Merge pull request #180 from jedwards4b/refactor/config_batch

split config_batch.xml per machine

machines/aleph/config_batch.xml

@@ -0,0 +1,10 @@
+<batch_system MACH="aleph" type="pbs" >
+  <directives>
+    <directive>-l nodes={{ num_nodes }}</directive>
+    <directive>-q iccp</directive>
+    <directive> -V </directive>
+  </directives>
+  <queues>
+    <queue walltimemax="24:00:00" default="true" >iccp</queue>
+  </queues>
+</batch_system>

machines/athena/config_batch.xml

@@ -0,0 +1,20 @@
+<!-- athena is lsf -->
+<batch_system MACH="athena" type="lsf">
+  <submit_args>
+    <argument> -q $JOB_QUEUE </argument>
+    <argument> -W $JOB_WALLCLOCK_TIME </argument>
+    <argument> -P $PROJECT </argument>
+  </submit_args>
+  <directives>
+    <directive                       > -R "span[ptile={{ tasks_per_node }}]"</directive>
+    <directive                       > -N  </directive>
+    <directive default="poe"         > -a {{ poe }} </directive>
+  </directives>
+  <queues>
+    <queue walltimemin="00:00" walltimemax="02:00">poe_short</queue>
+    <queue walltimemin="02:00" walltimemax="04:00" default="true">poe_medium</queue>
+    <queue walltimemin="04:00" walltimemax="08:00">poe_long</queue>
+  </queues>
+</batch_system>
+
+

machines/aws-hpc6a/config_batch.xml

@@ -0,0 +1,12 @@
+<batch_system MACH="aws-hpc6a" type="slurm">
+  <batch_submit>sbatch</batch_submit>
+  <submit_args>
+    <argument> --time $JOB_WALLCLOCK_TIME </argument>
+    <argument> -p $JOB_QUEUE </argument>
+  </submit_args>
+  <queues>
+    <queue walltimemax="144:00:00" nodemin="1" nodemax="96">regular</queue>
+    <queue walltimemax="1:00:00"   nodemin="1" nodemax="1">build</queue>
+    <queue default="true" walltimemax="4:00:00"   jobmin="1" jobmax="1">serial</queue>
+  </queues>
+</batch_system>

machines/casper/config_batch.xml

@@ -0,0 +1,21 @@
+<!-- casper pbs -->
+<batch_system MACH="casper" type="pbs">
+  <batch_submit>qsub</batch_submit>
+  <submit_args>
+    <argument> -l gpu_type=$GPU_TYPE </argument>
+  </submit_args>
+  <directives queue="casper" compiler="nvhpc" gpu_enabled="true">
+    <directive default="/bin/bash" > -S {{ shell }} </directive>
+    <directive> -l select={{ num_nodes }}:ncpus={{ max_tasks_per_node }}:mpiprocs={{ tasks_per_node }}:ompthreads={{ thread_count }}:mem=700GB:ngpus={{ ngpus_per_node }}:mps=1 </directive>
+  </directives>
+  <directives queue="casper" gpu_enabled="false">
+    <directive default="/bin/bash" > -S {{ shell }} </directive>
+    <directive> -l select={{ num_nodes }}:ncpus={{ max_tasks_per_node }}:mpiprocs={{ tasks_per_node }}:ompthreads={{ thread_count }}:mem=300GB:ngpus={{ ngpus_per_node }} </directive>
+  </directives>
+  <!-- Unknown queues use the batch directives for the regular queue -->
+  <unknown_queue_directives>casper</unknown_queue_directives>
+  <queues>
+    <queue walltimemax="12:00:00" nodemin="1" nodemax="10">casper</queue>
+  </queues>
+</batch_system>
+

machines/coeus/config_batch.xml

@@ -0,0 +1,24 @@
+  <!-- coeus slurm -->
+  <batch_system MACH="coeus" type="slurm" >
+    <batch_query per_job_arg="-j">squeue</batch_query>
+    <batch_submit>sbatch</batch_submit>
+    <batch_cancel>scancel</batch_cancel>
+    <batch_directive>#SBATCH</batch_directive>
+    <jobid_pattern>(\d+)$</jobid_pattern>
+    <depend_separator>,</depend_separator>
+    <walltime_format>%H:%M:%S</walltime_format>
+    <batch_mail_flag>--mail-user</batch_mail_flag>
+    <batch_mail_type_flag>--mail-type</batch_mail_type_flag>
+    <batch_mail_type>none, all, begin, end, fail</batch_mail_type>
+    <directives>
+      <directive> --job-name={{ job_id }}</directive>
+      <directive> --nodes={{ num_nodes }}</directive>
+      <directive> --ntasks-per-node={{ tasks_per_node }}</directive>
+      <directive> --output={{ job_id }}   </directive>
+      <directive> --exclusive                        </directive>
+    </directives>
+    <queues>
+      <queue nodemin="1" nodemax="96" default="true">medium</queue>
+    </queues>
+  </batch_system>
+