[Gluster-devel] Local vs unify

Gareth Bult gareth at encryptec.net
Mon Apr 28 08:58:59 UTC 2008


>Is it just me or does it seem like your latest mails have been sort of
>insulting, and now this mail looks like you are attempting to turn
>people away from glusterfs because you could not get it to work in
>your situation?

Well, that all depends on your point of view. It is I guess human nature for people to feel insulted if someone is critical of something in which you have a personal interest.

But just to clarify (a) I could get it to work and indeed have spent around 2 man months personally testing and experimenting with it and (b) I'm not attempting to "turn people away". I have a fairly heavy time investment in the project.

>Im not a developer so I cant speak for anyone else, but I dont
>remember reading anywhere that glusterfs was production ready, a final
>project, or anything to that matter? Is this not a project in the
>works? If it works for you in its current state, super, but otherwise
>it is an actively developed project? I also dont remember seeing that
>you had to PAY for it anywhere, it is FREE. So to complain about it or
>get rude over it, especially when you have not contributed or added
>anything constructive to it, well thats kind of lame in my eyes.

Ok, let's be clear here, so far as I can seem GlusterFS is available as a commercial product.
http://www.zresearch.com/subscription.php

One of my points was that there is a HEAVY implication that this IS a production ready environment.
This is liable to encourage people who are doing a little more than just "playing".

Is this not a typical open sourced project model, release a free product and feed the free testing results back into the commercial model? I have nothing against this, this is normal. All I am attempting to do is to point out issues that in the interests of both the open source and commercial users - could do with being addressed.

>Ive read the last couple mails and just thought to myself, god, i hope
>the devs dont get pissed or bummed by this guy in any way.
>If glusterfs doesnt work in your case, well, great, move on! There
>were more than one suggestions about alternatives that were given, but
>instead the thread just kept going on with no point in sight.

So you'd rather everyone said "wonderful", and not have anyone report problems or bugs ?

>Not meaning to be a troll myself, or start a flame war but c'mon, its
>free software, if it doesnt work for you dont bitch about it, move on
>to something else, or help write a translator or whatever that does
>what YOU want it to do.

Problems only get fixed if people report them. Sometimes developers don't realize how much of a problem something is unless people make a fuss.. development test harnesses tend to be fairly focussed and one of the reasons the open source model is so valuable to developers is the real-world feedback. (I am a developer by the way ...)

>Kudos to the guys/gals that have worked so hard on glusterfs, it seems
>like such a kickass project. I have not implemented it into a
>production environment myself YET, but I except there to be bugs and
>things not working perfectly yet. As far as I know this is still a
>fairly young project. My biggest hurdles has just been the learning
>curve and how to set up the bricks in the various ways. But not once
>have I complained that it didnt work right when it hasnt worked the
>way I wanted it, most likely due to my lack of knowing how to properly
>configure it.

Sure, and when you do have it all working and find that there is a reason why it won't work, you will want to report the fault with a view to having someone look at it and either recommend a solution or fix a bug... especially after having invested the time in actually getting it all going.

OR , do you (a) expect all open source projects to be perfect by the time you arrive at that stage, or (b) do you spend your time playing with projects and then dump them as soon as you find there is an issue with them?

Gareth.

----- Original Message -----
From: "Brandon Lamb" <brandonlamb at gmail.com>
To: gluster-devel at nongnu.org
Sent: Sunday, April 27, 2008 4:52:25 PM GMT +00:00 GMT Britain, Ireland, Portugal
Subject: Re: [Gluster-devel] Local vs unify

On Sun, Apr 27, 2008 at 5:33 AM, Gareth Bult <gareth at encryptec.net> wrote:
> >Are there some benchmrks available about this ?
>
> I'm not entirely sure how useful benchmarks are, certainly in this context. As demonstrated by the benchmarks currently on the site, it's quite possible to make them show pretty much whatever you want if you pick your own context.
>
> Gluster REALLY is quick in some contexts, certainly I can make Gluster look quick compared to NFS if we're talking about access to a single file or copying larger files. If on the other hand we're talking smaller files (i.e. many real world situations) then Gluster falls flat. "find" on a large gluster FS can take minutes rather than seconds on a local FS (for example).
>
> It may of course be I'm doing it all wrong .. however (!) if I am, given the time I've spent and the fact that I do have it all working, there may be room for improvement when it comes to the documentation (!)
>
> >local storeage CAN be notable
>
> Note; gluster (in particular AFR) on large files is currently flawed (IMHO). gluster on lots of small files is, as far as I can see flawed in the context of being too slow compared to local file-system access. This is not to say that Gluster is not useful or that issues cannot be fixed.
>
> You might be advised to setup your own test framework and test with your own data to get a true measure of how Gluster will perform for "you" in "your" environment ... Gluster is SO flexible, generic benchmarks can often be nothing more than speculation..
>
> Just to clarify; I think Gluster's general design is second to none .. I just there there are still a few implementation glitches to work through ...
>
> Gareth.
>
>
>
> ----- Original Message -----
> From: "Alain Baeckeroot" <alain.baeckeroot at laposte.net>
> To: gluster-devel at nongnu.org
> Sent: Sunday, April 27, 2008 8:04:55 AM GMT +00:00 GMT Britain, Ireland, Portugal
> Subject: Re: [Gluster-devel] Local vs unify
>
> Le samedi 26 avril 2008, Gareth Bult a écrit :
> > Technically, if you have local storage on each node then GlusterFS/Unify is a useful solution, but the performance overhead compared to local storeage can be notable.
> >
>
> Are there some benchmrks available about this ?
>
> Regards
> Alain Baeckeroot
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>

Is it just me or does it seem like your latest mails have been sort of
insulting, and now this mail looks like you are attempting to turn
people away from glusterfs because you could not get it to work in
your situation?

Im not a developer so I cant speak for anyone else, but I dont
remember reading anywhere that glusterfs was production ready, a final
project, or anything to that matter? Is this not a project in the
works? If it works for you in its current state, super, but otherwise
it is an actively developed project? I also dont remember seeing that
you had to PAY for it anywhere, it is FREE. So to complain about it or
get rude over it, especially when you have not contributed or added
anything constructive to it, well thats kind of lame in my eyes.

Ive read the last couple mails and just thought to myself, god, i hope
the devs dont get pissed or bummed by this guy in any way.

If glusterfs doesnt work in your case, well, great, move on! There
were more than one suggestions about alternatives that were given, but
instead the thread just kept going on with no point in sight.

Not meaning to be a troll myself, or start a flame war but c'mon, its
free software, if it doesnt work for you dont bitch about it, move on
to something else, or help write a translator or whatever that does
what YOU want it to do.

Kudos to the guys/gals that have worked so hard on glusterfs, it seems
like such a kickass project. I have not implemented it into a
production environment myself YET, but I except there to be bugs and
things not working perfectly yet. As far as I know this is still a
fairly young project. My biggest hurdles has just been the learning
curve and how to set up the bricks in the various ways. But not once
have I complained that it didnt work right when it hasnt worked the
way I wanted it, most likely due to my lack of knowing how to properly
configure it.


_______________________________________________
Gluster-devel mailing list
Gluster-devel at nongnu.org
http://lists.nongnu.org/mailman/listinfo/gluster-devel





More information about the Gluster-devel mailing list