[squeak-dev] OmniBase/Magma - how big??

Hello all,

Any success or horror stories with OmniBase and/or Magma?  Can you
report that either worked well at or got into trouble beyond any
particular size (either in object count or bytes of data)?  Other
comments?

Bill




Wilhelm K. Schwab, Ph.D.
University of Florida
Department of Anesthesiology
PO Box 100254
Gainesville, FL 32610-0254

Email: [hidden email]
Tel: (352) 846-1285
FAX: (352) 392-7029

Hi Bill,

        AFAIK you can't use omnibase in multiuser in linux. I saw in squeak
source a package OSProcess.filelockcache from Avi may be is to make it able to?

        I'm trying Magma and is fine by now. I'll add something I made to make
it able to maintain indexes automatically that I used to use with OmniBase.

        cheers,

Sebastian Sastre

 

Hi!

(ccing Magma list)

"Bill Schwab" <[hidden email]> wrote:
> Hello all,
>
> Any success or horror stories with OmniBase and/or Magma?  Can you
> report that either worked well at or got into trouble beyond any
> particular size (either in object count or bytes of data)?  Other
> comments?
>
> Bill

>From the Gjallar project we can report that Magma works really nice. We
have had perhaps 2-3 buglets and Chris fixed them all quickly and
promptly. IIRC they were related mainly to the query stuff which Chris
pushed out fast because of our popular demand - so he was definitely not
to blame. :)

Generally I would say that the only "issue" we had was slow(ish) read
performance. So if you write an "interactive" application with Magma you
need to take two things into account:

1. Since Magma is an ODB it builds a cache of objects into your session.
So when/if the session is cold you get punished with load times. But
when it is hot they are zero. Essentially what this means in Gjallar is
that when a user logs in things will take a bit of time the first time
she does something. But from then on things go much faster and since a
Gjallar user typically is logged in for a long time - typically all day
- it works fine.

2. Even with a warm session (with lots of cached objects) you can still
get punished with longish load times. This can be remedied a LOT with
setting up good read strategies.

IMHO I think Magma would SHINE if we somehow could boost the
serialization/deserialization mechanism in it. Because I suspect this is
where the bottleneck is regarding the above. An Exuperified Magma could
be the answer - or some other HARD approach - I dunno. Chris?

regards, Göran

Göran,

Can you give me a sense of what slow-ish means?  Unfair, I realize, but any guesstimates with context would be greatly appreciated.

Were there any particular reasons you chose Magma over OmniBase?  In my case, much of what I would do first with an OODB would happen on Dolphin, which AFAICT favors OmniBase.

Bill




============================

>From the Gjallar project we can report that Magma works really nice. We
have had perhaps 2-3 buglets and Chris fixed them all quickly and
promptly. IIRC they were related mainly to the query stuff which Chris
pushed out fast because of our popular demand - so he was definitely not
to blame. :)

Generally I would say that the only "issue" we had was slow(ish) read
performance. So if you write an "interactive" application with Magma you
need to take two things into account:

1. Since Magma is an ODB it builds a cache of objects into your session.
So when/if the session is cold you get punished with load times. But
when it is hot they are zero. Essentially what this means in Gjallar is
that when a user logs in things will take a bit of time the first time
she does something. But from then on things go much faster and since a
Gjallar user typically is logged in for a long time - typically all day
- it works fine.

2. Even with a warm session (with lots of cached objects) you can still
get punished with longish load times. This can be remedied a LOT with
setting up good read strategies.

IMHO I think Magma would SHINE if we somehow could boost the
serialization/deserialization mechanism in it. Because I suspect this is
where the bottleneck is regarding the above. An Exuperified Magma could
be the answer - or some other HARD approach - I dunno. Chris?

regards, Göran

Wilhelm K. Schwab, Ph.D.
University of Florida
Department of Anesthesiology
PO Box 100254
Gainesville, FL 32610-0254

Email: [hidden email]
Tel: (352) 846-1285
FAX: (352) 392-7029

>  IMHO I think Magma would SHINE if we somehow could boost the
>  serialization/deserialization mechanism in it. Because I suspect this is
>  where the bottleneck is regarding the above. An Exuperified Magma could
>  be the answer - or some other HARD approach - I dunno. Chris?

Magma users will appreciate the new r41Beta1 just posted Wednesday.
Profiling had shown Magma spending quite a bit of time in
IdentityDictionary's, largely for oid lookups.  I'm still
benchmarking, but with Igors new IdentityDictionary replacements I am
seeing tremendous gains across the board, for both reads and commits.

I'm also excited for single-clause Readers now improved #size and #at:
by a factor of 100x.. again!  They're so fast now, as a joke, I
demonstrated a scrolling list with more than two-hundred thousand
objects (two million would work as well)  The list widget only needs
its underlying domain to understand #size and #at:, so the Reader
plugs-in seamlessly.  Scrolling performance is good, even with remote
connections.

Great new Crhis. Great news

        thanks for that

Sebastian Sastre
 

Hi,
my name is gildas, i'm a student in 4th year in IT Management and i would like to use Magma for my project of implementation of an accounting software in squeak... The problem is that i'm used to use sql for the databases and i'm a little bit lost with Magma, i would like to know if there is any tutorial for magma which can explain how to create a server, how to create data and how to read it...

Sorry i'm really a begginer, but i really want to use magma which seems to be really integrated to squeak instead of an sql database.

Thank you,
regards,

Gildas.
_________________________________________________________________
Votre contact a choisi Hotmail, l'e-mail nouvelle génération. Créez un compte.
http://www.windowslive.fr/hotmail/default.asp

Gildas,


Magma isn't too difficult to get going. The main swiki page is located
here:

http://wiki.squeak.org/squeak/2665

And you may want to start with 'Getting Started with Magma':

http://wiki.squeak.org/squeak/2689

I've not seen any full tutorials for Magma, but maybe someone else has.

Magma is well factored and I would recommend browsing the categories for
the server and client api. Main classes of interest will be;

MagmaRepositoryController
MagmaSession
MaReadStrategy
MagmaCollectionReader

I hope this helps as a start,

Travis

Travis

On Sat, 2008-03-01 at 17:00 +0100, Gildas Le Mentec wrote:

Hi Gildas.

Welcome to the community! I assume Travis's answer is suffiencient.

There's a special mailing list for beginners where people such as yourself can get advice on Squeak. The details for joining are at http://www.squeak.org/Community.

Gulik.

--
http://people.squeakfoundation.org/person/mikevdg
http://gulik.pbwiki.com/

Also you will find some guidelines on the context of using with
Seaside, on the tutorial:

http://www.swa.hpi.uni-potsdam.de/seaside/tutorial (See Chapter 8 -
Persistence).

hth.
gsa.


2008/3/2, Travis Kay <[hidden email]>:

Newbie question, but posted where? I can't find it in Monticello or on squeaksource.
...Stan

Chris Muller-3 wrote

Magma users will appreciate the new r41Beta1 just posted Wednesday.

I hope this helps.

http://lists.squeakfoundation.org/pipermail/magma/2008-March/001030.html

Cheers.

On Wed, 2 Apr 2008 05:35:36 -0700 (PDT)
stan shepherd <[hidden email]> wrote:


--
Giuseppe Luigi Punzi <[hidden email]>
<http://www.lordzealon.com>