Thread: Press Release review comments
Just reading the Press Release for 8.3 http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/press/pr/releases/8.3/en/release.txt Few thoughts spring to mind - Just in time background writer changes are nice, but since we already mention the load distributed checkpoints (a closely related feature), it seems we are saying too much at the expense of other features. - Asynch Commit is a seriously important feature, not necessarily for existing users but definitely for people used to MySQL's speed and lack of robustness with myisam tables. It's an important message that we can get the speed of myisam and at the same time the robustness of Postgres. I don't suggest that we specifically mention mysql, but it is a feature that will alter many people's perception of Postgres, so we should talk about it somewhere. So I suggest we swap Just-in-time for Async Commit. (To forestall any negative comment, I would add: Yes, I did write that code, but I wrote it *because* of its importance to a wide range of people that might now consider PostgreSQL. No, I'm not asking for my name in brackets anywhere.) Lastly, the press contacts listed are all on the North American continent and within a few hours of each other. I don't think we gain much by having a 3rd contact from that part of the world. There are English speaking countries all around the world in almost every timezone, obviously England, but also South Africa, Australia, New Zealand and many more. If we mention only North America, it detracts significantly from the "Global Presence" message we would wish to convey. That is a point all by itself, I guess. The total European market size is the roughly the same as the North American market in terms of people, money etc. South Pacific market is also large and probably further into open source than some parts of Europe. It really is high time we got the idea across that open source and Postgres, in particular, is not solely a US project. I mean no disrespect at all to any of my fine colleagues, but the geographical location of the chair in which we sit has very little to do with our mission, goals, teamwork, communication channels or whatever. The same isn't true of our external communications, which definitely are affected by local issues, markets and definitely by timezones. So I suggest we have a press contact in either England or Australia listed, instead of one of the three contacts. In time, I hope that we would see an American, European and Australasian contact on the list, so we have a much wider distribution of support. Or others. If other languages want to have global representation that would be good also, for example Spanish-speaking contacts in the Americas and Europe. (I haven't checked). I don't see any reason why a country has to speak a particular language for it to be represented either. For example the Japanese press release might reasonably wish to have a European contact also, since many businesses and their leaders speak Japanese, even if the countries in which they operate do not. The same thinking might apply to a range of other languages. -- Simon Riggs 2ndQuadrant http://www.2ndQuadrant.com
Simon Riggs wrote: > Just reading the Press Release for 8.3 > http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/press/pr/releases/8.3/en/release.txt > > Few thoughts spring to mind > > - Just in time background writer changes are nice, but since we already > mention the load distributed checkpoints (a closely related feature), it > seems we are saying too much at the expense of other features. > > - Asynch Commit is a seriously important feature, not necessarily for > existing users but definitely for people used to MySQL's speed and lack > of robustness with myisam tables. It's an important message that we can > get the speed of myisam and at the same time the robustness of Postgres. > I don't suggest that we specifically mention mysql, but it is a feature > that will alter many people's perception of Postgres, so we should talk > about it somewhere. +1 > > So I suggest we swap Just-in-time for Async Commit. > > (To forestall any negative comment, I would add: Yes, I did write that > code, but I wrote it *because* of its importance to a wide range of > people that might now consider PostgreSQL. No, I'm not asking for my > name in brackets anywhere.) > Ahh man why did you have to bring that up... No I have to take back my +1 :P > Lastly, the press contacts listed are all on the North American > continent and within a few hours of each other. Canada != U.S. I think this makes sense. > The total European market size is the roughly the same as the North > American market in terms of people, money etc. South Pacific market is > also large and probably further into open source than some parts of > Europe. It really is high time we got the idea across that open source > and Postgres, in particular, is not solely a US project. I mean no That's true. > disrespect at all to any of my fine colleagues, but the geographical > location of the chair in which we sit has very little to do with our > mission, goals, teamwork, communication channels or whatever. The same > isn't true of our external communications, which definitely are affected > by local issues, markets and definitely by timezones. > I think this is thread hi-jacking :). There are a whole lots of issues that you just brought up that are worth discussing but certainly not on this thread :) Sincerely, Joshua D. Drake
