Jun-2018

• Hardware

  • No news

• Software

  • Updated packages
    • Key: intel/2018.1
      Description: Intel Cluster Toolkit Compiler Edition provides Intel C,C++ and fortran compilers, Intel MPI and Intel MKL -
    • Key: pgi/17.10
      Description: The Portland Group compilers for C, C++ and Fortran. It also supports OpenACC. http://www.pgroup.com/resources/docs.htm
    • Key: pgi/18.1
      Description: The Portland Group compilers for C, C++ and Fortran. It also supports OpenACC. http://www.pgroup.com/resources/docs.htm

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Jan-2018

• Hardware

  • No news

• Software

  • Updated packages
    • Key: ansys/18.1
      Description: The ANSYS software suites include Fluent, CFX and ANSYS Mechanical. It's only available to LSU users.
    • Key: impi/2017.3.196/intel64
      Description: This is an intel MPI implementation. It is included in the Intel-2017.4 compiler. It is a good compiler for MPI programming on Xeon Phi.
    • Key: impi/2018.0.128/intel64
      Description: This is an intel MPI implementation. It is included in the Intel-2018.0 compiler. It is a good compiler for MPI programming on Xeon Phi.
    • Key: python/3.6.2-anaconda-tensorflow
      Description: Python is a programming language that lets you work more quickly and integrate your systems more effectively. -

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Oct-2017

• Hardware

  • no news

• Software

  • Updated packages
    • Key: amber/16/INTEL-140-MVAPICH2-2.0
      Description: "Amber" refers to two things: a set of molecular mechanical force fields for the simulation of biomolecules (which are in the public domain, and are used in a variety of simulation programs); and a package of molecular simulation programs which includes source code and demos.
    • Key: cuda/8.0
      Description: Compute Unified Device Architecture (CUDA) is a parallel computing architecture developed by Nvidia for graphics processing
    • Key: fftw/3.3.6/INTEL-140-MVAPICH2-2.0
      Description: FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions, of arbitrary input size, and of both real and complex data.
    • Key: gromacs/5.1.4/INTEL-140-MVAPICH2-2.0
      Description: GROMACS (GROningen MAchine for Chemical Simulations) is a versatile package to perform molecular dynamics with Plumed installed.
    • Key: intel/2016.3
      Description: Intel Cluster Toolkit Compiler Edition provides Intel C,C++ and fortran compilers, Intel MPI and Intel MKL
    • Key: intel/2017.4
      Description: Intel Cluster Toolkit Compiler Edition provides Intel C,C++ and fortran compilers, Intel MPI and Intel MKL
    • Key: intel/2018.0
      Description: Intel Cluster Toolkit Compiler Edition provides Intel C,C++ and fortran compilers, Intel MPI and Intel MKL

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Jul-2017

• Hardware

  • no news

• Software

  • Newly installed packages
    • Key: pigz/2.3.4
      Description: pigz is parallel implementation of gzip, is a fully functional replacement for gzip that exploits multiple processors and multiple cores to the hilt when compressing data.
    • Key: remora/1.8
      Description: Resource Monitoring for Remote Applications
    • Key: xdinfo/1.3-1
      Description: Adds XSEDE Information Services Think Client (xdinfo) tool to paths remove-path PATH /usr/local/packages/xsede/xdinfo-1.3-1/bin
  • Updated packages
    • Key: matlab/r2017a
      Description:
    • Key: python/2.7.13-anaconda-tensorflow
      Description: Python is a programming language that lets you work more quickly and integrate your systems more effectively. -
    • Key: gateway-usage-reporting/2.0
      Description: adds gateway-submit_attributes tool to paths in the login shell environment remove-path PATH /etc/globus/gateway_submit_attributes-2.0
    • Key: xdusage/2.0-3
      Description: adds XSEDE xdusage tool to paths in the login shell environment remove-path PATH /usr/local/packages/xsede/xdusage/xdusage-2.0-3/bin

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Mar-2017

• Hardware

  • A "single" queue is added to the queuing system, whereas ppn is allowed to be set as 1, 2, 4, 6 and 8 cores
  • The "bigmem" queue has 3 nodes with 256GB RAM each

• Software

  • Newly installed packages
    • Key: gnuparallel/20170122
      Description: GNU parallel is a shell tool for executing jobs in parallel using one or more computers
    • Key: swift/0.96.2
      Description: Swift is a simple scripting language for executing many instances of ordinary application programs on distributed and parallel resources.
  • Updated packages
    • Key: cuda/7.5
      Description: Compute Unified Device Architecture (CUDA) is a parallel computing architecture developed by Nvidia for graphics processing
  • Retired packages
    • Key: globus/globus-gram5-xsede.module
      

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • At the moment, Intel Xeon Phi cards support native and offload mode, but the symmetric mode is not working properly

Jan-2017

• OS Update

• Upgraded to RHEL 6.8 with the latest security patches

• Hardware

  • Bigmem Queue (3 big memory nodes) is added: Each node has 28 cores, 256GB Ram, 1TB HD

• Software

  • No Change

• Maintenance Scheduled

  • Downtime to replace batteries on Tuesday Jan 31

• Known Problems

  • At the moment, Intel Xeon Phi cards support native and offload mode, but the symmetric mode is not working properly. We are working on it.

Jun-2016

• Hardware

  • no news

• Software

  • Newly installed packages
    • Key: amos/3.1.0/INTEL-14.0.2
      Description: The AMOS consortium is committed to the development of open-source whole genome assembly software -
    • Key: blat/0.35/INTEL-14.0.2
      Description: Blat is an alignment tool like BLAST, but it is structured differently. On DNA, Blat works by keeping an index of an entire genome in memory. Thus, the target database of BLAT is not a set of GenBank sequences, but instead an index derived from the assembly of the entire genome. Blat is commonly used to look up the location of a sequence in the genome or determine the exon structure of an mRNA, but expert users can run large batch jobs and make internal parameter sensitivity changes by installing command line Blat on their own Linux server. -
    • Key: bwa/0.7.4/INTEL-14.0.2
      Description: Burrows-Wheeler Aligner (BWA) is an efficient program that aligns relatively short nucleotide sequences against a long reference sequence such as the human genome. -
    • Key: expat/2.1.0/INTEL-14.0.2
      Description: Expat is an XML parser library written in C. It is a stream-oriented parser in which an application registers handlers for things the parser might find in the XML document (like start tags) -
    • Key: ray/2.3.1/INTEL-140-MVAPICH2-2.0
      Description: Ray is a parallel software that computes de novo genome assemblies with next-generation sequencing data. -
    • Key: rnnotator/3.4.0
      Description: Rnnotator is an automated software pipeline that generates transcript models by de novo assembly of RNA-Seq data without the need for a reference genome. The contigs produced by Rnnotator are highly accurate and reconstruct full-length genes when transcripts are sequenced sufficiently deep, roughly 30X for a given transcript. Rnnotator was designed to assemble Illumina single or paired-end reads. Rnnotator is also able to incorporate strand-specific RNA-Seq reads into the assembly in order to further improve the assembly. -
    • Key: sparsehash/2.0.2/INTEL-14.0.2
      Description: An extremely memory-efficient hash_map implementation. 2 bits/entry overhead! The SparseHash library contains several hash-map implementations, including implementations that optimize for space or speed. -
  • Updated packages
    • Key: abyss/1.9.0/INTEL-140-MVAPICH2-2.0
      Description: Assembly By Short Sequences - a de novo, parallel, paired-end sequence assembler -
    • Key: ansys/17.0
      Description: The ANSYS software suites include Fluent, CFX and ANSYS Mechanical. It's only available to LSU users.
    • Key: gromacs/5.0.7/INTEL-140-OPENMPI-1.8.4
      Description: GROMACS (GROningen MAchine for Chemical Simulations) is a versatile package to perform molecular dynamics.
    • Key: wq/272
      Description: WQ is a support tool that creates supports a dispatcher/worker model of distributed computing. Serial, multi-thread OpenMP, and small multi-process MPI are supported. Tasks are defined by input file names, or other command line strings, contained in an input file - one line per task.
  • Retired packages
    • Key: openmpi/1.8.1/INTEL-14.0.2
      

• Maintenance Scheduled

  • System will be offline for file system update. Downtime might last for a couple of weeks. Details are coming soon

• Known Problems

  • No impact on users

Apr-2016

• Hardware

  • no news

• Software

  • Newly installed packages
    • Key: bbcp/15.02.03
      Description: BBCP is an alternative to Gridftp when transferring large amounts of data, capable of breaking up your transfer into multiple simultaneous transferring streams, thereby transferring data much faster than single-streaming utilities such as SCP and SFTP. -
    • Key: wq/264
      Description: WQ is a support tool that creates supports a dispatcher/worker model of distributed computing. Serial, multi-thread OpenMP, and small multi-process MPI are supported. Tasks are defined by input file names, or other command line strings, contained in an input file - one line per task.
  • Updated packages
    • Key: pgi/15.10
      Description: The Portland Group compilers for C, C++ and Fortran. It also supports OpenACC. http://www.pgroup.com/resources/docs.htm
    • Key: globus/5.7-8
      Description: globus gram5 xsede 5.7-8
    • Key: globus/globus-gram5-xsede.module
      Description: globus gram5 xsede 5.7-8
  • Retired packages
    • Key: globus/5.7-6
      

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Jan-2016

• Hardware

  • no news

• Software

  • Newly installed packages
    • Key: gaussian/g09-d01
      Description:
    • Key: octave/3.8.2/INTEL-14.0.2
      Description: Description: GNU Octave is a high-level interpreted language, primarily intended for numerical computations. -
    • Key: pcre/8.36/INTEL-14.0.2
      Description: Description:
    • Key: vtune/2015.3.0
      Description: This is an intel VTune Amplifier XE for parallel_studio-2015u3
    • Key: gateway-usage-reporting/1.0
      Description: adds gateway-usage-reporting tool to paths in the login shell environment
  • Updated packages
    • Key: ansys/16.0
      Description: Description:
    • Key: espresso/5.2.1/INTEL-14.0.2-mvapich-2.0
      Description: Quantum Espresso is an integrated suite of Open-Source computer codes for electronic-structure calculations and materials modeling at the nanoscale. It is based on density-functional theory, plane waves, and pseudopotentials.
    • Key: impi/5.0.1.035/intel64
      Description: This is an intel MPI implementation. It is included in the Intel-15.0.1 compiler. It is a good compiler for MPI programming on Xeon Phi.
    • Key: mathematica/10.1
      Description: Description:
    • Key: matlab/r2015b
      Description: Description:
    • Key: python/2.7.10-mkl-mic
      Description: Description: Python is a programming language that lets you work more quickly and integrate your systems more effectively. Within this version, the numpy/scipy packages are linked to Intel Math Kernel Library (MKL), so that they can be accelerated with Xeon Phi. -
    • Key: globus/5.7-6
      Description: globus gram5 xsede 6.0-1
    • Key: perl/5.22.0
      Description: Description: Larry Wall's Practical Extraction and Report Language -
    • Key: xsede/1.1
      Description:
  • Retired packages
    • Key: mathematica/10.0
      

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Aug-2015

• Hardware

  • no news

• Software

  • Newly installed packages
    • Key: amber/14/INTEL-140-MVAPICH2-2.0
      Description: "Amber" refers to two things: a set of molecular mechanical force fields for the simulation of biomolecules (which are in the public domain, and are used in a variety of simulation programs); and a package of molecular simulation programs which includes source code and demos.
    • Key: boost/1.55.0/INTEL-14.0.2
      Description: Boost provides free peer-reviewed portable C++ source libraries. -
    • Key: boost/1.55.0/INTEL-14.0.2-mic
      Description: Boost provides free peer-reviewed portable C++ source libraries. Intel Xeon Phi version (mic) - URL: http://www.boost.org/
    • Key: espresso/5.1.2/INTEL-140-OPENMPI-1.8.4
      Description: Quantum Espresso is an integrated suite of Open-Source computer codes for electronic-structure calculations and materials modeling at the nanoscale. It is based on density-functional theory, plane waves, and pseudopotentials.
    • Key: hpx/0.9.10/impi-4.1.3.048-intel64
      Description: HPX is a general purpose C++ runtime system for parallel and distributed applications of any scale. - Built with Intel MPI for CPU only
    • Key: hpx/0.9.10/impi-4.1.3.048-intel64-mic
      Description: HPX is a general purpose C++ runtime system for parallel and distributed applications of any scale. - Intel Xeon Phi (mic) version built with impi
    • Key: hpx/0.9.10/mvapich2-2.0-INTEL-14.0.2
      Description: HPX is a general purpose C++ runtime system for parallel and distributed applications of any scale. - Build with MVAPICH2
    • Key: paraview/4.3.1/GCC-4.9.0
      Description: An open-source, multi-platform data analysis and visualization application. -
    • Key: paraview/4.3.1/INTEL-140-MVAPICH2-2.0
      Description: An open-source, multi-platform data analysis and visualization application. -
    • Key: siesta/3.2/INTEL-140-OPENMPI-1.8.4
      Description: SIESTA is both a method and its computer program implementation, to perform efficient electronic structure calculations and ab initio molecular dynamics simulations of molecules and solids.
  • Updated packages
    • Key: blast/2.2.31/INTEL-14.0.2
      Description: Basic Local Alignment Search Tool, or BLAST, is an algorithm
    • Key: pgi/14.10
      Description: The Portland Group compilers for C, C++ and Fortran. It also supports OpenACC. http://www.pgroup.com/resources/docs.htm
    • Key: pgi/15.4
      Description: The Portland Group compilers for C, C++ and Fortran. It also supports OpenACC. http://www.pgroup.com/resources/docs.htm
    • Key: globus/6.0
      Description: globus client xsede 6.0-1

• Maintenance Scheduled

  • Not scheduled

• Known Problems

  • No impact on users

Feb 2015

• Software

  • Updated
    • NAMD version 2.10b2 with Xeon Phi support installed.
      Module: namd/2.10b2/INTEL-14.0.2-ibverbs-mic
    • LAMMPS 21Jan15 with Xeon Phi support installed.
      Module: lammps/21Jan15/INTEL-14.0.2-impi-4.1.3.048-mic
    • Matlab version r2013a
      Module: matlab/r2013a
    • Mathematica version 10.0
      Module: mathematica/10.0
  • Retired
    • Module: namd/2.10b2/INTEL-14.0.2-ibverbs
    • Module: namd/2.10b2/INTEL-14.0.2-ibverbs-mic
  • Benchmark Results
    

    NAMD APOA1 Benchmark:
    
    NumberOfNodes   CPUonly  CPU+Phi  Speedup
    1  0.621938 days/ns  0.254949 days/ns  2.44
    2  0.471777 days/ns  0.142642 days/ns  3.31
    4  0.310066 days/ns  0.0823795 days/ns  3.76
    8  0.220733 days/ns  0.0601484 days/ns  3.67
    
    NAMD STMV Benchmark:
    
    NumberOfNodes   CPUonly  CPU+Phi  Speedup
    1  6.22028 days/ns  2.42398 days/ns  2.57
    2  3.35414 days/ns  1.37301 days/ns  2.44
    4  1.96454 days/ns  0.858824 days/ns  2.29
    8  1.28667 days/ns  0.510283 days/ns  2.51
    16  0.652524 days/ns  0.360255 days/ns  1.81
    
    LAMMPS Rhodopsin Benchmark:
    
    Number of Nodes;Without package (Baseline);OMP package;
       INTEL+OMP;Offload to Phi-Auto Balance;Offload to Phi-Fixed Balance
    1 node	Loop time (sec)	28.5529	24.1188	20.7227	13.6826	12.7717
    32 node	Loop time (sec)	34.8826	30.0743	25.7038	18.0796	17.5099
    1 node	Speedup	1	1.18	1.38	2.09	2.24
    32 node	Speedup	1	1.16	1.36	1.93	1.99
    
    LAMMPS Liquid Crystal Benchmark:
    1 node	Loop time (sec)	26.9558	26.4598	6.80273	4.97776	4.64265
    32 node	Loop time (sec)	26.9836	26.4459	7.7801	6.27055	7.63933
    1 node	Speedup	1	1.02	3.96	5.42	5.81
    32 node	Speedup	1	1.02	3.47	4.30	3.53
    

• Maintenance Scheduled

  • None scheduled.

• Known Problems

  • None impacting users