Text:USA has 1417 > 1000 celLs

Danielle et Jean Thierry-Mieg mieg at kaa.crbm.cnrs-mop.fr
Fri Sep 22 02:18:59 EST 1995

Dave says:

When I load my .ace files into ace4_1 I get no overt error messages but
log.wrm contains a sprinkling of messages like

Class 30: object Text:USA has 1417 > 1000 celLs, please edit the models
Class 76: object Germplasm:INIA F 66 has 1650 > 1000 celLs, please edit the models
Class 62: object Map_Data:Barley, SxM has 21849 > 1000 celLs, please edit the models

The point is that the performance of acedb degrades extremelly badly when the number
of nodes in an object becomes large.
I beleive that 10^4 is already quite bad.
In IGD we went much over resulting in a quasi collapse of the performances.

For this reason, i introduced a message system in log.wrm issued once per class
Most probably, these long lists are created by XREF

Text:USA could be modified to be a tag, it does not seem so crucial to XREF back
from that text to the object or be downgraded from ?Text to Text in the objects where it occurs
In the case of Map_Data:Barley, i would also question the model.

Generally speaking, acedb tells you that your model is counter-productive
and that you may wish to modify it a bit.

These warnings have nothing to do with the actual cachesize
By the way, CACHE1 is now dynamic and can be started at a lower value (1000) if you
have 16M or less of RAM.

More information about the Acedb mailing list

Send comments to us at biosci-help [At] net.bio.net