<html>
<head>
<meta http-equiv="Content-Type" content="text/html;
charset=windows-1252">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>Sure. Here is what we have:</p>
<p>########################## Scheduling
#####################################<br>
### This section is specific to scheduling<br>
<br>
### Tells the scheduler to enforce limits for all partitions<br>
### that a job submits to.<br>
EnforcePartLimits=ALL<br>
<br>
### Let's slurm know that we have a jobsubmit.lua script<br>
JobSubmitPlugins=lua<br>
<br>
### When a job is launched this has slurmctld send the user
information<br>
### instead of having AD do the lookup on the node itself.<br>
LaunchParameters=send_gids<br>
<br>
### Maximum sizes for Jobs.<br>
MaxJobCount=200000<br>
MaxArraySize=10000<br>
DefMemPerCPU=100<br>
<br>
### Job Timers<br>
CompleteWait=0<br>
<br>
### We set the EpilogMsgTime long so that Epilog Messages don't
pile up all <br>
### at one time due to forced exit which can cause problems for
the master.<br>
EpilogMsgTime=3000000<br>
InactiveLimit=0<br>
KillWait=30<br>
<br>
### This only applies to the reservation time limit, the job must
still obey<br>
### the partition time limit.<br>
ResvOverRun=UNLIMITED<br>
MinJobAge=600<br>
Waittime=0<br>
<br>
### Scheduling parameters<br>
### FastSchedule 2 lets slurm know not to auto detect the node
config<br>
### but rather follow our definition. We also use setting 2 as
due to our geographic<br>
### size nodes may drop out of slurm and then reconnect. If we
had 1 they would be<br>
### set to drain when they reconnect. Setting it to 2 allows them
to rejoin with out<br>
### issue.<br>
FastSchedule=2<br>
SchedulerType=sched/backfill<br>
SelectType=select/cons_res<br>
SelectTypeParameters=CR_Core_Memory<br>
<br>
### Govern's default preemption behavior<br>
PreemptType=preempt/partition_prio<br>
PreemptMode=REQUEUE<br>
<br>
### default_queue_depth should be some multiple of the
partition_job_depth,<br>
### ideally number_of_partitions * partition_job_depth, but
typically the main<br>
### loop exits prematurely if you go over about 400. A
partition_job_depth of<br>
### 10 seems to work well.<br>
SchedulerParameters=\<br>
default_queue_depth=1150,\<br>
partition_job_depth=10,\<br>
max_sched_time=50,\<br>
bf_continue,\<br>
bf_interval=30,\<br>
bf_resolution=600,\<br>
bf_window=11520,\<br>
bf_max_job_part=0,\<br>
bf_max_job_user=10,\<br>
bf_max_job_test=10000,\<br>
bf_max_job_start=1000,\<br>
bf_ignore_newly_avail_nodes,\<br>
kill_invalid_depend,\<br>
pack_serial_at_end,\<br>
nohold_on_prolog_fail,\<br>
preempt_strict_order,\<br>
preempt_youngest_first,\<br>
max_rpc_cnt=8<br>
<br>
################################ Fairshare
################################<br>
### This section sets the fairshare calculations<br>
<br>
PriorityType=priority/multifactor<br>
<br>
### Settings for fairshare calculation frequency and shape.<br>
FairShareDampeningFactor=1<br>
PriorityDecayHalfLife=28-0<br>
PriorityCalcPeriod=1<br>
<br>
### Settings for fairshare weighting.<br>
PriorityMaxAge=7-0<br>
PriorityWeightAge=10000000<br>
PriorityWeightFairshare=20000000<br>
PriorityWeightJobSize=0<br>
PriorityWeightPartition=0<br>
PriorityWeightQOS=1000000000</p>
<p>I'm happy to chat about any of the settings if you want, or share
our full config.</p>
<p>-Paul Edmon-<br>
</p>
<div class="moz-cite-prefix">On 5/29/19 10:17 AM, Julius, Chad
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:DM6PR06MB496947306BA65FD33CAC7700FC1F0@DM6PR06MB4969.namprd06.prod.outlook.com">
<meta http-equiv="Content-Type" content="text/html;
charset=windows-1252">
<meta name="Generator" content="Microsoft Word 15 (filtered
medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]-->
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:#0563C1;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
p.msonormal0, li.msonormal0, div.msonormal0
{mso-style-name:msonormal;
mso-margin-top-alt:auto;
margin-right:0in;
mso-margin-bottom-alt:auto;
margin-left:0in;
font-size:12.0pt;
font-family:"Times New Roman",serif;}
span.EmailStyle18
{mso-style-type:personal-compose;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;
font-family:"Calibri",sans-serif;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
<div class="WordSection1">
<p class="MsoNormal">All, <o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">We rushed our Slurm install due to a short
timeframe and missed some important items. We are now looking
to implement a better system than the first in, first out we
have now. My question, are the defaults listed in the
slurm.conf file a good start? Would anyone be willing to
share their Scheduling section in their .conf? Also we are
looking to increase the maximum array size but I don’t see
that in the slurm.conf in version 17. Am I looking at an
upgrade of Slurm in the near future or can I just add
MaxArraySize=somenumber?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">The defaults as of 17.11.8 are:<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"># SCHEDULING<o:p></o:p></p>
<p class="MsoNormal">#SchedulerAuth=<o:p></o:p></p>
<p class="MsoNormal">#SchedulerPort=<o:p></o:p></p>
<p class="MsoNormal">#SchedulerRootFilter=<o:p></o:p></p>
<p class="MsoNormal">#PriorityType=priority/multifactor<o:p></o:p></p>
<p class="MsoNormal">#PriorityDecayHalfLife=14-0<o:p></o:p></p>
<p class="MsoNormal">#PriorityUsageResetPeriod=14-0<o:p></o:p></p>
<p class="MsoNormal">#PriorityWeightFairshare=100000<o:p></o:p></p>
<p class="MsoNormal">#PriorityWeightAge=1000<o:p></o:p></p>
<p class="MsoNormal">#PriorityWeightPartition=10000<o:p></o:p></p>
<p class="MsoNormal">#PriorityWeightJobSize=1000<o:p></o:p></p>
<p class="MsoNormal">#PriorityMaxAge=1-0<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><b>Chad Julius</b><o:p></o:p></p>
<p class="MsoNormal">Cyberinfrastructure Engineer Specialist<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><b>Division of Technology & Security</b><o:p></o:p></p>
<p class="MsoNormal">SOHO 207, Box 2231<o:p></o:p></p>
<p class="MsoNormal">Brookings, SD 57007<o:p></o:p></p>
<p class="MsoNormal">Phone: 605-688-5767<o:p></o:p></p>
<p class="MsoNormal"> <o:p></o:p></p>
<p class="MsoNormal"><a href="http://www.sdstate.edu/"
moz-do-not-send="true"><span style="color:#0563C1">www.sdstate.edu</span></a><o:p></o:p></p>
<p class="MsoNormal"><img style="width:2.6041in;height:.75in"
id="Picture_x0020_1"
src="cid:part2.730CB0C7.9E4716A0@cfa.harvard.edu"
alt="cid:image007.png@01D24AF4.6CEECA30" class=""
width="250" height="72" border="0"><o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</blockquote>
</body>
</html>