Month: September 2020

by

Encountering Error when pip install TensorToolbox

I’m using Python-3.8.7  When I do a pip install for TensorFlowbox with Intel Optimized Toolbox, I received errors.

% pip install TensorFlowbox

But it failed with its SpectralToolbox Dependencies.

.....
.....

Building wheels for collected packages: SpectralToolbox, orthpol-light
Building wheel for SpectralToolbox (setup.py) ... error
ERROR: Command errored out with exit status 1:
command: /usr/local/python/intel/2017u3/intelpython3/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-p5qjlor0/spectraltoolbox/setup.py'"'"'; __file__='"'"'/tmp/pip-install-p5qjlor0/spectraltoolbox/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-4k6ars4d
cwd: /tmp/pip-install-p5qjlor0/spectraltoolbox/
Complete output (56 lines):

Somehow the later version of Python3 has issues with SpectralToolbox and TensorToolbox. To compile TensorToolbox, you have to go back to earlier version of Python 3. I chose Python-3.6.9 (https://www.python.org/downloads/release/python-369/).

And it works.

Perquisites:

openmpi-3.1.4
gnu-6.5
m4-1.4.18
gmp-6.1.0
mpfr-3.1.4
mpc-1.0.3
isl-0.18
gsl-2.1

 

Compile

% tar -zxvf Python-3.6.9
% cd Python-3.6.9
% ./configure --prefix=/usr/local/python-3.6.9 --enable-optimizations
% make -j 16
% make install

Installing TensorToolbox

% pip install numpy scipy matplotlib
% pip install mpi4py
% pip install TensorToolbox

For more information, see TensorToolbox-1.0.22 (https://pypi.org/project/TensorToolbox/#description)

 

Advertisement
by

Fixing can’t load screen14 issues for ANSYS 2020-R1

I was having this strange issue when running ANSYS ICEM

"Fixing can't load screen14 issues, using variable
Signal 11 caught!
Segmentation violation - exiting after doing an emergency save"

The Issue can easily resolved. It is due to errors due to missing fonts libraries

# yum install xorg-x11-fonts-*
Loaded plugins: fastestmirror, langpacks
Loading mirror speeds from cached hostfile
* base: mirror.newmediaexpress.com
* epel: download.nus.edu.sg
* extras: mirror.newmediaexpress.com
* updates: mirror.newmediaexpress.com
Package xorg-x11-fonts-ISO8859-1-75dpi-7.5-9.el7.noarch already installed and latest version
Package xorg-x11-fonts-Type1-7.5-9.el7.noarch already installed and latest version
Resolving Dependencies
--> Running transaction check
---> Package xorg-x11-fonts-100dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-75dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-1-100dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-14-100dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-14-75dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-15-100dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-15-75dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-2-100dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-2-75dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-9-100dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ISO8859-9-75dpi.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-cyrillic.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-ethiopic.noarch 0:7.5-9.el7 will be installed
---> Package xorg-x11-fonts-misc.noarch 0:7.5-9.el7 will be installed
--> Finished Dependency Resolution

The Issue should be resolved.

by

MemVerge GA Launch – Breakthrough in Big Memory

Presentations from Intel, Penguin Computing, Banco Intesa Sanpaolo and industry Visual Effects expert

Speaker Timestamps:
(3:32 min) – Intel, VP/GM – Alper Ilkbahar
(9:40 min) – MemVerge, CEO – Dr. Charles Fan
(30:37 min) – Penguin Computing, VP – Dr. Kevin Tubbs
(45:30 min) – Visual Effects Expect – Dr. Hank Driskill
(54:40 min) – Banco Intesa Sanpaolo, Head of Cloud – Nicola Carotti

by

Using xsos to summarise System Information

Introduction

The goal of xsos is to make it easy to instantaneously gather information about a system together in an easy-to-read-summary, whether that system is the localhost on which xsos is being run or a system for which you have an unpacked sosreport.

xsos will attempt to make it easy, parsing and calculating and formatting data from dozens of files (and commands) to give you a detailed overview about a system.

 

Installation

Manual Install

% git clone https://github.com/ryran/xsos.git
Cloning into 'xsos'...
remote: Enumerating objects: 6, done.
remote: Counting objects: 100% (6/6), done.
remote: Compressing objects: 100% (5/5), done.
remote: Total 946 (delta 1), reused 5 (delta 1), pack-reused 940
Receiving objects: 100% (946/946), 907.12 KiB | 728.00 KiB/s, done.
Resolving deltas: 100% (450/450), done.

