Обсуждение: SAP DB: The unsung Open Source DB
Anyone seen this ad lately? ;) http://ads.osdn.com/?ad_id=83&alloc_id=1055&site_id=2&request_id=1792929&1059014249841 Wonder why SAP is so unsung ... Jean-Christian Imbeault
The unknown is only in North American, but not in Europe. That is what I learnt more than a year ago. It has some advantagessover PG from on-line information. It is quicker than PG for one. -- --------- Original Message --------- DATE: Thu, 24 Jul 2003 11:43:48 From: Jean-Christian Imbeault <jc@mega-bucks.co.jp> To: pgsql-general@postgresql.org Cc: >Anyone seen this ad lately? ;) > >http://ads.osdn.com/?ad_id=83&alloc_id=1055&site_id=2&request_id=1792929&1059014249841 > >Wonder why SAP is so unsung ... > >Jean-Christian Imbeault > > >---------------------------(end of broadcast)--------------------------- >TIP 5: Have you checked our extensive FAQ? > > http://www.postgresql.org/docs/faqs/FAQ.html > ____________________________________________________________ Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail! http://login.mail.lycos.com/r/referral?aid=27005
Vernon Smith wrote: > The unknown is only in North American, but not in Europe. > That is what I learnt more than a year ago. It has some > advantagess over PG from on-line information. It is > quicker than PG for one. No. PG is 2.5 times faster. See how easy it is for people to make unsubstantiated claims? Mike Mascari mascarm@mascari.com >>Wonder why SAP is so unsung ... >> >>Jean-Christian Imbeault >>
OPEN magazine has an interview with the head of SAP DB development, and talks quite a bit about the MySQL strategy: http://www.open-mag.com/8422483279.shtml ----- Original Message ----- From: "Jean-Christian Imbeault" <jc@mega-bucks.co.jp> To: <pgsql-general@postgresql.org> Sent: Wednesday, July 23, 2003 10:43 PM Subject: [GENERAL] SAP DB: The unsung Open Source DB > Anyone seen this ad lately? ;) > > http://ads.osdn.com/?ad_id=83&alloc_id=1055&site_id=2&request_id=1792929&1059014249841 > > Wonder why SAP is so unsung ... > > Jean-Christian Imbeault > > > ---------------------------(end of broadcast)--------------------------- > TIP 5: Have you checked our extensive FAQ? > > http://www.postgresql.org/docs/faqs/FAQ.html > >
"Vernon Smith" <vwu98034@lycos.com> writes:
> The unknown is only in North American, but not in Europe. That is what
> I learnt more than a year ago. It has some advantagess over PG from
> on-line information. It is quicker than PG for one.
The one PG-vs-SAP benchmark that I've seen came to the opposite conclusion.
No doubt each has cases where it is quicker ... but if you don't specify
what you are measuring, there's not a lot of content in such statements.
regards, tom lane
On Thursday 24 July 2003 04:42, Mike Mascari wrote: > Vernon Smith wrote: > > The unknown is only in North American, but not in Europe. > > That is what I learnt more than a year ago. It has some > > advantagess over PG from on-line information. It is > > quicker than PG for one. > > No. PG is 2.5 times faster. > > See how easy it is for people to make unsubstantiated claims? Be fair Mike, it's actually 2.487 times faster Unsubstantiated claims should be implausibly accurate too ;-) -- Richard Huxton Archonet Ltd
On Thursday 24 July 2003 05:00, Ned Lilly wrote: > OPEN magazine has an interview with the head of SAP DB development, and > talks quite a bit about the MySQL strategy: > > http://www.open-mag.com/8422483279.shtml Interesting, but I'm not sure it's cleared anything up in my mind. "What is significant about the MySQL/SAP deal is that the two companies’ strategic mix of strengths makes market growth, when it does start to happen, pretty much inevitable" Eh? Once X has happened, X is inevitable? SAP AG still own and will support SAP DB (fair enough) but MySQL will have commercial rights and will rebrand it. So I can buy SAP from MySQL but they're not going to do the development on it, SAP AG will (but I can't buy it from them). Presumably the support for MySQL's customers will be via SAP's team. There's a multi-year plan to "bring the code bases closer together" which sounds like one of those big projects that always make me nervous. The main thrust seems to be: 1. MySQL have a simple DB with a lot of users 2. SAP have a complex DB with few users 3. Let's bring the two together and get the best of both worlds! That's fine, but my understanding of SAP DB's failure to attract a large community was that: - it had a lot of competition (MySQL/PG/Firebird...) - it was tricky to compile/install - the codebase was far from easy to get to grips with I'm not clear how MySQL are better equipped to solve those problems than SAP AG. Actually, I'm not clear that they're going to if SAP AG are going to handle development. Maybe it's me, but other than a marketing announcement, I don't get this. -- Richard Huxton Archonet Ltd
I think your assessment is a pretty fair one. It's definitely an announcement driven by marketing, and beating the drumof MySQL's wide (but perhaps not too deep) installed base. Two points I think the story missed: 1) As has been discussed on this list before, it's far from clear to me that MySQL is a good model of open source development. Theirs is largely a community of users, not code contributors, and we've heard lots of stories about patchesbeing rejected or substantially rewritten. By contrast, I've always been impressed with how the PG community managescomplexity, and how hackers can earn their way from bug reports, into minor peripheral hacks, into their first realTODO item, into major functionality enhancements. 2) As you mentioned, the SAP DB codebase - while very sophisticated in terms of functionality - is said to be a bit of amess, a major reason it had trouble attracting open source developers. So you've got a company with a product they control very tightly, jointly developing a multi-year, next-gen product withanother company that controls its product tightly. I think the chances are reasonably good that with time and patientinvestors, they'll be able to come up with a good new product. And if they pursue a dual-licensing strategy likethey have today, they'll probably have a pretty good crop of users. But I'm very skeptical that they'll ever realize the level of developer contributions that Postgres has today. That's the"X factor" in successful open source projects - an ever-increasing level of code review, fixes, and enhancements froma highly skilled, self-selecting group of experts. That's what enables Linux to outpace Windows, despite the fact thatMicrosoft is sitting on $40B+ in cash. And that's why I'm still bullish on Postgres to stay ahead of MySQL in termsof features/functionality, and to continue closing the gap with Oracle. Where Postgres continues to be vulnerable,as we've seen from this and other recent press coverage of MySQL (front page of the Wall Street Journal for goodness'sake) - is in the area of marketing, and specifically the lack of a corporate sponsor of a certain size and stature. But that's a topic for another list ;-) Cheers, Ned ----- Original Message ----- From: "Richard Huxton" <dev@archonet.com> To: "Ned Lilly" <ned@nedscape.com>; <pgsql-general@postgresql.org> Sent: Thursday, July 24, 2003 6:22 AM Subject: Re: [GENERAL] SAP DB: The unsung Open Source DB On Thursday 24 July 2003 05:00, Ned Lilly wrote: > OPEN magazine has an interview with the head of SAP DB development, and > talks quite a bit about the MySQL strategy: > > http://www.open-mag.com/8422483279.shtml Interesting, but I'm not sure it's cleared anything up in my mind. "What is significant about the MySQL/SAP deal is that the two companies’ strategic mix of strengths makes market growth, when it does start to happen, pretty much inevitable" Eh? Once X has happened, X is inevitable? SAP AG still own and will support SAP DB (fair enough) but MySQL will have commercial rights and will rebrand it. So I can buy SAP from MySQL but they're not going to do the development on it, SAP AG will (but I can't buy it from them). Presumably the support for MySQL's customers will be via SAP's team. There's a multi-year plan to "bring the code bases closer together" which sounds like one of those big projects that always make me nervous. The main thrust seems to be: 1. MySQL have a simple DB with a lot of users 2. SAP have a complex DB with few users 3. Let's bring the two together and get the best of both worlds! That's fine, but my understanding of SAP DB's failure to attract a large community was that: - it had a lot of competition (MySQL/PG/Firebird...) - it was tricky to compile/install - the codebase was far from easy to get to grips with I'm not clear how MySQL are better equipped to solve those problems than SAP AG. Actually, I'm not clear that they're going to if SAP AG are going to handle development. Maybe it's me, but other than a marketing announcement, I don't get this. -- Richard Huxton Archonet Ltd
Richard Huxton <dev@archonet.com> writes:
> There's a multi-year plan to "bring the code bases closer together" which
> sounds like one of those big projects that always make me nervous.
Just between us chickens, I hope they do spend multiple man-years trying
to merge those two codebases. It'll keep them distracted from
accomplishing anything useful ;-).
I only went to a couple of the MySQL talks at O'Reilly earlier this
month, but I could not help noticing that not one word was said about
SAP in those talks, while plenty was said about their development plans
for the MySQL codebase. Whatever the marketing guys may think, I
believe Monty et al intend to press straight ahead with improving their
own code.
regards, tom lane
I've never seen this work before. the only version of this scenario that works, is to buy the competitor, and learn
theircode, put the competitor out of business, and force the competitor's company to use a half ass bridge version for
onerev of a vision, then the next version force them to your supposedly improved flagship product. Can anyone say:
Microsoft?
Oracle?
Richard Huxton wrote:
> On Thursday 24 July 2003 05:00, Ned Lilly wrote:
>
>>OPEN magazine has an interview with the head of SAP DB development, and
>>talks quite a bit about the MySQL strategy:
>>
>>http://www.open-mag.com/8422483279.shtml
>
>
> Interesting, but I'm not sure it's cleared anything up in my mind.
>
> "What is significant about the MySQL/SAP deal is that the two companies’
> strategic mix of strengths makes market growth, when it does start to happen,
> pretty much inevitable"
>
> Eh? Once X has happened, X is inevitable?
>
> SAP AG still own and will support SAP DB (fair enough) but MySQL will have
> commercial rights and will rebrand it. So I can buy SAP from MySQL but
> they're not going to do the development on it, SAP AG will (but I can't buy
> it from them). Presumably the support for MySQL's customers will be via SAP's
> team.
>
> There's a multi-year plan to "bring the code bases closer together" which
> sounds like one of those big projects that always make me nervous.
>
> The main thrust seems to be:
> 1. MySQL have a simple DB with a lot of users
> 2. SAP have a complex DB with few users
> 3. Let's bring the two together and get the best of both worlds!
>
> That's fine, but my understanding of SAP DB's failure to attract a large
> community was that:
> - it had a lot of competition (MySQL/PG/Firebird...)
> - it was tricky to compile/install
> - the codebase was far from easy to get to grips with
>
> I'm not clear how MySQL are better equipped to solve those problems than SAP
> AG. Actually, I'm not clear that they're going to if SAP AG are going to
> handle development.
>
> Maybe it's me, but other than a marketing announcement, I don't get this.
Couldn't agree more: <tom wrote> Just between us chickens, I hope they do spend multiple man-years trying to merge those two codebases. It'll keep them distracted from accomplishing anything useful . </tom wrote> Tom Lane wrote: > Richard Huxton <dev@archonet.com> writes: > >>There's a multi-year plan to "bring the code bases closer together" which >>sounds like one of those big projects that always make me nervous. > > > Just between us chickens, I hope they do spend multiple man-years trying > to merge those two codebases. It'll keep them distracted from > accomplishing anything useful ;-). > > I only went to a couple of the MySQL talks at O'Reilly earlier this > month, but I could not help noticing that not one word was said about > SAP in those talks, while plenty was said about their development plans > for the MySQL codebase. Whatever the marketing guys may think, I > believe Monty et al intend to press straight ahead with improving their > own code. > > regards, tom lane > > ---------------------------(end of broadcast)--------------------------- > TIP 7: don't forget to increase your free space map settings >
And in documentation and web support.
1/ It's hard to find a good search engine that's PG centric. And sorry, Google only works so well when doing a
search for highly specific technical information.
2/ Postgres documentation does not seem to be in any of the standard, open source formats. Take a look at the
Apache documentation for an example.
Ned Lilly wrote:
> I think your assessment is a pretty fair one. It's definitely an announcement driven by marketing, and beating the
drumof MySQL's wide (but perhaps not too deep) installed base.
