[Magma] performance (was: test result)

Cees de Groot cg at cdegroot.com
Wed Aug 7 14:08:16 UTC 2002


Chris Muller <afunkyobject at yahoo.com> said:
>I didn't wait for this one to finish, but I expect you'll see something very
>close to five hours.
>
Benchmark: VW has a problem here, too (maybe not a good excuse...). It has
been pumping away at your sample code for a couple of minutes now. The
SmallInteger version takes 37ms on my PII/400 with VW, the LargeInteger
version a couple of minutes (still holding on). 

>Thankfully, I believe my solution is simple.  I'm going to change my oid map so
>that new and user objects are below 32-bits.  It'll limit a Magma repository to
>32 million objects (excluding integers, floats and characters) but hey, I don't
>think Squeak can currently write past just a few gigs anyway.
>
32 bits is 4billion objects, or are you using a lot of bits at the front to
get it down to 32m?

I'm happy with 32m as well (although you always have the risk that you maybe
don't have 32m at a certain point of time, but surely that you cycle through
all the 32m oids in a reasonable amount of time.

Visual is still pumping. Instead of boring you with the Usenet equivalent of
Muzak while waiting for the answer, I'm going to fetch a drink... 

VW takes 622,918 ms on my box. 10 minutes is better than 4 hours, but still
quite slow...

-- 
Cees de Groot               http://www.cdegroot.com     <cg at cdegroot.com>
GnuPG 1024D/E0989E8B 0016 F679 F38D 5946 4ECD  1986 F303 937F E098 9E8B
Cogito ergo evigilo



More information about the Squeak-dev mailing list