[RELEASE] dask-cuda v0.13 #267
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[gpuCI] Auto-merge branch-0.12 to branch-0.13 [skip ci]
This is not needed when setting the `pure=False` argument when submitting
to the workers
Fix UCX configurations
Adding `ignore_index` argument to `partition_by_hash()`
In Dask 2.9.0, support for computing the `sizeof` of Numba `DeviceNDArray`s was included. So bump our Dask version requirement so that we get this fix. Can drop our own vendored version of this code as a result.
As Dask 2.9.0+ has code for determining `sizeof` for Numba `DeviceNDArray`s, there is no need for us to carry around our own code for this. So go ahead and drop it.
Drop Numba `DeviceNDArray` code for `sizeof`
Support spilling of device objects in dictionaries
Update changelog for 0.13
As these serialization function names are important and are meant to be kept around, use proper names and not `_` (which is intended for values not to be kept).
As all `"cuda"` serializable objects are now also `"dask"` serializable, switch to just using `"dask"` serialization to perform device-to-host transfers to spill to host memory. Though continue falling back to `"pickle"` if nothing better can be found (after all that will be on host too). Similarly "unspilling" from host-to-device will happen naturally as part of the deserialization step. Should simplify what Dask-CUDA needs to keep track of/do.
This should ensure the frames can go through `"dask"` serialization when `DeviceSerialized` is serialized.
Note these frames are `bytes`-like. IOW C-contiguous buffers that can be easily loaded back into device memory or spilled to disk. https://docs.python.org/3/glossary.html#term-bytes-like-object
…o_host Use `"dask"` serialization to move to/from host
jakirkham
approved these changes
Mar 27, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
❄️ Code freeze for
branch-0.13and v0.13 releaseWhat does this mean?
Only critical/hotfix level issues should be merged into
branch-0.13until release (merging of this PR).What is the purpose of this PR?
branch-0.13intomasterfor the release