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

Skew Join (someone please implement) #182

Closed
johnynek opened this issue Sep 14, 2012 · 1 comment
Closed

Skew Join (someone please implement) #182

johnynek opened this issue Sep 14, 2012 · 1 comment

Comments

@johnynek
Copy link
Collaborator

Here is what we do:

  1. sample the left and right side way down (like 1/1000, but probably a parameter of the skewJoin).
  2. do a join of the left and right (maybe hashjoin if inner join is okay).
  3. esimate the number of items on the left and right for each key.
  4. set a replication factor for each key for each side, such that every (key, replication) bucket has about the same number of items.
  5. leftJoinWithTiny this replication factor for each key, if there is no replication factor choose 1.
  6. adapt blockJoin to read the key replication from the pipe, so that we can use different replications for each key.
  7. update Matrix API to have skew hinting, and if it is set, use skewJoin in matrix product.
  8. Profit.
@johnynek
Copy link
Collaborator Author

Done: #229

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant