Thread: Re: RFC: PostgreSQL and MySQL comparison.
Alexey Borzov writes: > I suppose PgSQL has to take a more active stance as well. Consider > "M$ vs Linux debates". Of course here both projects are Open Source so > the discussion should not be as heated... But I do think that > the statements in > http://www.mysql.com/doc/M/y/MySQL-PostgreSQL_features.html > should NOT go unanswered. Okay, I answered them: http://webmail.postgresql.org/~petere/comparison.html (The domain should move to www.ca.postgresql.org in the next few days. This is just a temporary location while the site meisters move things around.) I tried to be reasonable and biased at the same time. ;-) This article does not go into the advantages of PostgreSQL, since that's already done elsewhere, such as here: http://www.ca.postgresql.org/features.html -- Peter Eisentraut peter_e@gmx.net http://funkturm.homeip.net/~peter
On Wed, 29 Aug 2001, Peter Eisentraut wrote: > Okay, I answered them: > > http://webmail.postgresql.org/~petere/comparison.html Perhaps you should add some links to PostgreSQL's full-text searching solutions? It's hard to evaluate without some reference to what you're talking about. -sam
the following link is a good article on the real-world, multi-user performance of mysql and postgres. tim perdue ported source forge from mysql to postgre and performed a series of tests against the source forge code and database. one chart shows postgres scaling up to 100 concurrent users and serving pages very nicely while mysql craters at about 20. enjoy! rjsjr http://www.phpbuilder.com/columns/tim20001112.php3 > -----Original Message----- > From: pgsql-general-owner@postgresql.org > [mailto:pgsql-general-owner@postgresql.org]On Behalf Of Sam Tregar > Sent: Wednesday, August 29, 2001 3:01 PM > To: Peter Eisentraut > Cc: pgsql-general@postgresql.org > Subject: Re: [GENERAL] RFC: PostgreSQL and MySQL comparison. > > > On Wed, 29 Aug 2001, Peter Eisentraut wrote: > > > Okay, I answered them: > > > > http://webmail.postgresql.org/~petere/comparison.html > > Perhaps you should add some links to PostgreSQL's full-text > searching > solutions? It's hard to evaluate without some reference to > what you're > talking about. > > -sam > > > > ---------------------------(end of > broadcast)--------------------------- > TIP 2: you can get off all lists at once with the unregister command > (send "unregister YourEmailAddressHere" to > majordomo@postgresql.org) >
On Wed, 29 Aug 2001, Peter Eisentraut wrote: > Alexey Borzov writes: > > > I suppose PgSQL has to take a more active stance as well. Consider > > "M$ vs Linux debates". Of course here both projects are Open Source so > > the discussion should not be as heated... But I do think that > > the statements in > > http://www.mysql.com/doc/M/y/MySQL-PostgreSQL_features.html > > should NOT go unanswered. > > Okay, I answered them: > > http://webmail.postgresql.org/~petere/comparison.html Hmm... reading it suddenly I found that can't understand something. If table bigger than 1 GB are split in different files, there is no "one file per table". In that case, symlinking is not that safe... or is it? -- Alvaro Herrera (<alvherre[@]atentus.com>)
I think the quotes from Jan Wieck may be hard to understand, or even misleading, to a reader who did not see them in the context of the discussion in which they were written. They contain plenty of irony (sarcasm?), which when read literally, means something quite different from what was intended. Since the points made are valuable, perhaps they could be restated in the body of the text instead? ----- Original Message ----- From: "Sam Tregar" <sam@tregar.com> To: "Peter Eisentraut" <peter_e@gmx.net> Cc: <pgsql-general@postgresql.org> Sent: Wednesday, August 29, 2001 3:01 PM Subject: Re: [GENERAL] RFC: PostgreSQL and MySQL comparison. > On Wed, 29 Aug 2001, Peter Eisentraut wrote: > > > Okay, I answered them: > > > > http://webmail.postgresql.org/~petere/comparison.html > > Perhaps you should add some links to PostgreSQL's full-text searching > solutions? It's hard to evaluate without some reference to what you're > talking about. > > -sam > > > > ---------------------------(end of broadcast)--------------------------- > TIP 2: you can get off all lists at once with the unregister command > (send "unregister YourEmailAddressHere" to majordomo@postgresql.org) > >
> ... > ... mysql v postgres > ... In a book on marketing warfare (maybe the one by Ries & Trout) there was a question about who was the competitor: You own a hotel on a small desert island across the street from another hotel. The question is are they your real threat as a competitor and should you try to drive them down, maybe out of business? The real competitor is the other, next, nearest "island" 300 miles away - and not the hotel across the street. If either of the hotels go out, the whole island economy falters and you may go with it. What you really want is to compete against the other island. If the business of the hotel across the street goes up, then your own business is like to grow also. The issue in this case is who is the real, significant competitor. It reminds me about unix. Let see, is it Sun or HP that has the better unix. Opps, how about IBM, or Sony, NEC, FreeBSD, NetBSD, SCO, or ... and opps I forgot Linux, oh how could I do that? And it would be better to have start up run scripts where?, what is the better x-type windowing system?, ... But, MSDOS was MSDOS, was MSDOS... and MS Windows was MS Windows, is MS Windows So who is the real competitor - HP v Sun v Ibm v ... all unix and all slightly different in such miniscule amounts (I'll probably get flamed for that...) or unix v MS Windows... What would have happened if the unix vendors laid down their NIH and joined each other? So who is the real, important competitor? Is it so much important to "compete" between mysql and postgres, or is it more important to compete with the proprietary databases. They all must laugh themselves silly at times. But I tell you that there would be a WHOLE lot more concern if the HUGE amount of brains and talent on both these projects all laid down their editors, compilers, os's, and their deeply entrenched NIH attitude and join forces with each other. What would that be like? Is it possible? So you feel like competing, then compete against your own self to ask what you could learn from each other, then change direction and compete against your COMMON competitor - the proprietary databases. Just some rambling thoughts - and certainly not to try and tit-for-tat war. Regards, .. Otto
On Wed, 29 Aug 2001 21:32:59 +0200 (CEST), Peter Eisentraut wrote: >Okay, I answered them: > >http://webmail.postgresql.org/~petere/comparison.html Ok, sounds good. I hope that will calm down the discussion. regards Johann Zuschlag zuschlag@online.de
Bravo
Good points.
Most *nix systems are finaly working toward POSIX compliance.
Most SQL systems are working toward SQL92 {I think} compliance.
On the other hand just because you are compliant does not mean that your
code is portable. I don't want this point to turn into a thread, it just
isn't important enough.
There are even RDBMS systems that don't use SQL and that does not make
them bad, it just makes them different.
I used to write my databases in C using an algorythm I developed myself
that is simmilar to but before btree existed. My engine kept indexes in
a pointer of pointers type format in memory, that was quick and didn't
require a lot of memory. Those databases ran on an IBM PC-XT with 512 kB
ram and a 10 MB HD and easily handled 64K interlinked records with an
average of 8 probes per query. Oh and by the way that was on dos 3.2{if
my memory serves me correctly} with Borland Turbo C, but I forget wich
memory model I used {It has been 6 years since I did any non *nix
programming}.
Guy Fraser
Otto Hirr wrote:
>
> > ...
> > ... mysql v postgres
> > ...
>
> In a book on marketing warfare (maybe the one
> by Ries & Trout) there was a question about who
> was the competitor:
>
> You own a hotel on a small desert island
...snip...
> who is the real, significant competitor.
>
> It reminds me about unix. Let see, is it
> Sun or HP that has the better unix.
...snip...
> unix v MS Windows...
> What would have happened if the unix vendors
> laid down their NIH and joined each other?
>
> So who is the real, important competitor?
> Is it so much important to "compete"
> between mysql and postgres, or is it more
> important to compete with the proprietary
> databases. They all must laugh themselves
> silly at times.