Point 1: Get information on OS and Memory

Point 2: Get information on Network, Network Adapters and CPU

Point 3: Get Information on BIOS

Point 4: Get Information on sysctl

Point 5: Get information on kdump

References:

    1. xsos — a tool for sysadmins and support techs
    2. xsos Project Space
by

Altair acquires Univa and Ellexus

Altair, (Nasdaq: ALTR) a global technology company providing solutions in data analytics, product development, and high-performance computing (HPC), today announced the acquisition of Univa, a leading innovator in enterprise-grade workload management, scheduling, and optimization solutions for HPC and artificial intelligence (AI) on-premises and in the cloud.

For more information, see Altair Acquires Univa

by

How to increase the number of threads created by the NFS daemon for CENTOS 7

Taken from How to increase the number of threads created by the NFS daemon in RHEL 4, 5, 6 and 7?

In case of a NFS server with a high load, it may be advisable to increase the number of the threads created during the nfsd server start up.

Edit the following line in /etc/nfs.conf

% vim /etc/nfs.conf
#[nfsd]
# debug=0
threads=64
# host=
# port=0
# grace-time=90
# lease-time=90
# udp=y
# tcp=y

Testing whether it works….

% cat /proc/net/rpc/nfsd

According to the RH, “The first number is the total number of NFS server threads started. The second number indicates whether at any time all of the threads were running at once. The remaining numbers are a thread count time histogram.”

th 64 0 2.610 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
by

Find CPU and GPU Performance Headroom using Roofline Analysis

Join Technical Consulting Engineer and HPC programming expert Cedric Andreolli for a session covering:

  • How to perform GPU headroom and GPU caches locality analysis using Advisor Roofline extensions for oneAPI and OpenMP
  • An introduction to a new memory-level Roofline feature that helps pinpoint which specific memory level (L1, L2, L3, or DRAM) is causing the bottleneck
  • A walkthrough of Intel Advisor’s improved user interface

To see video, see https://techdecoded.intel.io/essentials/find-cpu-gpu-performance-headroom-using-roofline-analysis/#gs.fpbz93

by

NVIDIA to Acquire Arm for $40 Billion, Creating World’s Premier Computing Company for the Age of AI

NVIDIA and SoftBank Group Corp. (SBG) today announced a definitive agreement under which NVIDIA will acquire Arm Limited from SBG and the SoftBank Vision Fund (together, “SoftBank”) in a transaction valued at $40 billion. The transaction is expected to be immediately accretive to NVIDIA’s non-GAAP gross margin and non-GAAP earnings per share.

The combination brings together NVIDIA’s leading AI computing platform with Arm’s vast ecosystem to create the premier computing company for the age of artificial intelligence, accelerating innovation while expanding into large, high-growth markets. SoftBank will remain committed to Arm’s long-term success through its ownership stake in NVIDIA, expected to be under 10 percent.

For more information, see NVIDIA to Acquire Arm for $40 Billion, Creating World’s Premier Computing Company for the Age of AI

by

Checking nproc limits

One of our Linux Compute Server was showing when a particular was attempting to login on.

failed to execute /bin/bash: resource temporarily unavailable

We suspected that nprocs limits have been breached by the particular user. I found this write-up https://blog.dbi-services.com/linux-how-to-monitor-the-nproc-limit-1/ very prescriptive of the issue I faced.

Extracting information via ps is not useful unless you use the “-L” to show threads, possibly LWP (light-weight process).

 

% ps h -LA -o user | sort | uniq -c | sort -n
1 chrony
1 dbus
1 libstoragemgmt
1 nobody
1 rpc
1 rpcuser
2 avahi
2 user3
2 postfix
3 colord
3 rtkit
4 user1
4 user2
7 polkitd
23 user4
31 user5
34 user6
361 user7
442 user8
556 gdm
563 user9
922 user10
16384 user11
3319 root

You can see that user11 has 16384 threads!

To dig down into what is happening to a selected user. We will use user2 since it has one of the fewest LWP to

% ps -o nlwp,pid,lwp,args -u user2 | sort -n
NLWP PID LWP COMMAND
1 272705 272705 sshd: user2@pts/12
1 273054 273054 sshd: user2@notty
1 273216 273216 /usr/libexec/openssh/sftp-server
1 273406 273406 -bash

nlwp – Number of LWP
lwp – Process of ID of the LWP.

To eliminate the offending user11’s thousands of threads

% pkill -KILL -u user11

References

  1. Linux: how to monitor the nproc limit
  2. How is the nproc hard limit calculated and how do we change the value on CentOS 7