Platform LSF Administration Guide Version 6.2

Chapter 19
Reserving Resources
Administering Platform LSF
349
Time-based reservation and greedy reservation compared
Greedy
reservation
example
A cluster has four hosts: A, B, C and D, with 4 CPUs each. Four jobs are running in the
cluster:
Job1, Job2, Job3 and Job4. According to calculated job estimated start time,
the job finish times (FT) have this order: FT(
Job2) < FT(Job4) < FT(Job1) <
FT(
Job3).
The highest priority pending job requests –n 6 –R “span[ptile=2]”. Based on this, the
job needs three hosts and two CPUs on each host. The default greedy slot reservation
calculates job start time as the job finish time of
Job4; after Job4 finishes, requested
slots of pending job can be satisfied.
Start time prediction Time-based reservation Greedy reservation
Backfill scheduling if free
slots are available
Yes Yes
Correct with no job topology Yes Yes
Correct for job topology
requests
Yes No
Correct based on resource
allocation limits
Yes (guaranteed if only two limits
are defined)
No
Correct for memory requests Yes No
When no slots are free for
reservation
Yes No
Future allocation and
reservation based on earliest
start time
Yes No
bjobs displays best estimate Yes No
bjobs displays predicted
future allocation
Yes No
Absolute predicted start time
for all jobs
No No
Advance reservation
considered
No No
Job4 Job3 Job4 Job3
Job4 Job3 Job4 Job3
Job2 Job3 Job2
Job3 Job1 Job1
J
Current Inuse
Current available
A BCD