently, with a high of 314.15 transactions per second with eight users, which fell slightly to 288.37 transactions per second with 100 users. The other leading proprietary database also demonstrated consistency, running at 200.21 transactions per second with six users and 197.4 with 100."

"The other databases tested against the AS3AP benchmarks, open source competitors MySQL 3.22 and Interbase 6.0, demonstrated some speed with a low number of users but a distinct lack of scalability. MySQL reached a peak of 803.48 with two users, but its performance fell precipitously under the stress of additional users to a rate of 117.87 transactions per second with 100 users. Similarly, Interbase reached 424 transactions per second with four users, but its performance declined steadily with additional users, dropping off to 146.86 transactions per second with 100 users."

"'It's just astounding, and unexpected,' said Xperts' Brosnahan of Postgres' performance. 'I ran the test twice to make sure it was running right. Postgres is just a really powerful database.'"

"In the TPC-C tests, Postgres performed neck and neck with the two leading proprietary databases. The test simultaneously runs five different types of simulated transactions; the attached graph of test results shows steadily ascending intertwined lines representing all three databases, suggesting the applications scaled at comparable rates. With all five transactions running with 100 users, the three databases performed at a rate of slightly above five transactions per second."

"'The TPC-C is a challenging test with five transactions running at once while querying against the database and the stress of a growing number of users. It showed that all the databases we tested handle higher loads very well, the way they should," Brosnahan explained.'"

"Neither Interbase nor MySQL could be tested for TPC-C benchmarks. MySQL could not run the test because the application is not adequately compliant with minimal ANSI SQL standards set in 1992. Interbase 6.0, recently released as open source, does not have a stable ODBC driver yet; while Xperts was able to adapt the version 5 ODBC driver for the AS3AP tests, the TPC-C test would not run. 'With MySQL it's an inherent design issue. Interbase 6 should run the TPC-C test, and perhaps would with tweaking of the test's code,' said Brosnahan."

"Great Bridge's Gilbert attributes Postgres' high performance to a quality differential that comes from the open source development process; the source code for Postgres has been subjected to years of rigorous peer review by some of the best programmers in the world, many of whom use the product in their work environments. "Great Bridge believes that Postgres is by far the most robust open source database available. These tests provide strong affirmation of that belief," he said. The company intends to work with hardware vendors and other interested parties to continue larger-scale testing of Postgres and other leading open source technologies."

"Great Bridge LLC provides open source solutions powered by PostgreSQL, the world's most advanced open source database. Great Bridge delivers value-added open source software and support services based on PostgreSQL, empowering e-business builders with an enterprise-class database and tools at a fraction of the cost of closed, proprietary alternatives."

"Headquartered in Norfolk, Virginia, Great Bridge is a privately held company funded by Landmark Communications, Inc., the media company that also owns The Weather Channel, weather.com, and national and international interests in newspapers, broadcasting, electronic publishing, and interactive media."

  Current Newswire:
GroupIT site-content engine previewed

Lineo Availix Vertical Clustering 1.0 Ships

Apache Week issue 252 (22nd June 2001) is out

FreeOS: (Apache) Web server tutorial

PHP-Nuke 5.0 released

SECURITY: EnGarde Linux advisory: Apache directory listing vulnerability

ZDNet: Red Hat Tux 2.0 blows away Apache

O'Reilly: Industrial-Strength Webcasting with mod_mp3

Apache Module Registration: mod_gd

Expresso 3.1 open source application development framework unveiled

 Talkback(s) Name  Date
  I told you so...
HA HA! Nobody ever believes me when I tell them exactly
what this aritcle states. I told you so... HA HA!   
  Aug 14, 2000, 21:46:15
  The value of benchmarks...
Postgres is a good ORDBMS. Many people already knew this. Obviously, some people (usually called 'IT managers') are a bit leery to use an 'officially unsupported' (but in reality very well supported) piece of software for their business. This is where 'Great Bridge LLC' comes in, they offer commercial support for a free product. Also good, since it makes it possible for the mentioned 'IT manager' to use Postgres, and good software wants to be used...

But please, let's keep FAR, FAR away from the bechmark-pissing-contests which are so common amongst traditional vendors. This is a nice wake-up call for some people, but Postgres has evolved to its current state without having to prove its worth in benchmark rallies.   
  Aug 14, 2000, 21:47:56
  Errr...
Okay, there's only one way to say this, in the words of that immortal Wendy's commercial:

"Where's the beef?"