On Friday 14 December 2007 10:42, Joshua D. Drake wrote: > Simon Riggs wrote: > > Just reading the Press Release for 8.3 > > http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/press/pr/releases/8.3/en/rele > >ase.txt > > > > Few thoughts spring to mind > > > > - Just in time background writer changes are nice, but since we already > > mention the load distributed checkpoints (a closely related feature), it > > seems we are saying too much at the expense of other features. > > > > - Asynch Commit is a seriously important feature, not necessarily for > > existing users but definitely for people used to MySQL's speed and lack > > of robustness with myisam tables. It's an important message that we can > > get the speed of myisam and at the same time the robustness of Postgres. > > I don't suggest that we specifically mention mysql, but it is a feature > > that will alter many people's perception of Postgres, so we should talk > > about it somewhere. > > +1 > Simon, have you tested async_commit performance with any of the the beta releases? I was involved in a conversation on irc the other day that had some anecdotale reports that async_commit was not giving significant speed benfits, perhaps as a result of the other tuning and work that has gone into the release later in the development cycle. I've not had chance to verify this info, so I was wondering if you had. -- Robert Treat Build A Brighter LAMP :: Linux Apache {middleware} PostgreSQL
On Fri, 2007-12-14 at 15:22 -0500, Robert Treat wrote: > Simon, have you tested async_commit performance with any of the the beta > releases? I was involved in a conversation on irc the other day that had > some anecdotale reports that async_commit was not giving significant speed > benfits, perhaps as a result of the other tuning and work that has gone into > the release later in the development cycle. I've not had chance to verify > this info, so I was wondering if you had. Few of the features in 8.3 apply to all workloads, so such anecdotes could be applied to any of them. I don't think other tuning would mask that. I measured ~900% performance gain on a pure INSERT workload. Any transaction whose major component is the commit time will benefit from increased performance, i.e. short transactions get even faster. Many web apps fall into this category, especially those currently running on MySQL, Solid etc. But its also easy to come up with workloads that would show no gain at all. HOT has been similarly misunderstood, though I feel we now have a good understanding that it prevents performance drops, rather than simply being a boost in itself, as was originally thought. Of course, if anybody has some negative performance test cases for Async Commit, we'd be able to discuss those. But from all the things that have happened in recent weeks, I'm now wondering how much information is out there about all the performance features that have been added, and when to make use of them. -- Simon Riggs 2ndQuadrant http://www.2ndQuadrant.com
"Robert Treat" <xzilla@users.sourceforge.net> writes: > Simon, have you tested async_commit performance with any of the the beta > releases? I was involved in a conversation on irc the other day that had > some anecdotale reports that async_commit was not giving significant speed > benfits, perhaps as a result of the other tuning and work that has gone into > the release later in the development cycle. I've not had chance to verify > this info, so I was wondering if you had. It depends a lot on your workload. If you have many parallel connections keeping the i/o controller busy you might not see much change. Alternately if you have large updates then the 5-15ms waiting for fsync may make little impact. Where it helps a lot is on workloads like simple web applications which flood the server with tons of short updates on a handful of connections. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com Ask me about EnterpriseDB's On-Demand Production Tuning
