[RC5] When to Fetch Work ?
mikus at bga.com
Fri Sep 8 21:05:08 EDT 2000
I'm running OGR-25 off-line, meaning the dnetc program can't just
fetch work anytime it wants to. But it is hard for me to estimate
when the current work will be finished -- the smallest single stub
(unit of work) I have been given had 8 billion nodes, whereas the
largest single stub (unit of work) I've been given had THIRTY-FIVE
times as much work to do !! Is there some way for me to PLAN how
long it will take to process a given number (say 10) of OGR-25 units ?
Also, last year the software used its input buffers in LIFO fashion,
meaning that if new input was fetched *before* the input buffer was
exhausted, that new work was placed ahead of the work still there --
meaning the bottom unit in the input queue might stay unprocessed
"forever". Does the software still use its input buffers as LIFO ?
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest
More information about the rc5