There is no concrete information posted here about what exactly was tested in terms of "proprietary databases," and where's the benchmark report?

If someone could provide a URL, it would be immenseley more useful.

--ZS   
  Aug 14, 2000, 21:51:11
  ODBC
What this press release says is that Postgres has a solid ODBC driver.   
  Aug 14, 2000, 22:09:08
  Postgres Testing
Probably the most provocative remark in the story was:

"The two industry leaders cannot be mentioned by name because their restrictive licensing agreements prohibit anyone who
buys their closed source products from publishing their
company names in benchmark testing Results without the companies'
prior approval."

Has it come to that? Can a bookseller prohibit the newspapers from publishing
book reviews that haven't been approved?

Well, being under no restraint myself, let me name two industry leaders,
and then people can either confirm or deny that my two names are the
ones meant:

Oracle
MS SQL Server

I await correction, confirmation, or a subpoena.   
  Aug 14, 2000, 22:13:25
  Testing methodology
This article is all snake oil. I'm disappointed that I read the whole thing and didn't learn anything about comparative DBMS architecture and implementations. The simply *is* no *how or why* to the results. What a useless piece of sales drivel! I feel no better equipped to decide on a favorite database backend.   
  Aug 14, 2000, 22:22:16
  Your article
Great to hear about postgresql. We are currently developing in a Linux/Apache/php/postgres environment for the lowest cost possible to the end user. All are working great!   
  Aug 14, 2000, 22:30:34
  So who were they?
So, who were these "leading proprietary databases"? FoxPro and MS Access? Sybase & Oracle? Informix?

So far, this article just looks like some fluff for PostgreSQL. I looked into PostgreSQL about a year ago and thought it would be ready for prime time in a year or two, so it might be right on schedule...but this article certainly doesn't help me know that.   
  Aug 14, 2000, 22:35:25
  commercial db
Which were the 2 commercial databases?
  
  Aug 14, 2000, 22:35:31
  leading commercial applications
could these leading commercial applications be Oracle, Informix and MS SQL 7.0??????   
  Aug 14, 2000, 22:49:21
  Nice press release.
Very pretty. Work a bit more on the formatting and I'll give you a B+.

One *minor* point. This press release contains no facts of any relevance.

We all know the MySQL is a very fast file I/O tool, and not a proper RDBMS. We all know that Postgres is a real database that still has certain 'issues' (Rowsize limitations being the nastiest from my perspective).

We also know that MS SQL Server 7 is reasonably good. Oracle is outstanding. DB/2 is nice. Sybase has it's points. Ingres is actually very good, but nobody uses it. InterBase is dead. Honestly.

If you want a real benchmark, provide a budget to a set of DBAs for each database, let them buy, configure and tune the balls of their database server, then have a shoot-out. After each round, let them reconfigure the database. After a few rounds of each test you will have a *meaningful* result if you don't count reliability. Database reliability is measured in months of uptime. Not hours, or days.

Your benchmark is clearly tainted.

I like Postgres. But if a client's business is at stake (and, a database often holds *all* the client's data), I'll recommend Oracle or DB/2. Because I'm biased, like you are. But Oracle works in thousands of big companies.   
  Aug 14, 2000, 23:08:35
  very interesting
ok
nice to know   
  Aug 14, 2000, 23:09:44
  Show us the actual benchmark...
This sounds too much like the Microsoft/Mindcraft benchmarking we all know and loathe.


Give us the URLs to the actual benchmark results, which detail the testing environment, etc..., so that we can make up our own minds. Otherwise, don't waste our time.   

  Aug 14, 2000, 23:13:21
  No Data?!?!
I don't find the results unbelievable but no data was provided to substantiate this claim. This renders this article utterly useless. It might as well have not been published for all the good it does.

Michael   
  Aug 14, 2000, 23:18:23
  Postgres Benchmark
The test simultaneously runs five different types of simulated transactions; the attached graph of test results shows steadily ascending intertwined lines representing all three databases, suggesting the applications scaled at
comparable rates.

What attached graph?   
  Aug 14, 2000, 23:44:05
  Insert performance
I notice that insert performance was not broken out. In informal tests here, PostGreSQL could insert around 400 records per second max, while MySQL inserted several thousand records per second. For most applications this is not a problem -- they read far more than they write. For mine it was.

-E
  
  Aug 15, 2000, 00:07:34
  More tests to confirm results
Ok, this is all just great and fine. Being a happy PostgreSQL-user this is good news, but all these test seem to be picked apart some days later, when another "unpartial" group tests the same products and reach a different conclusion. If somebody could test the same stuff with different software, under different conditions etc. and verify the results that would be assuring.

Comparisons to MS SQL 7.0, Oracle, Sybase etc. would also be of great interest to see how PostgreSQL compares to other closed solutions.   
  Aug 15, 2000, 01:36:18
  Where are the numbers?
At the risk of sounding redundant, I still have to say I'm shocked. Postgres has proven to be every good thing but fast for me. Actually, PG's speed is roughly equivalent to a clydesdale with its hind legs gnawed off ;-)

Let's see some real numbers and hard facts. Even people like me who actually want to believe this benchmark are going to have a hell of a time of it.   
  Aug 15, 2000, 01:38:49
  Benchmarketing Open Sourced!
Whoopee, Benchmarketing has been opensourced too!   
  Aug 15, 2000, 02:20:12
  So, isn't it good?
Well, I've never used MySQL and people always bothered me about it's "better performance". Now I'm happy 'cause I know I've made the right choice. Even if those results aren't true, I'm sure that Pgsql is the best Open Source DBMS, ever. (Well, even Apache isn't THAT FAST, but does anyone doubts about it's superiority?).

A thing that makes me unhappy is... I've noticed that PHP 4.0.1 and up ALWAYS add MySQL support (Man, I didn't asked configure to add it, and there's no --no-mysql option. Not even --please-no-mysql). I know what I was doing, so I make no mistakes. Man... I've seen such things somewhere...   
  Aug 15, 2000, 02:46:35
  The actual press release with images..
Is here.
http://www.greatbridge.com/news/p_081420001.html
  
  Aug 15, 2000, 02:52:03
  So...
I use mySql because it is faster and easier to use than all the others.

I run a small web site on linux/apache/mysql/perl and php.

I don't think that I will ever have more than 2 simultaneous connections. When I get to the point of 10,000 hits per hour I'll switch to a High availabiliy server and upgrade to solaris/oracle.

Until then I don't care.

-- Andy   
  Aug 15, 2000, 03:04:06
   Re: Insert performance
Yes Postgres is slower on inserts. It is because of the sanity checking it does on each transaction. You can turn it off though with the -F flag which speed it up considerably.


> I notice that insert performance was not broken out. In informal tests here, PostGreSQL could insert around 400 records per second max, while MySQL inserted several thousand records per second. For most applications this is not a problem -- they read far more than they write. For mine it was.
-E
  
  Aug 15, 2000, 03:26:52
   Re: Postgres Testing
Interesting speculation. I was guessing that the 2 unmentionables were Oracle and DB2 or Sybase. I'd vouch for Sybase, since DB2 is supposed to be such a high performance product. But then so is Oracle... Hmmm...   
  Aug 15, 2000, 03:45:15
   Re: Insert performance
Yeah, that's great except for one thing. With Postgres, at least you can rest assured that the records you insert will actually get into the database in a robust fashion. I'd rather wait for a transaction to ensure that all of my inserts work, than cross my fingers hoping that MySQL actually inserts them right.

But that's just me. Personally, you can take Postgres and MySQL and shove 'em up your butt, cuz you'll have to pry my Oracle bible out of my cold, dead hands.

Rat Dog Bird   
  Aug 15, 2000, 04:18:29
  PostgreSQL, Interbase, MySQL
First off, kudos to the PostgreSQL development team for having come so far so fast. It wasn't that long ago that PostgreSQL was an unstable collection of functions written by a miscellaneous group of Stonebraker's students, no two of whom seemed to have the same exactly notion of what a DBMS should be doing. At this point, it's emerging as a strong competitor in the business market.

Secondly, the two market leading DBMSs are Oracle and DB2, each with about 30% of the market. The last time I saw independent numbers, MS SQL Server was a distant third with 13-14% of the market.

Now for the criticism. The test was conducted in a way that was unfair to both Interbase and MySQL. In regard to Interbase, it's well known that you can use the IB 5 ODBC driver with IB 6, and it is also well known that it doesn't work especially well. There are commercially available (for money) ODBC drivers that work better. The testers said that they were using the DBMS's preferred ODBC drivers, but that certainly wasn't the case with IB 6. IB 6 should never have been tested with an ODBC driver that was never intended to be used with it and is known to give inferior performance when used with it. (When last heard from, Jim Starkey was developing an ODBC driver for IB 6. I assume it will be out soon.)

In addition, the last time I read anything about IB 6, the multi-threaded version hadn't been ported to Linux and there were no immediate plans to do so. Lacking a shared cache, the only way it can handle 100 concurrent users is by dividing the available memory into 100 tiny units. On other operating systems, it doesn't have this problem. The article says that the tests were conducted on both Linux and NT. Given the expected divergence in results for IB on the two operating systems, the results would seem more trustworthy if the results for both systems were published.

In regard to MySQL, those of us who have been exposed to it for more than a week are aware that it is a specialized DBMS. The underlying C libraries were developed for use with databases that had very large tables, were read very heavily, but had few updates and didn't require transactions. When the C libraries were combined with an SQL interface to create a DMBS for web applications, the same conditions held. The AS3AP test actually consists of three parts: an OLTP section, an IR (Information Retrieval) section, and a mixed section. Nobody expects MySQL to do well on the OLTP section. It uses table level locks and can't do simultaneous updates very rapidly. On the other hand, it is blindingly fast with reads. I read a posting by a developer who claimed to have developed a OLAP application with MySQL that handled 10,000 concurrent users. I have no way of confirming that claim independently, but I would expect MySQL to beat anything else at information retrieval, simply because it is a specialized DBMS that is designed for IR. If you only have to do a few things well, you can do them much better. It is unfair to publish only that part of the test that MySQL can't possibly do well on, while withholding the results from the test section where MySQL would probably shine.

And finally, the tests aren't intended to cover everything of interest to a DBMS user. For example, version 6 of PostgreSQL had a reputation for corrupting its indices periodically, for no apparent reason. It's too early to tell if this has been corrected in version 7. MySQL has a reputation for protecting its data, even when it crashes. The Swedish company that developed it seems to have taken the Volvo approach to safety: the machine crumples, but the occupants are safe. (This in not to say that MySQL is safe under all conditions. The out-of-the-box version lacks transactions, which is certainly a safety issue in many applications.)

In summary, the testers should (1) run the full AS3AP test and release all the results, (2) either ensure that each DBMS is tested with a compatible ODBC driver or omit the drivers entirely, and (3) publish the results for different operating systems when a given DBMS has significantly different versions for each operating system.   
  Aug 15, 2000, 04:33:21
  These results are bogus.
These results are bogus. You have to follow the TPC's rules to have valid results. The rules include: (1) having the results audited by an independent TPC certified auditor, (2) publishing a full disclosure report with all of the details of the test, (3) submitting the results to the TPC where they are subject to review and challenge. Look at the TPC's web site (www.tpc.org) for a long list of valid results from legitimate vendors.   
  Aug 15, 2000, 05:07:37
  No Names, No Data, No Information
This is just marketing hype. I may even agree with the conclusions
but there is nothing here that would convince me if I did not already
know. Postgres IS about as good as Oracle or DB2 in pure SQL use but
the big comercial guys have lots of add ons that Postges lacks. They
also cost BIG $$$. MySQL is very fast for simple querries as much as
10x faster then Postgres by my tests but yes breaks down as the tasks
get bigger. So, No News and no information content in this article.

  
  Aug 15, 2000, 05:22:54
  Postgres is best of...?
To article :

"PR: Postgres Routs Competition in New Benchmark Tests "

When is Postgres OpenSource system, then is necessary assure full Open
for the test condition of this three DB systems.

Every DB system has his specific adjustment in many configuration parameters
in DB system in aplication (test app) and in the Operating System.
Choosing / Setting of this parameters is very complicated and requires high
skill and very good collaboration specialist on DB with specialist on OS
and Aplication (Test).

Without clarification all this condition are the results of this test
very low verity.

Have you any resource for this conditions on Internet.
  
