Saturday, November 30, 2013

SGI Receives Seven Top Honors in 2013 HPCwire Readers' and Editors' Choice Awards

SGI Receives Seven Top Honors in 2013 HPCwire Readers' and Editors' Choice Awards

Last week, SGI received seven awards for outstanding technology and implementation achievements as part of the 2013 HPCwire Readers' Choice Awards and HPCwire Editors' Choice Awards.


SGI was selected as a recipient of four HPCwire Readers’ Choice Awards, winning case studies that include:

  1. Top Supercomputing Achievement:  TOTAL S.A. and SGI “Pangea” the World’s Largest Commercial HPC Deployment for Oil and Gas Exploration
  2. Best use of HPC in Life Sciences:  Pittsburgh Supercomputing Center and SGI “Blacklight” for Next Generation Sequencing Assembly
  3. Best use of HPC in Financial Services:  PayPal and SGI, Mellanox, and DataDirect Networks technologies for fraud detection and prevention during PayPal transactions
  4. Best use of HPC in the Oil and Gas Industry:  TOTAL S.A. using SGI “Pangea” for oil and gas exploration

 HPCwire Editors’ Choice Awards selected SGI for two category wins based on customer implementation and standalone product offering, which include:

  1. Best use of HPC in Financial Services:  XSEDE PSC SGI “Blacklight” and SDSC Cray “Gordon” Research enables faster market analysis and imposes stricter regulations upon stock market trading
  2. Best HPC Storage Product or Technology:  SGI InfiniteStorage Gateway

For more information. See SGI Press release SGI Receives Seven Top Honors in 2013 HPCwire Readers' and Editors' Choice Awards

Wednesday, November 27, 2013

Intel Books24X7 Online Library

 
Key features of this unique online library configuration:
 
1. Choose two library options customized for your area of specialization:
  •  IT Pro™ (12,000 titles)
  • Engineering Pro™ (4,000 titles)
2. Search within any or all books by software source code (C, Java, C++ etc.)
3. Add notes and bookmarks (and share with your workgroup)
 
For more information, do look at the Intel Books24X7 Online Library 

Tuesday, November 26, 2013

Installing HTseq for python 26 for CentOS 6

Installing HTseq is very straightforward for CentOS 6. You will need to do just follow the installation manual

# yum install python-devel numpy python-matplotlib

Download and untar the HTSeq source files.
# tar -zxvf HTSeq-0.5.4p5.tar.gz
# cd HTSeq-0.5.4p5

Inside the expanded HTSeq home directories, to make HTSeq available for all users
# python setup.py build
# python setup.py install

For more information, do look at
  1. HTSeq Prerequisites and Installation

Monday, November 25, 2013

Error: php53-common conflicts with php-common when installing php53-xml on CentOS 5

I was trying to install php53-xml on CentOS 5.5. But instead I encountered the error instead. I', assuming  you have activated the webtatic repo and install the rest of the php 5.3 components like the How to yum install PHP 5.3 for CentOS 5

php53-common-5.3.3-21.el5.i386 from base has depsolving problems
  --> php53-common conflicts with php-common
Error: php53-common conflicts with php-common
 You could try using --skip-broken to work around the problem
 You could try running: package-cleanup --problems
                        package-cleanup --dupes
                        rpm -Va --nofiles --nodigest
The program package-cleanup is found in the yum-utils package.

The issue of the problem is that the current PHP from webtatic is 5.3.27, but the one from yum php53-xml is 5.3.3 which is higher. To solve the issue, you have to use the php-xml from webtatic

# yum install  php-xml

Wednesday, November 20, 2013

How to yum install PHP 5.3 for CentOS 5

There is interesting article from Rackspace How to install/upgrade PHP 5.3 for CentOS 5.x

Step 1: Basically, you have to install a repository "webstatic" which is found at 
# wget http://repo.webtatic.com/yum/centos/5/latest.rpm
# rpm -Uvh latest.rpm
# yum check-update

Step 2: Enable WebStatic (enabled=1)
[webtatic]
name=Webtatic Repository $releasever - $basearch
#baseurl=http://repo.webtatic.com/yum/centos/5/$basearch/
mirrorlist=http://mirror.webtatic.com/yum/centos/5/$basearch/mirrorlist
enabled=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-webtatic-andy

Step 3a: Do a Yum Install
# yum install php php-cli php-gd php-mysql php-mbstring

Step 3b: If you are upgrading, do a yum upgrade
# yum upgrade php

Step 4: Restart the Web Server
# service httpd restart

Step 5: Check the PHP version
[root@web ~]# php -v
PHP 5.3.27 (cli) (built: Jul 13 2013 22:19:28)
Copyright (c) 1997-2013 The PHP Group
Zend Engine v2.3.0, Copyright (c) 1998-2013 Zend Technologies

Friday, November 15, 2013

Fixing Authentication is requried to set the network proxy used for download packages for CentOS 6

I encountered this pop-up error today when I was xrdp into my CentOS 6.3. The error was something like this

authentication is required to set the network proxy used for downloading packages.  
An application is attempting to perform an action that requires privileges.  
Authentication as the super user is required to perform this action" and asking  
for the root password

Non-root users

Step 1: Launch a Terminal Console and type
# gnome-session-properties

Step 2: Uncheck PackageKit Update Applet .

See pix below.

Root User

Step 1: Disabled /etc/yum/pluginconf.d
# vim /etc/yum/pluginconf.d/refresh-packagekit.conf

[main]
enabled=0

Tuesday, November 12, 2013

Conflict when yum install perl-XML-SAX-0.96-7.el6.noarch on CentOS 6

When yum install perl-XML-SAX, you may encounter an error something like this

Transaction Check Error:
  file /usr/share/man/man3/XML::SAX::Base.3pm.gz conflicts between attempted 
installs of perl-XML-SAX-0.96-7.el6.noarch and perl-XML-SAX-Base-1.04-1.el6.rf.noarch
  file /usr/share/man/man3/XML::SAX::Exception.3pm.gz conflicts between attempted 
installs of perl-XML-SAX-0.96-7.el6.noarch and perl-XML-SAX-Base-1.04-1.el6.rf.noarch

The reason for this is due to the conflict of packages between the CentOS base repo (perl-XML-SAX-0.96-7)  and those found at rpmforge (perl-XML-SAX-Base-1.04-1).

The resolution is to disable the perl-XML-SAX-Base-1.04-1 found at rpmforge. You can do it by putting this line at /etc/yum.repos.d/rpmforge.repo

.....
.....
exclude=perl-XML-SAX-Base
.....
.....

Monday, November 11, 2013

Announcing the Release of MVAPICH2 2.0b, MVAPICH2-X 2.0b and OSU Micro-Benchmarks (OMB) 4.2

