llnl / llnl/scr

Using shared storage

Open
#133 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C
Stars
108
Forks
37
PR merge metrics
No merged PRs in 30d

Description

Hi,

I have following `scr.conf` which indicates that a checkpoint should periodically be written on shared storage after a checkpoint on local node's storage.

```
SCR_COPY_TYPE=FILE

STORE=/tmp GROUP=NODE COUNT=1
STORE=/dev/shm GROUP=NODE COUNT=2
STORE= GROUP=WORLD COUNT=1

CKPT=0 INTERVAL=1 GROUP=NODE STORE=/dev/shm TYPE=SINGLE COUNT=2
CKPT=1 INTERVAL=2 GROUP=WORLD STORE= TYPE=SINGLE COUNT=1
```

The problem is that after an interrupt the code will not be restarted from the shared storage (it starts from the beginning). However if I use XOR or PARTNER scheme the code will be restarted successfully from the last checkpoint in case of an interrupt. Am I missing something on the definition of the shared storage?

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the scr.conf storage and checkpoint definitions, then compare restart behavior after an interrupt using the FILE scheme versus XOR or PARTNER. Determine why the shared-storage checkpoint is not selected on restart; done means the configuration reliably resumes from the latest shared-storage checkpoint.

Written by the indexing model from the issue text.

Assessment

Tech stack
c
Domain
distributed-systems
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.