3pp-mirror/Meshroom
1
0
Fork 0
mirror of https://github.com/alicevision/Meshroom.git synced 2025-08-06 10:18:42 +02:00
Code Issues Projects Releases Packages Wiki Activity

ripple submitter slots count selection heuristic

Browse source
This commit is contained in:
fabien servant 2022-12-23 07:41:51 +01:00
parent 5cd7b49a64
commit 05e8b87200
1 changed files with 14 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

16
meshroom/submitters/rippleSubmitter.py
View file

@ -43,6 +43,18 @@ class RippleSubmitter(BaseSubmitter):
"NORMAL":",gpu8G", "NORMAL":",gpu8G",
"INTENSIVE":",gpu16G" "INTENSIVE":",gpu16G"
} }
#decide if we need multiple slots
minProcessors = 1
maxProcessors = 1
if Level.INTENSIVE in (node.nodeDesc.ram, node.nodeDesc.cpu):
#at least 2 slots
minProcessors = 2
#if more than 2 are available without waiting, use 3 or 4
maxProcessors = 4
elif Level.NORMAL in (node.nodeDesc.ram, node.nodeDesc.cpu):
#if 2 are available, otherwise 1
maxProcessors = 2
#Specify some constraints #Specify some constraints
requirements = "!\"rs*\",@.mem>25{gpu}".format(gpu=gpudict[node.nodeDesc.gpu.name]) requirements = "!\"rs*\",@.mem>25{gpu}".format(gpu=gpudict[node.nodeDesc.gpu.name])
@ -66,12 +78,12 @@ class RippleSubmitter(BaseSubmitter):
commandext = '{cmd} --iteration {iter}'.format(cmd=command, iter=iteration) commandext = '{cmd} --iteration {iter}'.format(cmd=command, iter=iteration)
#Create process task with parameters #Create process task with parameters
rippleproc = RippleProcessWithSlots(name='{name} iteration {iter}'.format(name=node.name, iter=iteration), discipline='ripple', appendKeys=True, keys=requirements, label=node.name, cmdList=[commandext], waitsFor=waitsFor, minProcessors=1, maxProcessors=1) rippleproc = RippleProcessWithSlots(name='{name} iteration {iter}'.format(name=node.name, iter=iteration), discipline='ripple', appendKeys=True, keys=requirements, label=node.name, cmdList=[commandext], waitsFor=waitsFor, minProcessors=minProcessors, maxProcessors=maxProcessors)
rippleprocs.append(rippleproc) rippleprocs.append(rippleproc)
rippleObj = RippleGroup(label="{name} Group".format(name=node.name), tasks=rippleprocs, name=node.name, waitsFor=waitsFor) rippleObj = RippleGroup(label="{name} Group".format(name=node.name), tasks=rippleprocs, name=node.name, waitsFor=waitsFor)
else: else:
rippleObj = RippleProcessWithSlots(name=node.name, discipline='ripple', appendKeys=True, keys=requirements, label=node.name, cmdList=[command], waitsFor=waitsFor, minProcessors=1, maxProcessors=1) rippleObj = RippleProcessWithSlots(name=node.name, discipline='ripple', appendKeys=True, keys=requirements, label=node.name, cmdList=[command], waitsFor=waitsFor, minProcessors=minProcessors, maxProcessors=maxProcessors)
return rippleObj return rippleObj