The MVAPICH team is pleased to announce the release of MVAPICH2 2.0b,
MVAPICH2-X 2.0b (Hybrid MPI+PGAS (OpenSHMEM) with Unified
Communication Runtime) and OSU Micro-Benchmarks (OMB) 4.2.
Features, Enhancements, and Bug Fixes for MVAPICH2 2.0b (since
MVAPICH2 2.0a release) are listed here.
* Features and Enhancements (since 2.0a):
    - Based on MPICH-3.1b1
    - Multi-rail support for GPU communication
    - Non-blocking streams in asynchronous CUDA transfers for
      better overlap
    - Initialize GPU resources only when used by MPI transfer
    - Extended support for MPI-3 RMA in OFA-IB-CH3, OFA-IWARP-CH3, and
      OFA-RoCE-CH3
    - Additional MPIT counters and performance variables
    - Updated compiler wrappers to remove application dependency
      on network and other extra libraries
        - Thanks to Adam Moody from LLNL for the suggestion
    - Capability to checkpoint CH3 channel using the Hydra process manager
    - Optimized support for broadcast, reduce and other collectives
    - Tuning for IvyBridge architecture
    - Improved launch time for large-scale mpirun_rsh jobs
    - Introduced retry mechanism in mpirun_rsh for socket binding
    - Updated hwloc to version 1.7.2
* Bug-Fixes (since 2.0a):
    - Consider list provided by MV2_IBA_HCA when scanning device list
    - Fix issues in Nemesis interface with --with-ch3-rank-bits=32
    - Better cleanup of XRC files in corner cases
    - Initialize using better defaults for ibv_modify_qp (initial ring)
    - Add unconditional check and addition of pthread library
    - MPI_Get_library_version updated with proper MVAPICH2 branding
        - Thanks to Jerome Vienne from the TACC for the report
MVAPICH2-X 2.0b software package provides support for hybrid MPI+PGAS
(UPC and OpenSHMEM) programming models with unified communication
runtime for emerging exascale systems. This software package
provides flexibility for users to write applications using the
following programming models with a unified communication runtime:
MPI, MPI+OpenMP, pure UPC, and pure OpenSHMEM programs as well as
hybrid MPI(+OpenMP) + PGAS (UPC and OpenSHMEM) programs.
Features, enhancements and bug-fixes for MVAPICH2-X 2.0b
(since MVAPICH2-X 2.0a) are as follows:
* Features and Enhancements (since 2.0a)
    - OpenSHMEM Features
        - Based on OpenSHMEM reference implementation 1.0e
        - Enhanced optimization of OpenSHMEM collectives (shmem_collect,
          shmem_fcollect, shmem_barrier, shmem_reduce, and shmem_broadcast)
        - Optimized shmalloc routine
    - UPC Features
        - Based on Berkeley UPC 2.18.0
        - Support for GUPC translator
    - MPI Features
        - Based on MVAPICH2 2.0b (OFA-IB-CH3 interface)
    - Unified Runtime Features
        - Based on MVAPICH2 2.0b (OFA-IB-CH3 interface). All the runtime
          features enabled by default in OFA-IB-CH3 interface of
          MVAPICH2 2.0b are available in MVAPICH2-X 2.0b
* Bug Fixes (since 2.0a):
    - OpenSHMEM Bug Fixes
        - Fixed synchronization issue in shmem_fence
        - Fixed issue in shmem_collect which prevented variable length
          collect routine
New features and Enhancements of OSU Micro-Benchmarks (OMB) 4.2 (since
OMB 4.1 release) are listed here.
* New Features & Enhancements
    - New OpenSHMEM benchmarks
        * osu_oshm_fcollect
    - Enable handling of GPU device buffers in all MPI collective benchmarks
    - Add device binding for OpenACC benchmarks
* Bug Fixes
    - Add upc_fence after memput in osu_upc_memput benchmark
    - Removed incorrect reuse of pSync in OpenSHMEM benchmarks
        - Thanks Sayan Ghosh for the report
    - Correct CUDA configuration example in README
    - Fix several warnings
For downloading MVAPICH2 2.0b, MVAPICH2-X 2.0b, OMB 4.2, associated
user guides, quick start guide, and accessing the SVN, please visit
the following URL:
http://mvapich.cse.ohio-state.edu

RM Failure, rc: 15041, msg: 'Execution server rejected request MSG=cannot send job to mom, state=PRERUN'

I'm using Torque 4.2.5 and MAUI 3.3.1

When I was submitting a job, I always seem to get my job deferred and stuck. I used the MAUI "checkjob jobid" function and the detailed information will come up something like

RM Failure, rc: 15041, msg: 'Execution server rejected request 
MSG=cannot send job to mom, state=PRERUN'

The solution is quite simple. The compute nodes that the scheduler threw to have its iptables on and thus the job was blocked. To solve the issue, just turn off the iptables and it works.