>
> Two points I think the story missed:
>
> 1) As has been discussed on this list before, it's far from clear to me that MySQL is a good model of open source
development. Theirs is largely a community of users, not code contributors, and we've heard lots of stories about
patchesbeing rejected or substantially rewritten. By contrast, I've always been impressed with how the PG community
managescomplexity, and how hackers can earn their way from bug reports, into minor peripheral hacks, into their first
realTODO item, into major functionality enhancements.
>
> 2) As you mentioned, the SAP DB codebase - while very sophisticated in terms of functionality - is said to be a bit
ofa mess, a major reason it had trouble attracting open source developers.
>
> So you've got a company with a product they control very tightly, jointly developing a multi-year, next-gen product
withanother company that controls its product tightly. I think the chances are reasonably good that with time and
patientinvestors, they'll be able to come up with a good new product. And if they pursue a dual-licensing strategy
likethey have today, they'll probably have a pretty good crop of users.
>
> But I'm very skeptical that they'll ever realize the level of developer contributions that Postgres has today.
That'sthe "X factor" in successful open source projects - an ever-increasing level of code review, fixes, and
enhancementsfrom a highly skilled, self-selecting group of experts. That's what enables Linux to outpace Windows,
despitethe fact that Microsoft is sitting on $40B+ in cash. And that's why I'm still bullish on Postgres to stay ahead
ofMySQL in terms of features/functionality, and to continue closing the gap with Oracle. Where Postgres continues to
bevulnerable, as we've seen from this and other recent press coverage of MySQL (front page of the Wall Street Journal
forgoodness' sake) - is in the area of marketing, and specifically the lack of a corporate sponsor of a certain size
andstature. But that's a topic for another list ;-)
>
> Cheers,
> Ned
>
>
> ----- Original Message -----
> From: "Richard Huxton" <dev@archonet.com>
> To: "Ned Lilly" <ned@nedscape.com>; <pgsql-general@postgresql.org>
> Sent: Thursday, July 24, 2003 6:22 AM
> Subject: Re: [GENERAL] SAP DB: The unsung Open Source DB
>
>
> On Thursday 24 July 2003 05:00, Ned Lilly wrote:
>
>>OPEN magazine has an interview with the head of SAP DB development, and
>>talks quite a bit about the MySQL strategy:
>>
>>http://www.open-mag.com/8422483279.shtml
>
>
> Interesting, but I'm not sure it's cleared anything up in my mind.
>
> "What is significant about the MySQL/SAP deal is that the two companies’
> strategic mix of strengths makes market growth, when it does start to happen,
> pretty much inevitable"
>
> Eh? Once X has happened, X is inevitable?
>
> SAP AG still own and will support SAP DB (fair enough) but MySQL will have
> commercial rights and will rebrand it. So I can buy SAP from MySQL but
> they're not going to do the development on it, SAP AG will (but I can't buy
> it from them). Presumably the support for MySQL's customers will be via SAP's
> team.
>
> There's a multi-year plan to "bring the code bases closer together" which
> sounds like one of those big projects that always make me nervous.
>
> The main thrust seems to be:
> 1. MySQL have a simple DB with a lot of users
> 2. SAP have a complex DB with few users
> 3. Let's bring the two together and get the best of both worlds!
>
> That's fine, but my understanding of SAP DB's failure to attract a large
> community was that:
> - it had a lot of competition (MySQL/PG/Firebird...)
> - it was tricky to compile/install
> - the codebase was far from easy to get to grips with
>
> I'm not clear how MySQL are better equipped to solve those problems than SAP
> AG. Actually, I'm not clear that they're going to if SAP AG are going to
> handle development.
>
> Maybe it's me, but other than a marketing announcement, I don't get this.