Joshua D. Drake wrote: >> Lastly, the press contacts listed are all on the North American >> continent and within a few hours of each other. > > Canada != U.S. I think this makes sense. No, but both are North America, and many people on this side of the pond don't consider there to be a huge difference between the two. There is a much more clear difference between North America and Europe or Australaisa. /D
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 14 Dec 2007 21:27:16 +0000 Dave Page <dpage@postgresql.org> wrote: > Joshua D. Drake wrote: > >> Lastly, the press contacts listed are all on the North American > >> continent and within a few hours of each other. > > > > Canada != U.S. I think this makes sense. > > No, but both are North America, and many people on this side of the > pond don't consider there to be a huge difference between the two. > There is a much more clear difference between North America and > Europe or Australaisa. One could be said the same about France and Germany, although I don't the French and Germans feel that way... Or the Brits and Irish. Sincerely, Joshua D. Drake - -- The PostgreSQL Company: Since 1997, http://www.commandprompt.com/ Sales/Support: +1.503.667.4564 24x7/Emergency: +1.800.492.2240 Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate SELECT 'Training', 'Consulting' FROM vendor WHERE name = 'CMD' -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHYvbLATb/zqfZUUQRAv/fAJ0dUyXds6dDaL9d0wJNE0RtIG6DnACfV4lE OWbucV4l4avY5Rmmk5vzB70= =Q1uL -----END PGP SIGNATURE-----
Joshua D. Drake wrote: > One could be said the same about France and Germany, although I don't > the French and Germans feel that way... Or the Brits and Irish. France and Germany are seperated by language so I don't think they count. As a Brit I certainly feel much closer (not just geographically) to our Irish neighbours than the US or Canada. In any case, I agree with Simon on this one - we should list geographically diverse contacts on the press releases whereever possible, if only to illustrate how global our presence is. /D
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 14 Dec 2007 21:42:15 +0000 Dave Page <dpage@postgresql.org> wrote: > Joshua D. Drake wrote: > > One could be said the same about France and Germany, although I > > don't the French and Germans feel that way... Or the Brits and > > Irish. > > France and Germany are seperated by language so I don't think they > count. As a Brit I certainly feel much closer (not just > geographically) to our Irish neighbours than the US or Canada. > > In any case, I agree with Simon on this one - we should list > geographically diverse contacts on the press releases whereever > possible, if only to illustrate how global our presence is. I believe the contacts should be geographically located to the region in which the PR will be presented. E.g; if you have UK press that this will go to, then absolutely lets get the UK contact listed for those. However if it is U.S. Press it needs to be North American Contacts, preferrably U.S. contacts. Sincerely, Joshua D. Drake P.S. Canadians and Americans do not think they speak the same language :P - -- The PostgreSQL Company: Since 1997, http://www.commandprompt.com/ Sales/Support: +1.503.667.4564 24x7/Emergency: +1.800.492.2240 Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate SELECT 'Training', 'Consulting' FROM vendor WHERE name = 'CMD' -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHYv9vATb/zqfZUUQRArTpAJ4zCnFLOBLZvkYAKobEaQSkpSlvewCglTDq G5qTluKl52Kw2xQM5zchNy8= =GmiS -----END PGP SIGNATURE-----
Joshua D. Drake wrote: > I believe the contacts should be geographically located to the region > in which the PR will be presented. > > E.g; if you have UK press that this will go to, then absolutely lets > get the UK contact listed for those. However if it is U.S. Press it > needs to be North American Contacts, preferrably U.S. contacts. Yes, exactly. I assume we're not saying we only want the US/Canadian press to pickup our release, but also want English language European/Australaisian publications to pick it up? That's exactly why it makes sense to try to include contacts from some of those regions. > P.S. Canadians and Americans do not think they speak the same > language :P Yeah, so I've heard :-) /D
