Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Datasets] Remove stats actor #31571

Closed
c21 opened this issue Jan 10, 2023 · 4 comments
Closed

[Datasets] Remove stats actor #31571

c21 opened this issue Jan 10, 2023 · 4 comments
Assignees
Labels
data Ray Data-related issues data-observability stale The issue is stale. It will be closed within 7 days unless there are further conversation

Comments

@c21
Copy link
Contributor

c21 commented Jan 10, 2023

What happened + What you expected to happen

Stats actor is responsbile to keep the stats for read tasks only, and be accessed by driver when driver RPC to stats actor. This model has the weird timing issue that if ds.stats() is called immediately after read, the stats may not be up-dated. The reason is when ds.stats() is called and stats actor does not get the metadata from read tasks yet. Read tasks makes RPC to stats actor for sending the metadata. The read stats is stored in stats actor, and fetched into driver memory. so causes this timing issue here.

Plan here is to kill stats actor and move the read stats into driver memory directly.

Versions / Dependencies

master

Reproduction script

on master, this script may or may not print execution stats depending on timing:

import ray
import time

ds = ray.data.range_table(5)

for batch in ds.iter_batches():
    print(batch)

#time.sleep(3)

print(ds.stats())
print(ds.stats())

Issue Severity

Medium: It is a significant difficulty but I can work around it.

@c21 c21 added data Ray Data-related issues data-observability labels Jan 10, 2023
@jianoaix
Copy link
Contributor

Just to note that to keep parity after removal, we need to improve the walltime tracking for stages in block metadata, potentially recording event timestamps. @scottjlee

@stale
Copy link

stale bot commented May 20, 2023

Hi, I'm a bot from the Ray team :)

To help human contributors to focus on more relevant issues, I will automatically add the stale label to issues that have had no activity for more than 4 months.

If there is no further activity in the 14 days, the issue will be closed!

  • If you'd like to keep the issue open, just leave any comment, and the stale label will be removed!
  • If you'd like to get more attention to the issue, please tag one of Ray's contributors.

You can always ask for help on our discussion forum or Ray's public slack channel.

@stale stale bot added the stale The issue is stale. It will be closed within 7 days unless there are further conversation label May 20, 2023
@scottjlee
Copy link
Contributor

keep

@stale stale bot removed the stale The issue is stale. It will be closed within 7 days unless there are further conversation label May 30, 2023
@stale
Copy link

stale bot commented Oct 15, 2023

Hi, I'm a bot from the Ray team :)

To help human contributors to focus on more relevant issues, I will automatically add the stale label to issues that have had no activity for more than 4 months.

If there is no further activity in the 14 days, the issue will be closed!

  • If you'd like to keep the issue open, just leave any comment, and the stale label will be removed!
  • If you'd like to get more attention to the issue, please tag one of Ray's contributors.

You can always ask for help on our discussion forum or Ray's public slack channel.

@stale stale bot added the stale The issue is stale. It will be closed within 7 days unless there are further conversation label Oct 15, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
data Ray Data-related issues data-observability stale The issue is stale. It will be closed within 7 days unless there are further conversation
Projects
None yet
Development

No branches or pull requests

4 participants