<div dir="ltr"><div>Hi Chris,</div><div><br></div>re: "can't run more than 1 job per node at a time. " <div><br></div><div>try "scontrol show config" and grep for defmem</div><div><br></div><div>IIRC by default the memory request for any job is all the memory in a node.</div><div><br></div><div>Regards,</div><div>Alex</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Apr 4, 2019 at 4:01 PM Andy Riebs <<a href="mailto:andy.riebs@hpe.com">andy.riebs@hpe.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div>
in slurm.conf, on the line(s) starting "NodeName=", you'll want to
add specs for sockets, cores, and threads/core.<br>
<br>
<div id="gmail-m_5522972163051033896smartTemplate4-quoteHeader">
<hr> <b>From:</b> Chris Bateson <a class="gmail-m_5522972163051033896moz-txt-link-rfc2396E" href="mailto:cbateson@vt.edu" target="_blank"><cbateson@vt.edu></a> <br>
<b>Sent:</b> Thursday, April 04, 2019 5:18PM <br>
<b>To:</b> Slurm-users <a class="gmail-m_5522972163051033896moz-txt-link-rfc2396E" href="mailto:slurm-users@lists.schedmd.com" target="_blank"><slurm-users@lists.schedmd.com></a><br>
<b>Cc:</b> <br>
<b>Subject:</b> [slurm-users] Slurm 1 CPU <br>
</div>
<div class="gmail-m_5522972163051033896replaced-blockquote" type="cite">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">I should start out by saying that I am
extremely new to anything HPC. Our end users purchased
a 20 node cluster which a vendor set up for us with
Bright/Slurm.
<div><br>
</div>
<div>After our vendor said everything was complete and
we started migrating our users workflow to the new
cluster they discovered that they can't run more than
1 job per node at a time. We started researching
enabling consumable resources which I believe we've
done so however we're getting the same result. </div>
<div><br>
</div>
<div>I've just discovered today that both <b>scontrol
show node</b> and <b>sinfo -lNe</b> show that each
of our nodes have 1 CPU. I'm guessing that's why we
can't submit more than 1 job at a time. I'm trying to
determine where is it getting this information and how
can I get it to display the correct CPU information.</div>
<div><br>
</div>
<div>Sample info:</div>
<div><br>
</div>
<div><b>scontrol show node</b></div>
</div>
<blockquote style="margin:0px 0px 0px 40px;border:none;padding:0px">
<div>
<div>
<div>NodeName=cnode001 Arch=x86_64 CoresPerSocket=1</div>
</div>
</div>
<div>
<div> CPUAlloc=0 CPUErr=0 CPUTot=1 CPULoad=0.01</div>
</div>
<div>
<div> AvailableFeatures=(null)</div>
</div>
<div>
<div> ActiveFeatures=(null)</div>
</div>
<div>
<div> Gres=(null)</div>
</div>
<div>
<div> NodeAddr=cnode001 NodeHostName=cnode001
Version=17.11</div>
</div>
<div>
<div> OS=Linux 3.10.0-693.el7.x86_64 #1 SMP Thu Jul
6 19:56:57 EDT 2017</div>
</div>
<div>
<div> RealMemory=192080 AllocMem=0 FreeMem=188798
Sockets=1 Boards=1</div>
</div>
<div>
<div> State=IDLE ThreadsPerCore=1 TmpDisk=2038
Weight=1 Owner=N/A MCS_label=N/A</div>
</div>
<div>
<div> Partitions=defq</div>
</div>
<div>
<div> BootTime=2019-03-26T14:28:24
SlurmdStartTime=2019-03-26T14:29:55</div>
</div>
<div>
<div> CfgTRES=cpu=1,mem=192080M,billing=1</div>
</div>
<div>
<div> AllocTRES=</div>
</div>
<div>
<div> CapWatts=n/a</div>
</div>
<div>
<div> CurrentWatts=0 LowestJoules=0 ConsumedJoules=0</div>
</div>
<div>
<div> ExtSensorsJoules=n/s ExtSensorsWatts=0
ExtSensorsTemp=n/s</div>
</div>
</blockquote>
<br>
<div><b>sinfo -lNe</b></div>
<blockquote style="margin:0px 0px 0px 40px;border:none;padding:0px">
<div>
<div>NODELIST NODES PARTITION STATE CPUS
S:C:T MEMORY TMP_DISK WEIGHT AVAIL_FE REASON</div>
<div>cnode001 1 defq* idle 1
1:1:1 192080 2038 1 (null) none</div>
</div>
</blockquote>
<div dir="ltr"><br>
</div>
<div><b>lscpu</b></div>
</div>
</div>
<blockquote style="margin:0px 0px 0px 40px;border:none;padding:0px">
<div>
<div>
<div>