  Aug 15, 2000, 06:40:45
  Hype
I must say that this is too thick, even humorous :) It's like one of
those sales leaflets they push your hand in exhibitions. I mean it's a nice
article but I don't understand what it's got to do with benchmarking.

All the big databases involved are good in some tasks and not so great in some other tasks and I won't even bother to go there. Where is explained how this test is _exactly_ made? Where are the numbers?   
  Aug 15, 2000, 07:43:24
   Re: So...
Remember that these tests benchmark the ODBC performance of the DBs.
In most setups with MySQL you are using the MySQL native db and API, not
ODBC. So the benchmarkings are, in fact, a huge pile of rubbish.

"I've got a benchmark proving my 1990 Ford Escort is faster than a 1990 Porsche 911 when I run the test using leaded fuel."

Quju   
  Aug 15, 2000, 08:33:03
  No value
This is really a vague story. Why aren't the "2 commercial" DB's named in this story ? Ok, one of them is Oracle, the other one probably M$ SQL server, or IBM's DB2. I think this article has no value without concrete information, benchmarks, etc.

But, if it's true that postgress is faster than say, Oracle, that would be great. Now all we have to do is wait for 5 years till postgress is WIDELY supported with drivers (spare me the ODBC crap), and then we can all use an open source RDBMS !   
  Aug 15, 2000, 10:22:00
   Re: Postgres Testing
Oracle and Sybase?   
  Aug 15, 2000, 11:48:52
  RE; JDBC Drivers ?
Does Postgres have JDBC drivers available ?   
  Aug 15, 2000, 14:18:24
  TPC-C looks bogus
Two things about the TPC-C strike me as fishy:
1) legally the TPC-C doesn't allow reporting of results that aren't audited... You can't claim a benchmark is TPC-C unless you follow their proceedures or you are breaking their trademark or something.
2) They are bragging about TPC-C transactions exceeding "5 transactions per second"... but TPC-C measures transactions per minute, and even multiplying by 60, 300 transactions per minute is a woefully low score... most Pentium 3 results are say, above 25,000 tpmC on a real TPC-C benchmark.

I would view any of their TPC-C claims with extreme suspicion. And I'm not familiar with this AS3AP benchmark... is it a new one?   
  Aug 15, 2000, 14:27:00
  Ok, Sure... I Believe You!
Interbase falls about after 100 users? Hmm... Tell that to the nearly 300 users that bombard my InterBase database 24 hours, 5 days a week, and at least 100 users that continue to bombard my InterBase database 24 hours during the weekend. Aside from intention downing for added table/key maintenance, InterBase has only gone donw when I've had to Reboot our NT server, which unfortunately is once every couple of months, not longer than 6 months or bad things start to happen... But I wouldn't blame that on InterBase.
  
  Aug 15, 2000, 14:40:25
  Be calm
Hi everybody,
I'm not a kind of "guru", but I think it's not good to start a kind of "dist war". If I'd make a(n independent) test I'd write everything that needed to repeat it by anyone who wants to do it. I think this article should be good for marketing and not for engineers or DBA's. It's not a technical documentation in this form so I couldn't take it seriously. I think if anyone's (DBA) interested in a real test (s)he could perform it and (s)he could believe that (s)he see.
Gabor
ps:sorry but my English is not perfect   
  Aug 15, 2000, 14:54:33
   Re: Errr...
www.tpc.org   
  Aug 15, 2000, 16:14:55
   Re: Re: So...
> Remember that these tests benchmark the ODBC performance of the DBs. In most setups with MySQL you are using the MySQL native db and API, not ODBC.

I can think of many cases where it's totally undesirable to use native interfaces. Why should any developer/company want to lock itself into supporting a particular DBMS? ODBC or JDBC are much wiser interface choices.   
  Aug 15, 2000, 16:58:34
  Re: No value (or, Why don't they name names?)
Jasper wrote:


This is really a vague story. Why aren't the '2 commercial' DB's named in this story?


As stated in the press release, they aren't named because their licensing agreements prohibit publishing their names in benchmarks without their permission. My guess, like most others I'd bet, is MS SQL Server 7, and Oracle 8. If so, then Larry Ellison needs to get off his hiney and either quit talking the talk, or start walking the walk, by letting others publish Oracle's name in benchmarks.