>
> But I tell you that there would be a WHOLE
> lot more concern if the HUGE amount of brains
> and talent on both these projects all laid
> down their editors, compilers, os's, and their
> deeply entrenched NIH attitude and join
> forces with each other. What would that be
> like? Is it possible? So you feel like
> competing, then compete against your own
> self to ask what you could learn from each
> other, then change direction and compete
> against your COMMON competitor - the
> proprietary databases.
>
> Just some rambling thoughts - and certainly
> not to try and tit-for-tat war.
>
> Regards,
>
> .. Otto
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
> (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
--
There is a fine line between genius and lunacy, fear not, walk the
line with pride. Not all things will end up as you wanted, but you
will certainly discover things the meek and timid will miss out on.
On Wed, Aug 29, 2001 at 09:32:59PM +0200, Peter Eisentraut wrote: > Alexey Borzov writes: > Okay, I answered them: > > http://webmail.postgresql.org/~petere/comparison.html In that document you say: > By the way, PostgreSQL does support multiple storage managers; not as > easily today as in the early days, but that is mostly due to the fact that > no one ever wanted to replace the current one. Does this mean that in the 7.1 tree one could include the storage manager for 7.0, thus being able to read both types, or is it more complicated than that? -- Martijn van Oosterhout <kleptog@svana.org> http://svana.org/kleptog/ > It would be nice if someone came up with a certification system that > actually separated those who can barely regurgitate what they crammed over > the last few weeks from those who command secret ninja networking powers.
Greetings, Peter! At 29.08.2001, 23:32, you wrote: >> But I do think that >> the statements in >> http://www.mysql.com/doc/M/y/MySQL-PostgreSQL_features.html >> should NOT go unanswered. PE> Okay, I answered them: PE> http://webmail.postgresql.org/~petere/comparison.html Looks good, but it's not exactly what I've had in mind. You see it is too often considered that if someone has to justify oneself, he is automatically guilty. I want to write an article that would make MySQL's developers justify themselves... Besides, one has to read the MySQL's article *before* this just to understand what's this all about... BTW, can I use parts of your answers in my work? And, a question to SQL standard gurus: is it necessary for a DBMS to implement subqueries and views to be considered entry-level SQL92 compliant. Besides, is there any document which has all PostgreSQL's deviations from standard in one place? PE> I tried to be reasonable and biased at the same time. ;-) PE> This article does not go into the advantages of PostgreSQL, since that's PE> already done elsewhere, such as here: PE> http://www.ca.postgresql.org/features.html -- Yours, Alexey V. Borzov, Webmaster of RDW.ru
On Wed, 29 Aug 2001, Robert J. Sanford, Jr. wrote: > http://www.phpbuilder.com/columns/tim20001112.php3 Now *that* was very informative, thank you. The best benefit to this, is that the optimization engine is supposedly vastly improved in the 7.2 tree, so that'll just increase the lead. If they clean up vacuum to actually get indexes, the planner will have a better chance at picking more optimal execution plans, too. I'm glad that development has picked up. It seemed like 6.5x would be around forever. Thanks for knocking down the walls guys. ^_^ -- +-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+ | Shaun M. Thomas INN Database Programmer | | Phone: (309) 743-0812 Fax : (309) 743-0830 | | Email: sthomas@townnews.com AIM : trifthen | | Web : hamster.lee.net | | | | "Most of our lives are about proving something, either to | | ourselves or to someone else." | | -- Anonymous | +-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
Martijn van Oosterhout writes: > > By the way, PostgreSQL does support multiple storage managers; not as > > easily today as in the early days, but that is mostly due to the fact that > > no one ever wanted to replace the current one. > > Does this mean that in the 7.1 tree one could include the storage manager > for 7.0, thus being able to read both types, or is it more complicated than > that? The storage manager hasn't changed in principle since 6.5. What has changed is the system catalog structure and contents. (Recall the pg_upgrade program that was available for 6.5->7.0 transition. It would copy the data files right over.) -- Peter Eisentraut peter_e@gmx.net http://funkturm.homeip.net/~peter