DokuWiki

User Tools

Site Tools

Trace:
cluster:159

Warning: Undefined array key -1 in /usr/share/dokuwiki/inc/html.php on line 1458

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
cluster:159 [2017/03/29 13:56]
hmeij07 [HPC Survey 2017] 		cluster:159 [2017/03/29 15:39] (current)
hmeij07 [HPC Survey 2017]
Line 4: Line 4:
 ===== HPC Survey 2017 ===== ===== HPC Survey 2017 =====
  
-"High-Performance Computing," or HPC, is the application of "supercomputers" to computational problems that are either too large for standard computers or would take too long. HPC typically consists of a system manager server (SMS, also know as login node or master node) and compute nodes. HPC designs may differ but frequently offer high speed +"High-Performance Computing," or HPC, is the application of "compute nodes" to computational problems that are either too large for standard computers or would take too long individually. HPC typically consists of a system manager server (SMS, also know as login node or master node, or all combined) and compute nodes. HPC designs may differ but frequently offer high speed networks, large home directories, scratch space, archive space and a job scheduler. A provision application is used to (re)image the compute nodes when needed.
- networks, large home directories, scratch space, archive space and a job scheduler. A provision application is used to reimage the compute nodes when needed.+
  
-Please answer any General or Technical questions that apply to you or your group. The intent of this survey is to properly size and identify components our computational needs. The time horizon is "within the next year". +Please answer any **General** and/or **Technical** questions that apply to you and/or your group. The intent of this survey is to properly size and identify components our computational environment would look like. The time horizon is roughly "within the next three years". 
  
 **General** **General**
Line 13: Line 12:
 Q: Regarding your primary computational needs, identify your discipline and/or department. Q: Regarding your primary computational needs, identify your discipline and/or department.
  
-Q: In two sentence identify your primary research area/project. +Q: In two sentences identify your primary research area/project. 
  
-Q: If you do not currently use HPC do you anticipate using it in within next three years?+Q: If you do not currently use HPC, do you anticipate using it in within next three years?
  
-Q: IF you currently use HPC (local or remote), please give brief description (ie I use Jetstream/Stampede/...).+Q: If you currently use HPC (local or remote), please give brief description (ie I use Jetstream/Stampede/...my own cluster).
  
-Q: Does your Department anticipate expanding/initiating computational activities in the curriculum, recruiting, current research projects?+Q: Does your Department anticipate expanding/initiating computational activities in the curriculum, recruiting, or current research projects?
  
-Q: Do you currently use or anticipate the need for large storage capacity? What estimated size three years from now for your group?+Q: Do you currently use or anticipate the need for large storage capacity? 
  
 Q: Do you currently use or anticipate the need for high interprocess communications, frequently referred to as running parallel programs across multiple compute nodes? Q: Do you currently use or anticipate the need for high interprocess communications, frequently referred to as running parallel programs across multiple compute nodes?
  
 Q: Are your typical compute intensive applications CPU bound (in memory computations only), IO bound (perform lots of reads and writes to storage), or both? Q: Are your typical compute intensive applications CPU bound (in memory computations only), IO bound (perform lots of reads and writes to storage), or both?
 +
 +Q: Will be you be anticipating, or are you currently using, GPU computing?
  
 Q: Most applications will be floating point operations, however do you use integer based computing? Q: Most applications will be floating point operations, however do you use integer based computing?
  
-Q: How would you  like to participate in ongoing developments of our HPCC (ie email/list, meetings, named contact ...)+Q: How would you  like to participate in ongoing developments of our HPC (ie email/list, meetings, named contact ...)? 
 + 
 +Q: Would you support some level of periodic "monetary contributions" based on your groups' cpu usage to build a budget for future HPC maintenance.
  
 **Technical** **Technical**
  
-Q: What Linux distribution are you using or familiar with (ie CentOS, Redhat, Suse, Ubuntu ...)?+Q: What Linux distributions are you using or familiar with (ie CentOS, Redhat, Suse, Ubuntu ...)?
  
 Q: What Scheduler are you using or familiar with (ie SGE, Torque, BPS, Openlava ...)? Q: What Scheduler are you using or familiar with (ie SGE, Torque, BPS, Openlava ...)?
  
-Q: What commercial grade compiler are you using or anticipate needing (ie Intel's icc/ifort, PGI pgicc/pgfortran ...)+Q: What commercial grade compiler are you using or anticipate needing (ie Intel's icc/ifort, PGI's pgicc/pgfortran ...)
  
 Q: What commercial software do you use or anticipate using (ie SAS, Stata, Matlab, Mathematica, IDL/Envi ...)? How many concurrent licenses for your group? Q: What commercial software do you use or anticipate using (ie SAS, Stata, Matlab, Mathematica, IDL/Envi ...)? How many concurrent licenses for your group?
Line 43: Line 46:
 Q: What open source software do you use or anticipate using (ie Gromas, Amber, Lammps, R ...) Q: What open source software do you use or anticipate using (ie Gromas, Amber, Lammps, R ...)
  
-Q: For the two questions above: how important is it to you your job can restart where it left off when a crash happens (versus starting over)?+Q: For the two questions above: how important is it to you that your job can restart where it left off when a crash happens (versus starting over)?
  
-Q: Do you run parallel programs using MPI across multiple compute nodes, or anticipate doign so? Which MPI flavor (OpenMPI, MVApich, OpenMP ...)?+Q: Do you run parallel programs using MPI across multiple compute nodes, or anticipate doing so? Which MPI flavor (OpenMPI, MVApich, OpenMP ...)?
  
 Q: Do you run forked programs (threads) confined to a single compute node? Which application (ie Gaussian, Autodock ...)? Q: Do you run forked programs (threads) confined to a single compute node? Which application (ie Gaussian, Autodock ...)?
  
-Q: For the two question above: how many cpu cores/threads per job typically, typical total memory requirements per cpu core/tread.+Q: For the two question above: how many cpu cores or threads per job typically, typical total memory requirements per cpu core or thread (or total per job).
  
-Q: For serial jobs (one program requires on cpu core) typical total memory requirement per job.+Q: For serial jobs (one program requires one cpu core); typical total memory requirement per job.
  
-Q: For all types of jobs above, how many concurrent jobs running, for how long per individual job, in a typical research month. +Q: For all type of jobs listed above; how many concurrent jobs would be running, for how long per individual job, in a typical research month or quarter. 
  
 Q: Estimate the size of your home directory needs for actively used files assuming archive space is available (ie 100G, 1T ...). Q: Estimate the size of your home directory needs for actively used files assuming archive space is available (ie 100G, 1T ...).
  
-Q: Estimate your archive space needs at the end of years (ie 1T, 5T ...).+Q: Estimate your archive space needs at the end of three years (ie 1T, 5T ...). 
 + 
 +Q: Estimate the need for large scratch space requirements (ie jobs that write 100-500G of content per job). 
 + 
 +Q: For home and scratch space areas, do you require a parallel, clustered file system or are simple NFS mounts sufficient? 
 + 
 +**Optional**
  
-A: Give us any feedback we did not ask for.+A: Give us any feedback we did not ask for or forgot to ask m(
  
  
cluster/159.1490810200.txt.gz · Last modified: 2017/03/29 13:56 by hmeij07

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki