<?xml version="1.0" encoding="UTF-8"?>
<!-- generator="FeedCreator 1.8" -->
<?xml-stylesheet href="https://docs-archive.hpc.udel.edu/lib/exe/css.php?s=feed" type="text/css"?>
<rss version="2.0">
    <channel xmlns:g="http://base.google.com/ns/1.0">
        <title>hpc documentation - technical:whitepaper</title>
        <description></description>
        <link>https://docs-archive.hpc.udel.edu/</link>
        <lastBuildDate>Tue, 14 Apr 2026 22:00:56 +0000</lastBuildDate>
        <generator>FeedCreator 1.8</generator>
        <image>
            <url>https://docs-archive.hpc.udel.edu/_media/wiki/dokuwiki.svg</url>
            <title>hpc documentation</title>
            <link>https://docs-archive.hpc.udel.edu/</link>
        </image>
        <item>
            <title>automated_devshm_cleanup</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/automated_devshm_cleanup</link>
            <description>Automated /dev/shm cleanup

Historically, shared memory was an IPC facility whereby one process could mark memory segments as being shared, and other processes on the same machine could map those segments into their memory space.  Segments had a key that identified them, and the key had to be shared to dependent processes that wanted to map the segment.  This was often accomplished by:</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Thu, 13 Dec 2018 18:02:26 +0000</pubDate>
        </item>
        <item>
            <title>darwin_ucx_openmpi</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/darwin_ucx_openmpi</link>
            <description>Mellanox UCX and Open MPI on DARWIN

During early-access testing of the DARWIN cluster, several users reported unexpected crashes of their Open MPI applications.  The crashes were accompanied by a running stream of kernel messages:


  :
[Sat Feb  6 16:51:55 2021] infiniband mlx5_0: create_mkey_callback:148:(pid 0): async reg mr failed. status -12
[Sat Feb  6 16:51:55 2021] mlx5_core 0000:81:00.0: mlx5_cmd_check:794:(pid 0): CREATE_MKEY(0x200) op_mod(0x0) failed, status limits exceeded(0x8), syn…</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Fri, 12 Feb 2021 22:15:41 +0000</pubDate>
        </item>
        <item>
            <title>hpcc-open64-acml</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/hpcc-open64-acml</link>
            <description>HPCC with open64 compiler, ACML and base FFT

Make

Here are som modifications on Mills based on recommendations from  Using ACML (AMD Core Math Library) In High Performance Computing Challenge (HPCC)

Changes to the make file hpl/setup/Make.Linux_ATHLON_FBLAS copied to hpl/Make.open64-acml

	*  Comment lines beginning in MP or LA</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Tue, 27 Apr 2021 20:21:09 +0000</pubDate>
        </item>
        <item>
            <title>openmpi-psm2-spawn</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/openmpi-psm2-spawn</link>
            <description>Open MPI, PSM2, and MPI_Comm_spawn()

MPI programs have always started with the worker processes described as the world for the computation (MPI_COMM_WORLD).  The MPI-2 standard includes the concept of something larger than that world:  the universe encompasses all processing resources available, where the</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Tue, 29 Mar 2022 15:45:19 +0000</pubDate>
        </item>
        <item>
            <title>r-runtime-blas-lapack</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/r-runtime-blas-lapack</link>
            <description>R: Runtime-configuration BLAS/LAPACK

The R Project for Statistical Computing is used on our clusters by a wide variety of scientific disciplines.  Though the breadth of applications is wide, many of them require the functionality of BLAS/LAPACK libraries.  R provides its own baseline implementations that will build on any system; naturally, one cannot expect these</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Mon, 10 Dec 2018 17:40:38 +0000</pubDate>
        </item>
        <item>
            <title>rjava-gcc-optimization</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/rjava-gcc-optimization</link>
            <description>rJava: when compilers get too smart

In the midst of installing the thousands of packages available in the CRAN library, a subset of the packages displayed a curious behavior:  when R attempted to load the package to verify installation, the R subprocess would hang indefinitely.  With timeouts in place, the installation of the package eventually fails.</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Fri, 17 Jun 2022 18:58:18 +0000</pubDate>
        </item>
        <item>
            <title>start</title>
            <link>https://docs-archive.hpc.udel.edu/technical/whitepaper/start</link>
            <description>White Papers

Some of the content in this area will be in PDF format and may need to be downloaded before being read.

rJava:  When Compilers Get Too Smart

While installing all 16k (nearly 17k) CRAN packages on a recent R 4.1.3 build, many packages with a dependency on rJava would hang when being tested.</description>
            <author>anonymous@undisclosed.example.com (Anonymous)</author>
            <pubDate>Fri, 17 Jun 2022 16:25:18 +0000</pubDate>
        </item>
    </channel>
</rss>
