Resource Limits

Here is a summary of the system limits for users of our computing resources. The memory is the maximum you can allocate per node in practice. Nodes is the total number of nodes, and cores is the total maximum per user, not the cores per node.

These are the default values; individual users and units may have limits that are higher or lower than these (hardware permitting).

Deigo

Deigo has a number of partitions, each with its own restrictions on maximum cores and time.

Compute
Partition
Memory
per node
Nodes
total
Cores
per user
Memory
per user
Runtime
per job
Notes
Compute 500G 356 2000 7500G 4 days (1)
Short 500G 648 4000 6500G 2 hours (1)
Largemem 500G 84 5 nodes unrestricted (2)
Largemem 750G 14 5 nodes unrestricted (2)
Bigmem 2978G 1 8 unrestricted (2)
Largejob 500G 100 50 nodes 2 days Special (3)
datacp 185G 4 4 19G for moving data (4)

 

Storage
System
Amount    Notes
/home 50G   user limit
/flash 10T   Per-unit limit
/bucket 50T   Per-unit limit, expandable
naruto   Tape for archiving; expends as neeed

 

  1. If you need more compute time than the current limit, you can ask us to increase that limit in exchange for fewer cores, as detailed here.
     
  2. Largemem and Bigmem have no fixed maximum time. However, it's a bad idea to run for more than about a month. The risk increases greatly that a hardware or software error, an electical outage, or emergency maintenance will kill your job prematurely.
     
  3. Largejob is for many-core computations, and requires an application from the users' unit leader. You can apply using this form.
     
  4. Datacp partition has four nodes, and is only for transferring large data volumes between Bucket and Flash. Do not use these for any kind of computation.

Saion

Saion is a general accellerated computing system with three subsystems. You can read more about this on this page.

Partition Memory
per node
Nodes
partition
Cores
per user
GPUs
per node
Runtime Notes
test-gpu 497 6 8 4 preemptible (1)
Kannel 120G 16 4096 7 days KNL nodes (2)
GPU 497G 16 16 4 7 days P100, V100 (3)
PowerNV 500G 8 1216 4 7 days P100, V100 (4)

 

Storage
System
Amount    Notes
/home 50GB   user limit, same system as Deigo
/bucket 50TB   unit-limit; same system as Deigo
/work 10TB   Per-unit limit

 

  1. The test-gpu partition is accessible to anybody. However, it is a low-priority partition; if somebody using a restricted partition wants to use the hardware, your job may be suspended or killed. This is best used for development and testing, not long computations.
     
  2. You can use all of the knl nodes at once for massively parallel computations. This is an especially good resource for developing and testing parallel code. However, we may ask you to reduce the number if other users are waiting for the resource.
     
  3. You could take 16 cores on a single node; or up to 16 nodes, each with a single core. This is a popular system. If you are using too much resources at once, and blocking other users form doing their work we may terminate your job. Please be mindful of how much you use.
     
  4. The PowerNV system is meant for experienced users, and we keep a hands-off approach to maintaining it. If you are a PowerNV user and have specific needs that the default settings do not cover you are welcome to come by and discuss this with us.

Storage

This summarizes the public storage systems at OIST, and which system uses each one. Also see our pages on the research storage system for more up to date information.

System Storage Amount    Notes
Deigo, Saion /home 50G   Per-user
Deigo /flash 10T   Per-unit
Deigo, Saion /bucket 50T   Per-unit
Saion /work 10T   Per-unit
All naruto   Tape for archiving