The following statements is based on your own experience or not? >That's fine, but my understanding of SAP DB's failure to attract a large >community was that: > - it had a lot of competition (MySQL/PG/Firebird...) It is true that there are quite some open source free DBs out there. They, however, are for different people, or types ofapplications. If you are a PG guy, you may not be too keen on MySQL. > - it was tricky to compile/install What I know is the opposite. SAP DB is very easy to install, at least on Windows. > - the codebase was far from easy to get to grips with Since I don't know this, I won't say anything about it. ____________________________________________________________ Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail! http://login.mail.lycos.com/r/referral?aid=27005
Good to learn PG, in fact, is faster than SAP DB. But how the figure 2.487 comes up? A single select statement or somethingelse? --------- Original Message --------- DATE: Thu, 24 Jul 2003 11:00:30 From: Richard Huxton <dev@archonet.com> To: Mike Mascari <mascarm@mascari.com>, vwu98034@lycos.com Cc: pgsql-general@postgresql.org,Jean-Christian Imbeault <jc@mega-bucks.co.jp> >On Thursday 24 July 2003 04:42, Mike Mascari wrote: >> Vernon Smith wrote: >> > The unknown is only in North American, but not in Europe. >> > That is what I learnt more than a year ago. It has some >> > advantagess over PG from on-line information. It is >> > quicker than PG for one. >> >> No. PG is 2.5 times faster. >> >> See how easy it is for people to make unsubstantiated claims? > >Be fair Mike, it's actually 2.487 times faster > >Unsubstantiated claims should be implausibly accurate too ;-) > >-- > Richard Huxton > Archonet Ltd > >---------------------------(end of broadcast)--------------------------- >TIP 9: the planner will ignore your desire to choose an index scan if your > joining column's datatypes do not match > ____________________________________________________________ Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail! http://login.mail.lycos.com/r/referral?aid=27005
Nah, they were just pulling your leg, making the point that baseless assertions mean nothing. In fact, SAPDB and Postgresql are probably both reasonable fast, and both probably have corner cases where one is a clear winner over the other, but neither is likely to just win every race. On Thu, 24 Jul 2003, Vernon Smith wrote: > Good to learn PG, in fact, is faster than SAP DB. But how the figure 2.487 comes up? A single select statement or somethingelse? > > --------- Original Message --------- > > DATE: Thu, 24 Jul 2003 11:00:30 > From: Richard Huxton <dev@archonet.com> > To: Mike Mascari <mascarm@mascari.com>, vwu98034@lycos.com > Cc: pgsql-general@postgresql.org,Jean-Christian Imbeault <jc@mega-bucks.co.jp> > > >On Thursday 24 July 2003 04:42, Mike Mascari wrote: > >> Vernon Smith wrote: > >> > The unknown is only in North American, but not in Europe. > >> > That is what I learnt more than a year ago. It has some > >> > advantagess over PG from on-line information. It is > >> > quicker than PG for one. > >> > >> No. PG is 2.5 times faster. > >> > >> See how easy it is for people to make unsubstantiated claims? > > > >Be fair Mike, it's actually 2.487 times faster > > > >Unsubstantiated claims should be implausibly accurate too ;-) > > > >-- > > Richard Huxton > > Archonet Ltd > > > >---------------------------(end of broadcast)--------------------------- > >TIP 9: the planner will ignore your desire to choose an index scan if your > > joining column's datatypes do not match > > > > > > ____________________________________________________________ > Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail! > http://login.mail.lycos.com/r/referral?aid=27005 > > ---------------------------(end of broadcast)--------------------------- > TIP 9: the planner will ignore your desire to choose an index scan if your > joining column's datatypes do not match >
> What I know is the opposite. SAP DB is very easy to install, at least on Windows. I had a terrible time with it. Although the software installed fine, trying to get an actual DB up and running was a total nightmare. Maybe they've improved since last December, but I had tons of trouble. I tried to use their "demo setup" as a template to set up another database instance, but was unsuccessful. Jon
> Good to learn PG, in fact, is faster than SAP DB. But how the figure 2.487 comes up? A single select statement or somethingelse? I think it was just a more-exact unsubstantiated claim. Jon > > --------- Original Message --------- > > DATE: Thu, 24 Jul 2003 11:00:30 > From: Richard Huxton <dev@archonet.com> > To: Mike Mascari <mascarm@mascari.com>, vwu98034@lycos.com > Cc: pgsql-general@postgresql.org,Jean-Christian Imbeault <jc@mega-bucks.co.jp> > > >On Thursday 24 July 2003 04:42, Mike Mascari wrote: > >> Vernon Smith wrote: > >> > The unknown is only in North American, but not in Europe. > >> > That is what I learnt more than a year ago. It has some > >> > advantagess over PG from on-line information. It is > >> > quicker than PG for one. > >> > >> No. PG is 2.5 times faster. > >> > >> See how easy it is for people to make unsubstantiated claims? > > > >Be fair Mike, it's actually 2.487 times faster > > > >Unsubstantiated claims should be implausibly accurate too ;-) > > > >-- > > Richard Huxton > > Archonet Ltd > > > >---------------------------(end of broadcast)--------------------------- > >TIP 9: the planner will ignore your desire to choose an index scan if your > > joining column's datatypes do not match > > > > > > ____________________________________________________________ > Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail! > http://login.mail.lycos.com/r/referral?aid=27005 > > ---------------------------(end of broadcast)--------------------------- > TIP 9: the planner will ignore your desire to choose an index scan if your > joining column's datatypes do not match >
On Thursday 24 July 2003 19:12, Jonathan Bartlett wrote: > > Good to learn PG, in fact, is faster than SAP DB. But how the figure > > 2.487 comes up? A single select statement or something else? > > I think it was just a more-exact unsubstantiated claim. I like to think of it as the *definitive* unsubstantiated claim ;-) Incidentally, I'm available for government work. -- Richard Huxton Archonet Ltd
On Thu, 2003-07-24 at 10:32, Dennis Gearon wrote: > I've never seen this work before. the only version of this scenario > that works, is to buy the competitor, and learn their code, put the > competitor out of business, and force the competitor's company to > use a half ass bridge version for one rev of a vision, then the > next version force them to your supposedly improved flagship product. > Can anyone say: > > Microsoft? > Oracle? The exception that proves the rule: http://www.oracle.com/peoplesoft/Rdb_CaseStudyE.pdf While O would love us to migrate to 9i, there has been no pressure, and their engineering and tech support is still top quality. -- +-----------------------------------------------------------------+ | Ron Johnson, Jr. Home: ron.l.johnson@cox.net | | Jefferson, LA USA | | | | "I'm not a vegetarian because I love animals, I'm a vegetarian | | because I hate vegetables!" | | unknown | +-----------------------------------------------------------------+
DON'T make the mistake I just made, and install:
Adobe Acrobat Reader 6.0 ( on WindowsXP pro, at least ).
It has some wierd dependency on a piece of software that is only used on Microsoft TabletXP Operating system and just
keepspopping up multiple popups to try and install that software. For more on it, do a google on:
'microsoft journal viewer error adobe'
Ron Johnson wrote:
> On Thu, 2003-07-24 at 10:32, Dennis Gearon wrote:
>
>>I've never seen this work before. the only version of this scenario
>>that works, is to buy the competitor, and learn their code, put the
>>competitor out of business, and force the competitor's company to
>>use a half ass bridge version for one rev of a vision, then the
>>next version force them to your supposedly improved flagship product.
>>Can anyone say:
>>
>> Microsoft?
>> Oracle?
>
>
> The exception that proves the rule:
> http://www.oracle.com/peoplesoft/Rdb_CaseStudyE.pdf
> While O would love us to migrate to 9i, there has been no pressure,
> and their engineering and tech support is still top quality.
>
On 24 Jul 2003 at 10:00, Tom Lane wrote: > Richard Huxton <dev@archonet.com> writes: > > There's a multi-year plan to "bring the code bases closer together" which > > sounds like one of those big projects that always make me nervous. > > Just between us chickens, I hope they do spend multiple man-years trying > to merge those two codebases. It'll keep them distracted from > accomplishing anything useful ;-). IMO bigger problem mysql have is the attitude of being simple and doing things their own way. Whether or not they merge with SAP, they need to get rid of that attitude to move furthther. And once they git rid of that attitude, their success so far will become their enemy for keeping backword compatibility, simplicity and still achieve standard compliant, all over performing feature rich database. Open source or commercial, I doubt if it would be anytime sooner that mysql catches postgresql at least in terms of being on same scale.. Just a thought.. Bye Shridhar -- Mix's Law: There is nothing more permanent than a temporary building. There is nothing more permanent than a temporary tax.