I wonder if they could get away with 'a major RDBMS from a company based in Redmond, WA' and something similar for Oracle ('a major RDBMS from a company owned by Larry Ellison', say). -- Joe   

  Aug 15, 2000, 17:13:23
  sux
I use chat server (IP 166 MHZ / 64 RAM ) with MySQL. I have 10.000-30.000 query in hour. And MySQL ist slow ???   
  Aug 15, 2000, 19:46:47
  What dont you get?
After reading the various responses to this article, I have one question to all the people that commented with dissatisfaction:

What part of proprietary and non-disclosure don't you get?

These database companies consider thier database architecture proprietary and mostly secret...ie you buy the databse you cannot talk about it...after all someone might make a better product using thier trade secrets. Of course that might not be a solid belief as PostgreSQL has shown but it is part of thier mentality.

Having said that, it follows that since the testers had to buy the software they really can't tell us much more than this without being sued.

The simple fact that PostgreSQL has those numbers is enough to simply confirm what anyone familiar with OpenSource and the various databased available will tell you...PostgreSQL is a good solid database with excellent commercial technical support if you want it.

Randall   
  Aug 15, 2000, 20:30:52
   Re: sux
I doubt anyone is saying MySQL is slow, it's just that PostgreSQL is faster. You have to have an open mind. A lot may also depend on what you like. No body is telling you to quit using MySQL. If you're happy with it, use it. But try PostgreSQL and you may like it more. I like it because of ease of use. I tried MySQL and found it difficult compared to MySQL. It may all be a matter of preference.   
  Aug 15, 2000, 21:50:33
   Re: RE; JDBC Drivers ?
> Does Postgres have JDBC drivers available ?

Yes, they come in the PostgreSQL source tarball.

- rick   
  Aug 15, 2000, 21:55:36
  the databases identified
Well, tentatively identified, at any rate. The version numbers contained in the
full text of the benchmark configuration -- http://www.greatbridge.com/news/p_081620001.html -- are as follows:

What versions of the databases did you use?

PostgreSQL - 7.0 release version
Proprietary 1 - 8.1.5
Proprietary 2 - 7.0
MySQL - 3.22.32
Interbase - 6.0

My guess is that Proprietary 1 == Oracle, and Proprietary 2 == SQL Server. The latter guess is also bolstered by this sentence from earlier in the report:
"...Proprietary 2, which only runs on Windows NT".

I stress that these are only guesses, and you need to draw your own conclusions.   
  Aug 16, 2000, 10:03:55
  And the commercial DBs are...
The original article with graphs is at: http://www.greatbridge.com/news/p_081420001.html

For better or worse, they publish information on the versions they used, which provides enough information to completely identify the commercial DBs:

Proprietary 1 - 8.1.5 (based on version numbers, this is Oracle)
Proprietary 2 - 7.0 (based on version numbers (and that it
only was available for Windows), this
is MS SQL Server)
PostgreSQL - 7.0 release version
MySQL - 3.22.32
Interbase - 6.0

The numbers for TPC, as shown in http://www.greatbridge.com/img/tpc-c.gif, are not TPC-C type numbers, but the graph shows that PostgreSQL was tracking Oracle and MS-SQL very well. Obviously, since the curves are essentially straight lines, even at the 100 users level none of these DBs were being stressed. I presume they stopped at the 100 level because their test environment couldn't go higher.

A list of results from TPC-C is at: http://www.tpc.org/new_result/c-d-results.idc

My conclusion is that PostgreSQL is competitive with Oracle and MS-SQL at moderate loads, and we will have to test them in our environment to determine anything further. Since PostgresSQL supports transactions etc., we can (probably completely) automatically translate any PHP code that uses PostgreSQL into equivalent code that uses Oracle or Sybase if there turns out to be a problem.

../Dave   
  Aug 16, 2000, 14:50:46
  The Real World
I'm a web developper and I use database backends alot. For a mission critical application, I wouldn't use postgres if you put a gun to my head. MySQL is the only non-prop. DB server I trust. Period.   
  Aug 16, 2000, 17:00:20
  Don't you need a gigabit ethernet?
The "source" articles and explanations from the "benchmarking" company (somewhat indistinguishable from a body shop) reveal they're using a single 100 MB connection to the server, consising of a single PCI bus. Correct me if I'm wrong, but would it *really* matter which RDBMS performs better, since these operations are likely to be IO bound?
  