<div>Architecture: x86_64</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>CPU op-mode(s): 32-bit, 64-bit</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Byte Order: Little Endian</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>CPU(s): 48</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>On-line CPU(s) list: 0-47</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Thread(s) per core: 1</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Core(s) per socket: 24</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Socket(s): 2</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>NUMA node(s): 2</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Vendor ID: GenuineIntel</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>CPU family: 6</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Model: 85</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Model name: Intel(R) Xeon(R) Platinum
8168 CPU @ 2.70GHz</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Stepping: 4</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>CPU MHz: 2700.000</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>BogoMIPS: 5400.00</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Virtualization: VT-x</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>L1d cache: 32K</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>L1i cache: 32K</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>L2 cache: 1024K</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>L3 cache: 33792K</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>NUMA node0 CPU(s):
0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>NUMA node1 CPU(s):
1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>Flags: fpu vme de pse tsc msr pae
mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush
dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx
pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs
bts rep_good nopl xtopology nonstop_tsc aperfmperf
eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx
est tm2 ssse3 fma cx16 xtpr pdcm pcid dca sse4_1
sse4_2 x2apic movbe popcnt tsc_deadline_timer aes
xsave avx f16c rdrand lahf_lm abm 3dnowprefetch epb
cat_l3 cdp_l3 intel_pt tpr_shadow vnmi flexpriority
ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2
erms invpcid rtm cqm mpx rdt_a avx512f avx512dq
rdseed adx smap clflushopt clwb avx512cd avx512bw
avx512vl xsaveopt xsavec xgetbv1 cqm_llc
cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida
arat pln pts</div>
</div>
</div>
</div>
</blockquote>
<div dir="ltr">
<div dir="ltr">
<div><br>
</div>
<div dir="ltr"><b>slrum.conf SelectType Configuration</b><br>
</div>
</div>
</div>
</div>
<blockquote style="margin:0px 0px 0px 40px;border:none;padding:0px">
<div>
<div>
<div>
<div>SelectType=select/cons_res</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>SelectTypeParameters=CR_Core_Memory</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>PartitionName=defq Default=YES MinNodes=1
AllowGroups=ALL PriorityJobFactor=1 PriorityTier=1
DisableRootJobs=NO RootOnly=NO Hidden=NO Shared=NO
GraceTime=0 PreemptMode=OFF ReqResv=NO
AllowAccounts=ALL AllowQos=ALL LLN=NO ExclusiveUser=NO
OverSubscribe=YES OverTimeLimit=0 State=UP
Nodes=cnode[001-020]</div>
</div>
</div>
</div>
</blockquote>
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div style="font-weight:bold"><br>
</div>
<div style="font-weight:bold"><br>
</div>
<div>I can provide other configs if you feel that
it could help.</div>
<div><br>
</div>
<div>Any ideas? I would have thought that slurm
would grab the CPU information from the CPU instead of
the configuration.</div>
<div><br>
</div>
<div>Thanks</div>
<div>Chris</div>
<blockquote style="margin:0px 0px 0px 40px;border:none;padding:0px">
<div>
<div style="font-weight:bold"><br>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
<br>
</div>
</blockquote></div>