Difference between revisions of "Large-Memory SMP Servers"

From UFRC
Jump to navigation Jump to search
 
(12 intermediate revisions by 3 users not shown)
Line 1: Line 1:
[[Category:Infrastructure]][[Category:Hardware]]
+
[[Category:Scheduler]]
 
Research Computing currently maintains the following resources for calculations requiring large amounts of physical memory.
 
Research Computing currently maintains the following resources for calculations requiring large amounts of physical memory.
  
Line 5: Line 5:
 
|+ Large Memory Servers
 
|+ Large Memory Servers
 
|-
 
|-
! Host
+
! Nodes
 
! Architecture  
 
! Architecture  
 
! Vendor
 
! Vendor
Line 11: Line 11:
 
! Frequency
 
! Frequency
 
! Cores
 
! Cores
! Memory(GB)
+
! SLURM Memory(GB)
 
|-
 
|-
! scope="row" | i1a-s3
+
| 1 || amd64 || AMD || EPYC 7702 || 2.0 || 128 || 4023
| amd64 || AMD || 6174 || 2.2 || 48 || 512
 
 
|-
 
|-
| amd64 || Intel || E5-4607 || 2.2 || 24 || 750
+
| 2 || amd64 || Intel || E7-8850 || 3.2 || 192 || 1509
|-
 
! scope="row" | s5a-s23
 
| amd64 || Intel || E7-8850 || 2.0 || 80 || 1024 (1TB)
 
 
|}
 
|}
  
The big memory machines are available only via the HiPerGator "bigmem" queue. To submit jobs to this queue you will need to
+
For more details see automatically generated data in [[Available Node Features]]
<pre>
 
a) Log into HiPerGator (ssh gator.rc.ufl.edu)
 
  
b) Add the following PBS directive to your job submission script.
+
The big memory machines are available only via the HiPerGator "bigmem" partition. To submit jobs to this partition you will need to add the following directive to your job submission script.
 +
  #SBATCH --partition=bigmem
  
  #PBS -q bigmem
+
Since our regular nodes have 1TB of available memory we do not recommend using bigmem nodes for jobs with memory requests lower than that.
</pre>
 
Please also be aware that servers with hundreds of gigabytes of RAM are still fairly expensive. For this reason and so that resources can be allocated fairly, resource requests are considered on the basis of [[Processor Equivalents]]. If you submit a job to the ''bigmem'' queue and your group does not have an NCU allocation sufficient to accommodate the resource request, your job will be blocked and will not run.
 

Latest revision as of 15:33, 14 July 2022

Research Computing currently maintains the following resources for calculations requiring large amounts of physical memory.

Large Memory Servers
Nodes Architecture Vendor Processor Frequency Cores SLURM Memory(GB)
1 amd64 AMD EPYC 7702 2.0 128 4023
2 amd64 Intel E7-8850 3.2 192 1509

For more details see automatically generated data in Available Node Features

The big memory machines are available only via the HiPerGator "bigmem" partition. To submit jobs to this partition you will need to add the following directive to your job submission script.

 #SBATCH --partition=bigmem

Since our regular nodes have 1TB of available memory we do not recommend using bigmem nodes for jobs with memory requests lower than that.