  Aug 16, 2000, 20:19:36
   Re: The Real World
> I'm a web developper and I use database backends alot. For a mission critical application, I wouldn't use postgres if you put a gun to my head. MySQL is the only non-prop. DB server I trust. Period.


You are a funny man. I like your sense of humor.   
  Aug 17, 2000, 17:17:25
   Re: Errr...
Well, they can't say. But anyone who knows what databases are out there, OR who the two richest men in the world are, can guess pretty easy.

Instead of Brand X, they would be known as Brand O, and Brand M. Owned by the two stooges Larry, and Bill (respectively). Both of which are guilty of the High Treason of Bloatware. I love that the brand O media pak contains 18 CDs. 'Here take all of this software... Oh, and here is our licensing scheme...' (ouch)   
  Aug 31, 2000, 15:12:20
   Re: These results are bogus.
The membership of the TPC organization suggests that it would not be completely
without bias -- specifically, none of the members would be expected to
look with particular favor upon open source software, since many if not
all are either vendors of COTS database products or consultants who may
be VARs or whose expertise pertains to commercial products. Which makes the
sourness of the "These results are bogus" comment somewhat suspect.
Methinks he doth protest too much.   
  Sep 2, 2000, 02:14:27
  Great!!!
The things are getting better....

My company is developing a telemarketing software, and we have chossen Postgresql as the db for it, we will have aprox. 100 online users and our db server is a Pentium III 600mhz with 18.9 Gb HD and 128 MB of ram, probably i will need to upgrade the ram, running on RedHat 6.2.

For us is great to have access to a very mature sql db engine, that is opensource and now is tested to be much more faster than Oracle and Microsoft products.

Please i would like to ask every body to see how good is having a company that could advertise the product and can introduce it to the world, look to the RedHat or similar companies, and what they did in terms of popularization of linux operating system, and how the corporations are looking diferent as 3 or 4 years ago to it.

I use linux from version 0.12, and in the last 7 years i have deployed a lot of corporate and mission critical systems running under linux, but we had to hide for th corporations that we were using opensource or free software, and the only problem was the lack of good database systems. For some projects we used Essentia from Intersoft and for others MySql, both were great, the first we had to pay for it, and the second is real fast in simple database operations and doesn't have support for transactions.

Now we can choose PostgreSql for all our db needs, and i hope that it grows up to the level of Apache or Linux in terms of users.

Having postgresql available, is like a gift and we would like to thank every body in the development team, as well as the people that will be promoting postgresql as the opensource comercial grade sql db system that we could use in our next mission critical application or bussiness automation were Linux and OpenSource software is really behind the oracle and microsoft solutions.

The only thing is missing for postgresql is a set of tools, like a good form tools, and report generators, to automate some of the most dificult application development. Maybe this is available somewere else, would be great to have them integrated and properly configured to use like an application development enviroment for postgresql.

I hope every body could be a little more positive about this news and not to worry about the format or the missed links or the way it was conducted the benchmarks, the things is we have a very good competitor for comercial db servers, that is what counts.

Juan.
OpenComm do Brasil Ltda.

  
  Sep 12, 2000, 12:28:07
  Interbase
Think that the tests are comparing things that are not relevant in many cases. A database server is much more like a set of possibilities, compared to our needs. If I want a DB which is fast an integrated for use with ASP pages on IIS, i will use SQL server. If I want to be sure that my database will survive until 2999, and I have a complex design to express,i use Oracle. If I need a relativelly fast, well documented, small, comfortable DB, with Delphi/CBuilder clients running on Windows, and the database on Linux, and my deadline is next monday, i will use Interbase.   
  Oct 21, 2000, 13:20:04
Enter your comments below.
Your Name: Your Email Address:


Subject: CC: [will also send this talkback to an E-Mail address]
Comments:

See our talkback-policy for or guidelines on talkback content.

About Triggers Newsletters Media Kit Security Triggers Login


All times are recorded in UTC.
Linux is a trademark of Linus Torvalds.
Powered by Linux 2.2.12, Apache 1.3.9. and PHP 3.14
Copyright INT Media Group, Incorporated All Rights Reserved.
Legal Notices,  Licensing, Reprints, & Permissions,  Privacy Policy.
http://www.internet.com/