High Performance Computing encompasses using "supercomputers" with high numbers of CPUs, large parallel storage systems and advanced networks to perform time-consuming calculations. Parallel algorithms and parallelization of storage are essential to this field, as well as issues with complex, fast ...

learn more… | top users | synonyms

-2
votes
2answers
10 views

memory usage per user in SGE cluster

I would like to automate the estimation of monthly memory usage of all jobs performed by a given user in my cluster (SGE, ubuntu). I have seen there are many tools to compute the current memory usage ...
0
votes
0answers
8 views

Different compute schedulers running on VMs

I'm setting up a parallel computing cluster of about 15 nodes that will have to be shared by multiple research groups. Each group has their own software, licensing, and job scheduler requirements (...
0
votes
0answers
21 views

How can I launch graph500 benchmark on nVidia GPU with CUDA?

How can I launch graph500 benchmark on nVidia GPU with CUDA? There is written: https://devblogs.nvidia.com/parallelforall/gpus-graph-predictive-analytics/ At the IEEE Bigdata conference later in ...
0
votes
1answer
31 views

Slurm: Have two separate queues for GPU and CPU-only jobs

At the moment, we have set up Slurm to manage a small cluster of six nodes with four GPUs each. That has been working great so far, but now we want to utilize the Intel Core i7-5820K CPUs for jobs ...
0
votes
2answers
1k views

Numerous pbs_server errors in /var/log/messages

On supercomputer's management node we receive numerous errors such as: pbs_server: LOG_ERROR::is_request, bad attempt to connect from 10.10.0.254:1023 (address not trusted - check entry in ...
-5
votes
1answer
27 views

why all hard drives in a HPC cluster must be of the same size and part number? [closed]

One of the hard drives of my HPCC is broken and I have to buy another one. I've heard it must be of the exactly same size and part number. Could anyone explain me why? Thanks
0
votes
0answers
73 views

How to submit ABAQUS job on Rocks cluster?

I created ROCKS6 cluster and i installed ABAQUS on my cluster and it works well on frontend, but i can not submit job on my 2 other compute nodes. I just want to run ABAQUS on this cluster (frontend ...
1
vote
0answers
30 views

Torque queue issue

I am having troubles with Torque + Maui. The problem is the following: I have 2 queues, each queue has 10 associated nodes. If i submit 10k jobs to the first queue and i submit 1 job to the second ...
0
votes
0answers
77 views

The disadvantages of Windows hpc cluster?

What are the disadvantages for a cluster running windows OS for hpc applications? Why such hpc are so uncommon? The cost of windows server OS is rather small comparing to the hardware cost, and the ...
0
votes
0answers
50 views

Performance impact of blocking Infiniband

How does a small blocking in an Infiniband network affect the latency? More specifically, if I compare a 4xQDR non-blocking setup with an FDR setup with a blocking factor of 1:1.6, how would they ...
0
votes
0answers
16 views

HPC 2012 R2 - Define “priority” for servers

I've been digging through the documentation and I can't figure out my issue, so I'm coming here for help. I've got an HPC 2012 R2 cluster I've built for a customer. There is one head-node that has ...
0
votes
0answers
116 views

Deployment of Base-Node via iSCSI in Server 2012R2 HPC cluster fails (can not join domain)

We are currently evaluating Server 2012R2 with HPC Pack for an upcoming project. Sadly we are stuck at deploying the base node. The node boots via PXE (iPXE) and connects to iSCSI, installs Windows ...
1
vote
1answer
197 views

HPC / EC2 - optimizing NFS for relialibility

In AWS-EC2, I've set-up a cluster of linux virtual machines made of an NFS fileserver and many clients. If the number of clients is above ~20, under heavy I/O, I am experiencing loss of file integrity:...
1
vote
1answer
107 views

deploying base note for HPC cluster in server 2012R2 hangs

We are in the process of evaluating server 2012R2 standard with HPC pack for a small cluster of nodes (about 40 to start with - the current setup has only one compute node). For the moment we use old ...
2
votes
2answers
109 views

Why configure cluster nodes to reboot when out of memory?

I have access to a research HPC cluster which is configured so that if your job tries to use more memory than the node has available the node crashes and automatically reboots. This appears to be ...
10
votes
1answer
797 views

Why does requesting GPUs as a generic resource on a cluster running SLURM with the built-in plugin fail?

Disclaimer: This post is quite long as I tried to provide all relevant configuration information. Status and Problem: I adminster a gpu cluster and I want to use slurm for job management. ...
3
votes
2answers
784 views

Does Xeon Phi works with i7 CPUs

Does Xeon Phi coprossesors works with i7 CPUs ? It's advertised for use with Xeons, but for my app (WRF), i7-3930k performs better and is 3 times cheaper than high grade xeons. So I wonder if I could ...
1
vote
1answer
268 views

What cluster management software does there exist for Windows, except for MS HPC Server? [closed]

Do there exist any open-source solutions? Perhaps, cheaper commercial ones? I'm surprised that googling does not yield anything meaningful for me.
1
vote
1answer
41 views

No service for subscription

I am trying to set up bursting to Azure with a Windows HPC cluster. THe cluster already works fine and I can start jobs on the machines on that are on the local network. When I try and create a node ...
2
votes
1answer
89 views

Parallel Processing and Disk IO for performance. More cores or more servers?

I have a large analysis job on an AWS EC2 instance (c3.8xlarge) on Ubuntu 12.04. The objective is to load the server at 100% CPU, running as many jobs as memory allows (varying amounts but generally ...
2
votes
1answer
680 views

Building Windows server cluster hpc

I'm trying to set up a Windows Server cluster with a head node and 2x compute nodes. So far, i've managed to install : (thanks to this tutorial http://msdn.microsoft.com/en-us/library/jj884142.aspx) ...
3
votes
2answers
3k views

Alternative to ScaleMP?

Anyone know of an alternative to ScaleMP? They let several x86 boxes boot as one large box. Theoretically AMD's hypertransport should allow the same thing. Any other companies or OSS projects doing ...
1
vote
1answer
39 views

Parallel filesystem which schedules simultaneous file requests to mutually exclusive sets of OSSs

My environment is RHEL based, interconnect is infiniband. I have some experience with Lustre. What i want to know is: Is there a parallel file-system, where if simultaneous write request arrive, ...
0
votes
1answer
1k views

HPC Cluster (SLURM): recommended ways to set up a secure and stable system

I'm working with a SLURM driven HPC Cluster, containing of 1 control node and 34 computation nodes and since the current system is not exactly very stable I'm looking for guidelines or best practices ...
1
vote
1answer
223 views

Windows VMs on ScaleMP cluster?

I was wondering whether the following stack would work, and if so then how well, and what sort of problems I might expect to encounter when setting it up? Hardware layer - lots of cheap servers SMP ...
1
vote
1answer
357 views

Open MPI can't launch remote nodes via SSH

I am trying to set up Open MPI between a few machines on out network. Open MPI works fine locally, but I just can't get it to work on a remote node. I can ssh into the remote machine (without password)...
0
votes
1answer
74 views

How do we configure Lustre to block client requests when under load, rather than failing?

We are using Lustre in a cluster with approximately 200TB of storage, 12 Object Storage Targets (that connect to a DDN storage system using QDR Infiniband), and roughly 160 quad and 8-core compute ...
0
votes
1answer
49 views

Running jobs in a HPC cluster [closed]

I'm quite new to HPC environment. Is there any difference in running a job on a node utilizing 8 cores and running the same job on 8 nodes utilizing I core in terms of performance or walltime used. ...
6
votes
4answers
2k views

using i7 “gamer” cpu in a HPC cluster

I'm running WRF weather model. That's a ram intensive, highly parallel application. I need to build a HPC cluster for that. I use 10GB infiniband interconnect. WRF doesn't depends of core count, but ...
4
votes
4answers
9k views

Set up simple Infiniband Block Storage (SRP or iSER)

I'm trying to figure out how to set up a simple storage system which exports block storage over Infiniband, using either SRP or iSER. I'm very early in the process, and at the moment I'm basically ...
6
votes
1answer
2k views

XFS Adding Quotas - Skip Quota check on first mount/boot

We run a 14TB XFS fileserver on our cluster and want to add quota support. This is running 3.9.2-1.el6.elrepo.x86_64 kernel under CentOS 6.3 (Final). The issue is when we unmount the XFS RAID, and re-...
4
votes
2answers
2k views

View Infiniband routing table generated by OpenSM?

As I understand it, the subnet manager of an Infiniband network calculates the best routes between each pair of nodes on the network and provides these routes to the nodes when they want to ...
0
votes
0answers
97 views

Microsoft HPC: mixing windows and linux blades

I have a working Windows HPC cluster with 32 blades, all of them are using Windows HPC. My question is: can I install Linux on 16 blades and keep the other 16 on Windows? Is there a specific version ...
0
votes
2answers
346 views

Running ScaleMP on top of OpenStack [closed]

Looking for a feedback if anyone has already played with running ScaleMP linux appliances in OpenStack (KVM)? A short description of the setup (w/ or w/o InfiniBand, total amount of RAM, etc) and its ...
1
vote
3answers
193 views

Appropriate network file system for huge (5+ Gb) files

I've got a number of servers used for HPC / cluster computing and I noticed that, given the fact that part of the computations they run use huge files over NFS, this causes significant bottlenecks. I'...
0
votes
2answers
114 views

low performance on HPC cluster (sge) when running multiple jobs

O know this is a long-shot but I'm clueless here. I'm running several computer simulations on High Performance Computation cluster (HPC) of oracale grid engine (sge). A single job runs at a certain ...
2
votes
1answer
122 views

Microsoft HPC cluster - AD or AD LDS?

The instructions for deploying an HPC Cluster (e.g. step 1.5 on this page in TechNet) are very clear that HPC cluster nodes "must be members of an Active Directory domain". Does the Active Directory ...
2
votes
5answers
626 views

setting up a cluster

I have 5 PC(s) connected over a LAN through a switch. I want to connect them to form a HPC cluster. The OS may be any Linux version (currently I have installed Ubuntu 8.10, 9.10 and Fedora 10) Purpose ...
3
votes
1answer
448 views

What makes Lustre faster and more scalable than NFS?

I have read in various places (e.g. here and here) that NFS' I/O performance does not scale, while Lustre's does, and that Lustre can deliver better I/O rates in general. There seem to be various ...
9
votes
6answers
4k views

How to allow users to transfer files to other users on linux

We have an environment of a few thousand users running applications on about 40 clusters ranging in size from 20 compute nodes to 98,000 compute nodes. Users on these systems generate massive files (...
1
vote
1answer
1k views

Torque and maui node status

I am new for torque and maui. I was checking for node state to looking for which nodes are free and which nodes are in use. For torque one command is pbsnodes. Which gives status and other info ...
3
votes
1answer
382 views

Which is the fastest way to move 1Petabyte from one storage to a new one?

First of all, thanks for reading, and sorry for asking something related to my job. I understand that this is something that I should solve by myself but as you will see its something a bit difficult. ...
1
vote
1answer
62 views

HPC OSS Node issue with unreadable local hdd error

We have a HPC setup with four OSS server(OSS1 to OSS4) and two MDS Nodes(MDS1 to MDS2) It has been running till yesterday without any problem. Today morning i found that OSS4 is in shutdown condition....
0
votes
1answer
62 views

Windows HPC Server 2008 R2 SP3 upgrade

It's not clear to me from reading the documentation if I must update the clients, or whether I can just update the head node and compute nodes. Does anyone have any experience of this? I don't want ...
1
vote
2answers
536 views

Task spooler for computing server on Debian

Recently our university has bought an computing server with one multi-core Xeon and 4 powerfull GeForce videocard for lessons on discipline "High perfomance computing with CUDA". There is Debian ...
1
vote
4answers
2k views

A dozen mac mini's vs dell rack server for parallel image processing [closed]

I need to do some large scale image processing in parallel. I was thinking of running a dozen Mac Mini's in parallel to do the data processing. I need to run Microsoft Windows on the machines so I ...
1
vote
2answers
54 views

HPC 2008 Time Share Single Core (Fractional Resource Scheduling)

Windows HPC 2008 appears to be restricted to one task per core. Is there anyway to time share multiple tasks (or jobs) over a single core?
-3
votes
1answer
207 views

Why do supercomputers take up a whole room? [closed]

Some pictures regarding the Texas Advanced Computing Center (one I am currently interested in) here: http://www.tacc.utexas.edu/resources/hpc/ If you see the two supercomputers they have, Lonestar ...
1
vote
0answers
63 views

Programs running in an NX session seem to pause when session disconnects

We are currently running an interactive HPC application which presents a graphical interface to the user, attaches to an HPC cluster and allows them to run and observe some computation. The user logs ...
0
votes
1answer
529 views

Non-exclusive job scheduling in PBS/Torque

The cluster resource manager Torque typically allocates compute nodes on an exclusive basis. However, when you have a lot of small jobs (like we do) running against multi-core compute nodes, this can ...