P3 Description
The Go SDK produces correct results, but it also does redundant recomputations, or allocations. This Jira is to track performance improvements along any per-element path, where that overhead could be significant.
Contributions should include at least one benchmark with before and after values to demonstrate the value of the change.
Attachments
Issue Links
1.
|
Reduce ParDo per element Invoke overhead |
|
Resolved | Robert Burke |
|
|||||||
2.
|
Reduce metrics overhead |
|
Resolved | Robert Burke |
|
|||||||
3.
|
Use explicit receivers for method invocations. |
|
Resolved | Robert Burke | ||||||||
4.
|
Consolidate defer overhead per bundle |
|
Resolved | Holden Karau |
|
|||||||
5.
|
Consider using sync/atomic for Go SDK metrics. |
|
Resolved | Robert Burke | ||||||||
6.
|
Consider converting bundle & ptransform ids to ints eagerly. |
|
Resolved | Robert Burke | ||||||||
7.
|
Reduce Combine overhead |
|
Resolved | Robert Burke |
|
|||||||
8.
|
Reduce overhead of Go SDK side metrics |
|
Resolved | Robert Burke |
|
