Questions tagged [hpc]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
0 votes
0 answers
9 views

Dynamically checking and allocating SLURM nodes within a python script

I have a computationally expensive simulation function I am looking to distribute accross a multi-node cluster. The code looks something like this: input_tasks = [input_0, input_1, ..., input_n] for i ...
GTOgod's user avatar
  • 109
0 votes
1 answer
57 views

On which computer are mounted files executed?

This could be a few questions, but I feel it's a similar issue. On which computer are mounted executable files executed? And does this change with like sshfs, nfs, or ln? Like if you have a storage ...
user avatar
0 votes
0 answers
17 views

How can I port bind an interactive transient PBS session to the head node and then port bind the head node to my local machine?

The short of it: How can I port bind an interactive transient PBS session to the head node and then port bind the head node to my local machine? So, a chain of port bind of three machines. More ...
Sudoh's user avatar
  • 101
0 votes
0 answers
82 views

Why does 'module load intel' fail to load libfftw3?

I am in a cluster (centos) trying to run a program (pw.x) that requires loading three modules: intel, impi, and quantum-espresso. Now, I am getting an error saying pw.x: error while loading shared ...
Abdul Muhaymin's user avatar
1 vote
0 answers
63 views

Network management switch and data transfer switch in an HPC cluster

Shown in the image is a validated design from Dell for an HPC with AI workload. We are using this reference in our HPC design. The image is taken from this white paper by Dell on page 12: HPC ...
klaus's user avatar
  • 11
0 votes
0 answers
60 views

Why is SLURM reporting "NO NETWORK ACCESS" for two new hosts when they are, in fact, up and reachable?

I'm new to the operations side of HPC, after being asked to take over administration of a small slurm-based cluster. We recently added two new hosts to our cluster, but slurm has repeatedly drained ...
Mike Heinz's user avatar
0 votes
0 answers
597 views

Windows can't find SSHFS

When mapping a network drive on Windows 10, after installing SSHFS-Win and WinFsp, I get the error message "Windows can't find ssfhs". Any idea why this is happening? Here is the steps that ...
sam wolfe's user avatar
  • 101
0 votes
1 answer
79 views

Why a non-root installed software can work across the whole cluster?

I recently installed a new python3 and another python package locally in my account folder on a cluster with a dozen of nodes (each node with several cores). I originally thought that I can python3 ...
xiaohuamao's user avatar
5 votes
1 answer
86 views

Do writes and reads contend with each other for an SSD? [closed]

I have a disk I/O intensive application where I am reading and writing a lot concurrently. For a spinning disk, it makes sense that there is contention because the pointer has to move around a lot but ...
Niteya Shah's user avatar
0 votes
1 answer
513 views

passwd: Authentication token manipulation error using ssh and public key

I'm using ssh and public key (saved in my local computer) to connect to a HPC cluster as follows: $ ssh -i ~/narvi_key/xenial-narvi-key [email protected] Enter passphrase for key '/home/xenial/...
Farid Alijani's user avatar
1 vote
0 answers
1k views

How to make a host file in SLURM with $SLURM_JOB_NODELIST

I have access to a HPC with 40 cores on each node. I have a batch file to run a total of 35 codes which are in separate folders. Each code is an open mp code which requires 4 cores each. so how do I ...
Libin Varghese's user avatar
0 votes
1 answer
834 views

Getting error while using Rsync on Cygwin

I am using rsync, to transfer data to HPC cluster, using Cygwin for windows 7. I type the following command to transfer a folder from my computer to the HPC remote server. rsync -rzv /cygdrive/C/...
lsr729's user avatar
  • 123
1 vote
1 answer
1k views

slurmd: Invalid job credential

I'm having some problems with a test configuration of Slurm on my laptop. I'm trying to run four slurmd instances on one machine, which is also the same machine as slurmctld runs on. I have a local ...
lukas's user avatar
  • 11
1 vote
0 answers
51 views

HPC Visualization node GPU choice

On most HPC Visualization node, we find nvidia tesla cards. P100 come to mind. I'm not sure to follow why. Nvidia Tesla cards are designed for computing according to nividia documentation, not for ...
Mathieu Westphal's user avatar
0 votes
1 answer
1k views

mpirun tcp_peer_send_blocking: send() to socket X failed: Broken pipe (32)

I am using a HPC composed by 4 machines (master, slave1, slave2, slave3 and slave4) I try to run a script on a HPC structure : mpirun -report-uri - -host master,slave1,slave2,slave3,slave4 --map-by ...
Charlotte Rieux's user avatar
0 votes
0 answers
141 views

How to install WINE in a linux server using clone method from GitHub?