Sunday, November 10, 2013

PBS Error: Unable to change the status of compute nodes from down state when using Torque 4.2.5

 I did my installation according to Torque Administration 4.0.2
PBS_Server;LOG_ERROR::get_node_from_str, Node node-c00 is reporting on 
node node-c00.cluster.spms.ntu.edu.sg, 
which pbs_server doesn't know about

The solution is simple. Use the full hostname used by the client nodes and update the $TORQUE_HOME/server_priv/nodes

Restart the pbs_server, pbs_sched, trqauthd services
# service pbs_server restart
# service pbs_sched restart
# service trqauthd restart

Thursday, November 7, 2013

Resolving mmremote: Unknown GPFS execution environment when issuing mmdelnode commands

I was issuing a mmdelnode from the NSD node to a client node which are having issues, but somehow I keep getting this error.

[root@nsd1]# mmdelnode -N node-c01

Verifying GPFS is stopped on all affected nodes ...
node-c01-nas:  mmremote: Unknown GPFS execution environment
mmdelnode: Command failed.  Examine previous error messages to determine cause.

If you do a mmgetstate -a on the NSD node, you will get something like this
# mmgetstate -a 
.....
.....
05      node-c01   unknown
.....
.....

The resolution is not too difficult. According to IBM GPFS Administration and Programming References mmdelnode

You have to power down the affected node and do the mmdelnode again. You can see it is successful.
# mmdelnode -N node-c01

Verifying GPFS is stopped on all affected nodes ...
mmdelnode: Command successfully completed
mmdelnode: Propagating the cluster configuration data to all
affected nodes.  This is an asynchronous process.

Do a mmgetstate -a and you will see that the node deletion is successful

Tuesday, November 5, 2013

Using Intel® C++ Compiler with the Eclipse* IDE on Linux*

There is a very good tutorial on how to use Eclipse IDE with Intel C++ Compilers. Do look at the article by Intel

Using Intel® C++ Compiler with the Eclipse* IDE on Linux*

Saturday, November 2, 2013

GPFS Error: Unable to delete Client Node from GPFS Cluster

Today when we were delete a Client GPFS Node, the error received was something like:

 Fri Nov  1 11:55:03 SGT 2013: mmaddnode: Processing node node01
mmaddnode: Node node01 was not added to the cluster.
   The node appears to already belong to a GPFS cluster.
mmaddnode: mmaddnode quitting.  None of the specified nodes are valid.

When we do the command mmlcluster, the node was not in the list. So in other words, we are not able to add or delete the node.

Solution: Just reinstall. But you may have to remove old information
# rm -fr /usr/lpp
# rm -fr /var/mmfs

Reinstall and add the GPFS Node again. For more information, do take a look at  Adding nodes to a GPFS cluster


Friday, November 1, 2013

Compiling and Installing Meep-1.2.1 on CentOS 6 and OpenMPI

Meep (or MEEP) is a free finite-difference time-domain (FDTD) simulation software package developed at MIT to model electromagnetic systems, along with our MPB eigenmode package. The latest official version is 1.2 and can be found at Download Page for Meep

But there is a Lapack linking problem for 1.2 which is explained in Error when compiling Meep-1.2 on CentOS. It is strongly recommended to use the pre-release Meep 1.2.1 found at http://jdj.mit.edu/~stevenj/meep-1.2.1.tar.gz

Before you compile Meep 1.2.1, you need to first compile the libctl library. Compiling the libctl library is quite straightforward.

Step 1-3: For more information do see Compiling and Installing Meep-1.2.1 on CentOS 6 and OpenMPI (linuxcluster.wordpress.com)

Step 4: Compiling meep-1.2.1
# tar -zxvf meep-1.2.1.tar.gz
# cd meep-1.2.1
#  ./configure --prefix=/usr/local/meep-1.2.1 \
--with-libctl=/usr/local/libctl-3.2.1/share/libctl/ \
LDFLAGS=-I/usr/local/libctl-3.2.1/lib \
CPPFLAGS=-I/usr/local/libctl-3.2.1/include \
--with-mpi=/usr/local/openmpi-1.6.5/

References:
  1. Undefined reference to `dgetrf_’ error when compiling Meep-1.2 on CentOS
  2. crtbegin.o: No such file: No such file or directory Error when compiling Meep-1.2 on CentOS
  3. My opinion: Compiling Meep