2002-08-20 23:54:17 +00:00
|
|
|
<title>Handling long computations</title>
|
|
|
|
|
<body bgcolor=ffffff>
|
2002-07-30 23:48:08 +00:00
|
|
|
|
2002-08-20 23:54:17 +00:00
|
|
|
<h2>Handling long computations</h2>
|
2002-07-30 23:48:08 +00:00
|
|
|
|
2002-04-30 22:22:54 +00:00
|
|
|
<p>
|
2002-09-05 11:46:10 +00:00
|
|
|
Some applications have very long (several month) computations
|
2002-07-30 23:48:08 +00:00
|
|
|
that generate large amounts of local state,
|
|
|
|
|
and therefore are most efficiently done on a single host.
|
|
|
|
|
|
2002-04-30 22:22:54 +00:00
|
|
|
<p>
|
2002-07-30 23:48:08 +00:00
|
|
|
To accommodate such applications, BOINC allows workunits and results
|
|
|
|
|
to be linked together into <b>sequences</b>.
|
|
|
|
|
Each workunit and result may have a predecessor and a successor.
|
|
|
|
|
|
2002-04-30 22:22:54 +00:00
|
|
|
<p>
|
2002-07-30 23:48:08 +00:00
|
|
|
If a result A has a predecessor B,
|
|
|
|
|
the scheduling server will dispatch A only
|
|
|
|
|
to the host that has handled (or is currently handling) B.
|
|
|
|
|
If result A times out,
|
|
|
|
|
then the scheduling server will dispatch B to any eligible host.
|