I tried to install by downloading from linux terminal and running the commands $ cd wine-3.0-rc1 $ ./configure --enable-win64 $ make # make install and gives always error and fail to proceed. Since ...
Gokul Anand's user avatar
0 votes
0 answers
18 views

Prefer more resources on Windows HPC?

If I specify a min/max core or socket counts, is there a way to tell HPC to prefer to use more resources? i.e. if I have the below job/task, can I tell HPC to prefer more sockets? <Job Version="3....
Carbon's user avatar
  • 155
0 votes
1 answer
193 views

Where should I save the .exe file in Windows HPC Cluster head?

I am working on HPC Pack cluster. I want to use the HPC Pack Job Manager to run my tasks. Where should I save the .exe file? Task
Jack's user avatar
  • 1
0 votes
1 answer
52 views

What does the acronym CMC refer to in this context?

I am doing a research project on SGI High Performance Computers, specifically the SGI ICE XA platform. In the datasheet (found here), I came across this image. I am wondering what the acronym CMC ...
sawyermclane's user avatar
1 vote
1 answer
874 views

SLURM configuration: cons_res with CR_Core either cannot allocate resource or jobs end up in CG status

I am new to SLURM. I am trying to configure slurm in a new cluster. I have 4 nodes each has 14 cores. I wanted to share nodes in a way that every core can run independently (i.e., node01 can have 14 ...
Somesh's user avatar
  • 11
0 votes
1 answer
2k views

Is it possible to use a Xeon Phi in a Mac running OS X Yosemite?

A colleague of ours is willing to lend our lab a Xeon Phi (2012 model, the one with passive cooling) that he has no use for. The only machine we could physically put this beast into is a Mac Pro (2011 ...
András Aszódi's user avatar
2 votes
2 answers
439 views

Why is Xeon processor suited for HPC?

Among the different specs for Xeon, available here, http://ark.intel.com/compare/87720,84679, which aspects makes it relevant for HPC ? The processor base frequency is less than for Core i7. Why ...
kiriloff's user avatar
  • 223
1 vote
1 answer
127 views

Mirror with backup of changed files

we have some largeish storage requirements (genomic data) for which we need to buy some archival space (write once, read rarely, files remain easily accessible, each user should have access to their ...
Yannick Wurm's user avatar
3 votes
2 answers
477 views

Advantages of nVidia Tesla vs Xeon Phi

This maybe only theoretical, but are there any advantages of using one or another in terms of: ISA energy efficency Solution with using QPI looks better for me. Are there any advantages of using ...
user3313119's user avatar
1 vote
1 answer
1k views

Setting include paths in LSF-bsub

I am running a big job on a HPC. But the installed programs for the job are old versions and I have the new versions in my home directory. I can log in to any node, export the path to the new versions ...
WYSIWYG's user avatar
  • 201
0 votes
1 answer
3k views

How much memory is being used on the gpu by each PID?

How do I know how much memory on the GPU is being used by each process, CUDA compiled code using 1 processor, accessing 1 of 2 M2090 Tesla cards available on a 16 core node on an HPC? ~/code_dev/...
dmerz75's user avatar
0 votes
2 answers
1k views

Mathematica benchmark on two different system

I have two system one has a Xeon E5-1650 CPU, 24GB of RAM, and a 7200rpm hard drive. The other system which is a Lenovo G510 laptop, has a core i7-4700MQ CPU, 4GB of RAM, and 5400rmp hard drive. The ...
MOON's user avatar
  • 410
9 votes
5 answers
24k views

Why does a Xeon processor cost much more than a similar Core i7 part?

I want to know why the price of a Xeon CPU and a similar Core i7 CPU is so different. The CPU clock is more or less the same. I want to know this because I want to have an understanding of what kind ...
MOON's user avatar
  • 410
0 votes
1 answer
86 views

Solution for HPC with Matlab and Linux

I have the following problem. I want to install Matlab as root on a remote computer with graphical user interface installed. I've a campus license that can be installed on 4 different computers (or ...
linello's user avatar
  • 111
2 votes
0 answers
58 views

Windows Azure HPC Scheduler Architecture

So far I've found very little information on the scheduling policy, resource management policy of Azure HPC Scheduler. I would appreciate any kind of information regarding some of these questions: ...
Churianov Roman's user avatar
2 votes
0 answers
247 views

Sharing storage on Linux and Solaris

I'm looking for a solution in order to share a san mounted volume between several hosts running on Linux (RHEL) and/or Solaris (Sparc). Note that I basically need to share a set of directories ...
devlearn's user avatar
  • 121
0 votes
1 answer
459 views

Kerrighed cluster does not use all nodes when running OpenMP

I've been given access to a recently built cluster that uses Kerrighed. I'm the first user of this cluster, so setup and configuration problems may be a possibility. I'm also totally new to ...
Sevenless's user avatar
  • 101
1 vote
4 answers
3k views

Creating a home server cluster for data analysis

I've seen a couple of questions on here that are similar to what I'm thinking, but not quite. So, I figured I'd ask and see what falls out. Basically, I have 2 or 3 old, slower-ish laptops laying ...
user1074057's user avatar
4 votes
3 answers
742 views

Building an improvised 'supercomputer' by connecting a number of used PCs in a LAN

I am in ongoing development of a simple but processor intensive computer program that I use for scientific research. It would be really helpful to have more processing speed. Right now I'm running Mac ...
Matt Munson's user avatar
0 votes
1 answer
190 views

preparing large data sets for analysis [closed]

I have a fortran program that simulates a lot of ascii text file output. These are quite large to load the entirety of any single file into common data analysis programs (Matlab, R, SciPy). I think ...
hatmatrix's user avatar
  • 854
4 votes
0 answers
911 views

Installing new rolls on existing Rocks Cluster

I am on a supercomputer research team at a small university. We can't seem to find where the CUDA roll is installed on our Rocks cluster, and we're pretty sure it was never properly installed. We do ...
zr00's user avatar
  • 321
2 votes
2 answers
2k views

Which edition microsoft 2008 server should I use?

I have a MSDN (Accademic alliance) subscription, and I have a rack mountable box, (I can't remember whats in it anymore but it has 2 CPUs and 1 GB of DDR 2 ram. I remember when I was given it (12 ...
Frames Catherine White's user avatar
0 votes
2 answers
116 views

Homebuild cluster and apps efficiecy

I have 3 pc's and was wondering if i can connect them into a cluster. I read few articles about clustering, but there is one think which I'm not sure about. If I build my cluster can I run apps for "...
f1ames's user avatar
  • 101
9 votes
2 answers
8k views

Makefile fails to install file correctly, installing HPL

I started installing HPL a while ago and had a related question. I've been following along with this guide from Intel. I figure this warrants a whole new one. When I try to make the archive, the ...
zr00's user avatar
  • 321
8 votes
4 answers
10k views

How do different Linux distributions handle multiple cores and multi-threading?

I realize that by asking this question I have already labeled myself as "unprepared" for high-performance workstations, but whatever, you have to start somewhere! I would like to know how different ...
Jonathan's user avatar
  • 1,290
2 votes
1 answer
80 views

What technologies came to the PC world from High Performance Computing?

Recently I came to think about the technologies that supercomputers gave to the modern PCs. I found out that the instruction pipelining appeared in one of the first CDC machines and parallelism was ...
Massimiliano's user avatar
10 votes
2 answers
12k views

Wait for one or all LSF jobs to complete

With Platform LSF, is there a command that will wait for either a specific job or all my submitted jobs to finish before returning? In other words, I am looking for the LSF equivalent of the wait ...
Vebjorn Ljosa's user avatar