Minor reformatting work for SIGOPS.

doc/common/project.bib

@@ -78,10 +78,10 @@
     AUTHOR    = {D. Jackson and Q. Snell and M. Clement},
     TITLE     = {{Core Algorithms of the Maui Scheduler}},
     BOOKTITLE = {Job Scheduling Stategies for Parallel Processing},
-    PUBLISHER = {Springer-Verlag}
+    PUBLISHER = {Springer-Verlag},
     VOLUME    = {2221},
     PAGES     = {87-102},
-    ADDRESS   = {7th International Workshop, JSSP 2001, Cambridge, MA}, 
+    ADDRESS   = {{7th International Workshop, JSSP 2001, Cambridge, MA}}, 
     YEAR      = {2001},
 }
 
@@ -99,7 +99,7 @@
 @CONFERENCE
 {
     Jones2003, 
-    AUTHOR      = {T. Jones and S. Dawson, and R. Neely, and W. Tuel and L. Brenner and J. Fier and R. Blackmore and P. Caffrey and B. Maskell and P. Tomlinson and M. Roberts},
+    AUTHOR      = {T. Jones and S. Dawson and R. Neely and W. Tuel and L. Brenner and J. Fier and R. Blackmore and P. Caffrey and B. Maskell and P. Tomlinson and M. Roberts},
     TITLE       = {{Improving scalability of parallel jobs by adding parallel awareness to the operating system}},
     BOOKTITLE   = {Proceedings of Supercomputing 2003},
     ADDRESS     = {{Phoenix, AZ}},
@@ -213,7 +213,7 @@
     PUBLISHER = {Springer-Verlag},
     VOLUME    = {2862},
     PAGES     = {44-60},
-    ADDRESS   = {9th International Workshop, JSSP 2003, Seattle, WA}, 
+    ADDRESS   = {{9th International Workshop, JSSP 2003, Seattle, WA}}, 
     YEAR      = {2003},
 }
 

doc/sigops/Makefile

@@ -49,5 +49,5 @@ view: $(REPORT).ps
 	ghostview $(REPORT) &
 
 clean: 
-	rm -f *~ *.dvi *.log *.aux $(REPORT).ps *.blg *.bbl #*.eps #*.gif *.ps
+	rm -f *~ *.dvi *.log *.aux *.ps *.pdf *.blg *.bbl #*.eps #*.gif
 	      

doc/sigops/bio.txt

-Morris Jette is a computer scientist with the Integrated 
-Computational Resource Management Group at Lawrence Livermore 
-National Laboratory. His primary research interest computer 
-scheduling, from individual tasks and processors to distributed 
-applications running across a computational grid.
-
-Mark Grondona is a computer scientist with the Production 
-Linux Group at Lawrence Livermore National Laboratory, where 
-he works on cluster administration tools and resource management.

doc/sigops/report.tex

@@ -128,7 +128,8 @@ workload prioritization.
 \item {\tt Open Source}: SLURM is available to everyone and 
 will remain free.
 Its source code is distributed under the GNU General Public
-License~\cite{GPL2002}.
+License
+~\cite{GPL2002}.
 
 \item {\tt Portability}: SLURM is written in the C language, 
 with a GNU {\em autoconf} configuration engine.
@@ -235,7 +236,7 @@ While allocation of entire nodes to jobs is still a recommended mode of
 operation for very large clusters, an alternate SLURM plugin provides 
 resource management down the the resolution of individual processors.
 
-The SLURM's {\tt srun} command and  daemons are extensively 
+The SLURM's {\tt srun} command and the daemons are extensively 
 multi-threaded. 
 {\tt slurmctld} also maintains independent read and 
 write locks for critical data structures. 
@@ -266,8 +267,8 @@ This improves performance by distributing the communication workload.
 Note that every communication is authenticated and acknowleged 
 for fault-tolerance.
 
-A number of interesting papers 
-~\cite{Jones2003,Kerbyson2001,Petrini2003,Phillips2003,Tsafrir2005}
+A number of interesting papers
+~\cite{Jones2003,Kerbyson2001,Petrini2003,Phillips2003,Tsafrir2005} 
 have recently been written about
 the impact of system daemons and other system overhead on 
 parallel job performance. This {\tt system noise} can have a 
@@ -287,7 +288,7 @@ SLURM addresses this issue by:
 \item Making the {\tt slurmd} daemon resource requirements negligible 
 \item Supporting configurations that let the {\tt slurmd} daemon sleep 
 during the entire job execution period
-\item If the {\tt slurmd} daemons are given work, it is done on a 
+\item If the {\tt slurmd} daemons do perform work, it is done on a 
 highly synchronized fashion across all nodes
 \end{itemize}
 In addition, the default mode of operation is to allocate entire 
@@ -299,7 +300,6 @@ Allocation of resources to the resolution of individual processors
 on each node is supported by SLURM, but this comes at a higher cost 
 in terms of the data managed.
 The selection of resource resolution is provided by different plugins.
-These factors permit SLURM to effectively run highly parallel jobs.
 
 Resource management of large clusters entails the processing of 
 large quantities of data, both for the software and the