On Fri, 14 Dec 2007, Simon Riggs wrote: > I measured ~900% performance gain on a pure INSERT workload. Any > transaction whose major component is the commit time will benefit from > increased performance, i.e. short transactions get even faster. I saw something in the multi-hundred percent range as well for some simple benchmarks. The huge win is cases where right now the limiter is the rate of commits to disk, and in other situations I wouldn't be surprised to find little or no improvement. I'm not sure how to make that distinction clear to regular users though. I've been thinking about it since your first message in this thread, and so far I'm at a loss as to how to fairly and accurately characterize the improvement in a way that's short enough to put into the press release. It's hard to say any variant on "you can trade-off reliability for speed" without sounding like you don't care about reliability; you need the context that some competitors have been doing that already without making it so obvious to appreciate the feature. > Many web apps fall into this category, especially those currently > running on MySQL, Solid etc. When I was working on the big MySQL comparison piece over the summer, the only upcoming 8.3 feature I mentioned by name was async commit because it will shift the landscape more than anything else. You certainly have my vote that it's more noteworthy than the background writer changes, but writing a description blurb saying that may take some work. Maybe you should hire the PR people those guys who worked on the BGW change used, they managed to get a small tuning feature listed right up there in the big press release. -- * Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD
On Friday 14 December 2007 17:10, Joshua D. Drake wrote: > P.S. Canadians and Americans do not think they speak the same > language :P Especially when it's in French ;-) Robert
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Fri, 14 Dec 2007 18:32:05 -0500 Robert Bernier <robert.bernier5@sympatico.ca> wrote: > On Friday 14 December 2007 17:10, Joshua D. Drake wrote: > > P.S. Canadians and Americans do not think they speak the same > > language :P > > Especially when it's in French ;-) As I recall the French speaking part of Canada doesn't actually consider themselves true Canadians :P Joshua D. Drake - -- The PostgreSQL Company: Since 1997, http://www.commandprompt.com/ Sales/Support: +1.503.667.4564 24x7/Emergency: +1.800.492.2240 Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate SELECT 'Training', 'Consulting' FROM vendor WHERE name = 'CMD' -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHYxN1ATb/zqfZUUQRAjSIAJ9ANqfeGfXKczw1FtYkBPvIaVBo2ACfRCam ZzJAlQtQWtCru0tlb6MuICs= =5cx/ -----END PGP SIGNATURE-----
Simon, > Just reading the Press Release for 8.3 > http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/press/pr/releases/8.3/en/rel >ease.txt Note that this isn't publically accessible. People have to have CVS rights to it. That's on purpose, because it's important that the quotes *not* get distributed to the press before official release date -- and a couple reporters read this list archives. Hi, guys! > So I suggest we swap Just-in-time for Async Commit. I don't necessarily disagree with the priorities, but your changes are kind of untimely. That text was there for over a month of discussion -- it's gone out to the translators already. I'm sorry to be negative when you're trying to help, but I *begged* people to comment and correct in November. So, do you feel that making the change is worth potentially having some of the translated versions be inconsistent because they don't get updated? > Lastly, the press contacts listed are all on the North American > continent and within a few hours of each other. That's because the version you're looking at is the one I'm going to send out to the North American press, and is provided as an example. The one you send out would, of course, have your and Raymond's contact info, and the one Charles sends out would have his. To look at the *general* version, look in 8.3/translate -- that version will get filled in with the appropriate regional contacts for that language. > I don't see any reason why a country has to speak a > particular language for it to be represented either. For example the > Japanese press release might reasonably wish to have a European contact > also, since many businesses and their leaders speak Japanese, even if > the countries in which they operate do not. The same thinking might > apply to a range of other languages. I'm not clear on what you're suggesting we do differently? -- --Josh Josh Berkus PostgreSQL @ Sun San Francisco
On Fri, Dec 14, 2007 at 02:10:52PM -0800, Joshua D. Drake wrote: > > P.S. Canadians and Americans do not think they speak the same > language :P Yeah, eh? Like, those Merricans are, like, _losers_, but like, we're just a bucha hosers, eh? Now I gotta go get me a cruller to wash down this two-four. Or maybe the other way around. Andrew "McKenzie" Sullivan Tuhranna
On Fri, Dec 14, 2007 at 03:36:21PM -0800, Joshua D. Drake wrote: > As I recall the French speaking part of Canada doesn't actually > consider themselves true Canadians :P As any Acadian will tell you, the Province of Quebec is not all of French Canada. A
On Fri, Dec 14, 2007 at 09:42:15PM +0000, Dave Page wrote: > France and Germany are seperated by language so I don't think they So is Canada and, um, Canada ;-) A
On Friday 14 December 2007 18:49, Andrew Sullivan wrote: > On Fri, Dec 14, 2007 at 09:42:15PM +0000, Dave Page wrote: > > France and Germany are seperated by language so I don't think they > > So is Canada and, um, Canada ;-) so true and Belguim has French, Flemish and German and Switzerland has French and German and English (business only) and the Brits have the Welsh :-) Robert
> ------- Original Message ------- > From: Josh Berkus <josh@agliodbs.com> > To: pgsql-advocacy@postgresql.org > Sent: 14/12/07, 23:40:26 > Subject: Re: [pgsql-advocacy] Press Release review comments > > Simon, > > > Just reading the Press Release for 8.3 > > http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/press/pr/releases/8.3/en/rel > >ease.txt > > Note that this isn't publically accessible. People have to have CVS rights > to it. You might want to double check that. I can get it with no authentication at all. /D
On Fri, 2007-12-14 at 15:40 -0800, Josh Berkus wrote: > Note that this isn't publically accessible. People have to have CVS rights > to it. That's on purpose, because it's important that the quotes *not* > get distributed to the press before official release date -- and a couple > reporters read this list archives. Hi, guys! OK, sorry. > > So I suggest we swap Just-in-time for Async Commit. > > I don't necessarily disagree with the priorities, but your changes are kind > of untimely. That text was there for over a month of discussion -- it's > gone out to the translators already. I'm sorry to be negative when you're > trying to help, but I *begged* people to comment and correct in November. Apologies again. > So, do you feel that making the change is worth potentially having some of > the translated versions be inconsistent because they don't get updated? In short, Yes. > > Lastly, the press contacts listed are all on the North American > > continent and within a few hours of each other. > > That's because the version you're looking at is the one I'm going to send > out to the North American press, and is provided as an example. The one > you send out would, of course, have your and Raymond's contact info, and > the one Charles sends out would have his. To look at the *general* > version, look in 8.3/translate -- that version will get filled in with the > appropriate regional contacts for that language. OK, well this all makes sense. I didn't realise I was allowed to modify the UK version. That covers most of my thinking. I'll put myself, Ray and Dave on there if thats OK with them. I still think it would be helpful to have an International version for when recipients are multi-national, but thats just a nice to have. -- Simon Riggs 2ndQuadrant http://www.2ndQuadrant.com
Dave Page wrote: >>> Just reading the Press Release for 8.3 >>> http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/press/pr/releases/8.3/en/rel >>> ease.txt >> Note that this isn't publically accessible. People have to have CVS rights >> to it. > > You might want to double check that. I can get it with no authentication at all. As can I. Joshua D. Drake
All, > >> Note that this isn't publically accessible. People have to have CVS > >> rights to it. > > > > You might want to double check that. I can get it with no authentication > > at all. > > As can I. Crapola. Apparently there's no way to meaningfully secure CVSweb? Can we somehow exclude one pgfoundry project from CVSweb? -- Josh Berkus PostgreSQL @ Sun San Francisco
Josh Berkus wrote: > All, > >>>> Note that this isn't publically accessible. People have to have CVS >>>> rights to it. >>> You might want to double check that. I can get it with no authentication >>> at all. >> As can I. > > Crapola. Apparently there's no way to meaningfully secure CVSweb? > > Can we somehow exclude one pgfoundry project from CVSweb? We should be able to do it in the apache config. I will take a look at it this weekend (once I recover from a bout of some 24 hour virus). Joshua D. Drake
Simon Riggs wrote: > > > So I suggest we swap Just-in-time for Async Commit. > > > > I don't necessarily disagree with the priorities, but your changes are kind > > of untimely. That text was there for over a month of discussion -- it's > > gone out to the translators already. I'm sorry to be negative when you're > > trying to help, but I *begged* people to comment and correct in November. > > Apologies again. > > > So, do you feel that making the change is worth potentially having some of > > the translated versions be inconsistent because they don't get updated? > > In short, Yes. Yea, I have async commit as the first item the performance section of the release notes because it is our #1 performance improvment in terms of new performance capabilities. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +