[Swift-commit] r8284 - SwiftApps/subjobs

[ketan at ci.uchicago.edu]

Author: ketan
Date: 2014-10-22 11:05:08 -0500 (Wed, 22 Oct 2014)
New Revision: 8284

Modified:
   SwiftApps/subjobs/README.txt
   SwiftApps/subjobs/runswift.sh
Log:
doc changes

Modified: SwiftApps/subjobs/README.txt
===================================================================
--- SwiftApps/subjobs/README.txt	2014-10-22 15:29:51 UTC (rev 8283)
+++ SwiftApps/subjobs/README.txt	2014-10-22 16:05:08 UTC (rev 8284)
@@ -4,24 +4,29 @@
 Introduction
 ------------
 This document describes an approach to run multiple jobs over a single large
-block on BlueGene systems. The technique, called sub-block jobs lets users
-submit multiple, independent, varying size jobs within a single larger cobalt
-block. Sub-block jobs is a mode of running jobs on BlueGene systems wherein
-one can allocate a larger block of compute nodes and repeatedly submit jobs of
-smaller sized sub-blocks to this larger allocation.
+block of compute nodes on BlueGene/Q systems.
 
-This package provides tools, scripts and example use-cases to run Swift
-applications in sub-block mode over the BlueGene resources. The benefit of
-this approach is that the user does not have to invoke the sub-block specific
-routines involving the details of the underlying node interconnect hardware.
-The approach transparently allow user to run jobs directly via Swift.
-Additionally, users can run multiple 'waves' of jobs asynchronously and in
-parallel without restarting the outer block.
+The technique, called sub-block jobs lets users submit multiple, independent,
+repeated jobs within a single larger Cobalt block. Sub-block jobs is a mode of
+running jobs on BlueGene/Q systems wherein one can allocate a larger "outer"
+block of compute nodes and repeatedly submit jobs of smaller sized sub-blocks
+to this block.
 
+The current package provides tools, scripts and example use-cases to run Swift
+applications in sub-block mode over the ALCF BlueGene/Q resources: +Vesta+,
++Cetus+ and +Mira+. The benefit of this approach is that the user does not have
+to invoke the sub-block specific routines involving the details of the
+underlying node interconnect hardware. Additionally, with the same Swift script
+and configuration, user gets a flexibility to run jobs in sub-block or
+non-sub-block mode depending on the scale and size of a run. The approach
+transparently allows user to run jobs directly via Swift. Users can run
+multiple 'waves' of jobs asynchronously and in parallel without restarting the
+outer block.
+
 Swift sub-block jobs
 ---------------------
 
-To download the package, checkout the code as follows:
+To download the package, checkout the directory as follows:
 
 ----
 svn co https://svn.ci.uchicago.edu/svn/vdl2/SwiftApps/subjobs
@@ -50,7 +55,7 @@
 
 To convert an ordinary Swift application run in sub-block mode, the following changes are required:
 
-. Add bg.sh as the application invoker in place of sh. For example, if the app definition is as follows:
+First, Add bg.sh as the application invoker in place of +sh+ or any other invoker. For example, if the app definition is as follows:
 
 ----
 sh @exe @i @o arg("s","1") stdout=@sout stderr=@serr;
@@ -62,7 +67,7 @@
 bg.sh @exe @i @o arg("s","1") stdout=@sout stderr=@serr;
 ----
 
-. Add the +SUBBLOCK_SIZE+ environment variable to the sites file. For example:
+Second, add the +SUBBLOCK_SIZE+ environment variable to the sites file. For example:
 
 ----
 <profile key="SUBBLOCK_SIZE" namespace="env">16</profile>
@@ -70,7 +75,7 @@
 
 NOTE: The value of +SUBBLOCK_SIZE+ variable must be a power of 2 greater than 8 and less than the +maxnodes+ value.
 
-A complete example sites file for a sub-block job run on ALCF Vesta is shown below:
+A complete example sites file for a sub-block job run on ALCF +Vesta+ is shown below:
 
 ----
 <?xml version="1.0" encoding="UTF-8"?>
@@ -101,15 +106,15 @@
 
 Of note are the +SUBBLOCK_SIZE+ and the +mode+ properties which must be present
 in the sites definition. The former defines the size of the subblock needed and
-the latter specifies that the "mode" to run the outer cobalt job woul be in
+the latter specifies that the "mode" to run the outer cobalt job would be
 +script+ mode. In this particular example, we have the outer block size to be
-256 nodes where as the subblock size is 16 nodes. This results in a total of 16
+256 nodes whereas the subblock size is 16 nodes. This results in a total of 16
 subblocks resulting in +jobsPerNode+ value to be 16.
 
 Use-Case Applications
 ----------------------
 
-This section discusses the real-world use-case applications that are set up
+This section discusses some of the real-world use-case applications that are set up
 with this package. These applications are tested with subblock and non-subblock
 runs on ALCF Vesta, a 2-rack (2048 nodes) BlueGene/Q system.
 

Modified: SwiftApps/subjobs/runswift.sh
===================================================================
--- SwiftApps/subjobs/runswift.sh	2014-10-22 15:29:51 UTC (rev 8283)
+++ SwiftApps/subjobs/runswift.sh	2014-10-22 16:05:08 UTC (rev 8284)
@@ -7,7 +7,7 @@
 mpixlc mpicatnap.c -o mpicatnap
 
 #New method
-/home/ketan/swift-0.95/cog/modules/swift/dist/swift-svn/bin/swift -sites.file cobalt.xml -config cf -tc.file apps catsnsleepmpi.swift -n=600 -s=200
+/home/ketan/swift-0.95/cog/modules/swift/dist/swift-svn/bin/swift -sites.file cobalt.xml -config cf -tc.file apps catsnsleepmpi.swift -n=64 -s=2
 
 #A testblock run
 #qsub -A ATPESC2013 -n 32 -t 5 -q low --mode c16 --mode script testblock.sh # --disable_preboot