rapidsai/dask-cuda

View on GitHub

Utilities for Dask and CUDA interactions

Stars: 324
Forks: 112
Open beginner issues: 0
Indexed issues: 78
Avg merge: 9h 18m
Merged PRs (30d): 8

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 17, 2026
Latest indexed: Sep 19, 2026
Contributing guide: No contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

78 open issues indexed

[DOC] RAPIDS -> nvidia docs migration TODOs Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

rapidsai/dask-cuda#1685 · Opened Sep 11, 2026
Multi-node Dask-CUDA/cuGraph workload cannot form a shared GPU cluster across Kubeflow Trainer pods Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

rapidsai/dask-cuda#1681 · 3 comments · Opened Sep 2, 2026
LocalCUDACluster + .persist() produces confusing "Missing dependency" errors when downstream code uses scheduler="threads" Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

rapidsai/dask-cuda#1649 · 3 comments · Opened May 13, 2026
Intermittent failures in `test_worker_force_spill_to_disk` Open

bug non-breaking

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#1561 · 5 comments · Opened Sep 26, 2025
Add spill-to-disk tests for devices without a memory resource Open

Difficulty 3/5 1-2 days Newbie friendliness 42/100

rapidsai/dask-cuda#1510 · 1 reaction · Opened Jun 17, 2025
Nightly tests cannot download artifacts Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 32/100

rapidsai/dask-cuda#1474 · Opened Apr 25, 2025
`test_pre_import` fails in some environments Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

rapidsai/dask-cuda#1441 · Opened Feb 4, 2025
TypeError: Unsupported type `<class 'numpy.ndarray'>` Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

rapidsai/dask-cuda#1405 · 6 comments · Opened Nov 19, 2024
localcudacluster with onnxruntime inference Open

Difficulty 4/5 3-5 days Newbie friendliness 18/100

rapidsai/dask-cuda#1399 · Opened Oct 22, 2024
Allow enabling CUDF_SPILL via an environment variable Open

Difficulty 3/5 1-2 days Newbie friendliness 48/100

rapidsai/dask-cuda#1374 · Opened Aug 13, 2024
CPU Memory Usage for Tasks with CPU-GPU Transfer Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#1351 · 29 comments · Opened Jun 21, 2024
cudaErrorDevicesUnavailable: CUDA-capable device(s) is/are busy or unavailable Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

rapidsai/dask-cuda#1342 · 2 comments · Opened May 30, 2024
`test_dataframe_shuffle`: (CuPy's?) CUB emits `warning: cuda_runtime_api.h [jitify] File not found` Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

rapidsai/dask-cuda#1317 · 3 comments · Opened Feb 28, 2024
Support "dataframe.query-planning" config in ``dask.dataframe`` Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 25/100

rapidsai/dask-cuda#1311 · Opened Feb 12, 2024
New version bug Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

rapidsai/dask-cuda#1288 · 3 comments · Opened Nov 26, 2023
Failing tests on `distributed>2023.9.2` Open

rapidsai/dask-cuda#1265 · 3 comments · 1 assignee · Opened Oct 26, 2023
Add a cluster/worker option to log cuDF spilling statistics Open

Difficulty 5/5 Over a week Newbie friendliness 30/100

rapidsai/dask-cuda#1254 · 1 comment · Opened Sep 29, 2023
Dask-Cuda running out of memory with Cupy Sobel Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

rapidsai/dask-cuda#1250 · 8 comments · Opened Sep 28, 2023
UCX serialization errors after `dumps_task` removal from Distributed Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

rapidsai/dask-cuda#1246 · 4 comments · Opened Sep 26, 2023
`FutureWarning`s in `LocalCluster` not being raised as errors Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#1242 · Opened Sep 26, 2023
CI failures and `ConnectionPool` errors Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

rapidsai/dask-cuda#1236 · 5 comments · Opened Sep 25, 2023
Workers keep running regardless of retire all workers and close the client Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

rapidsai/dask-cuda#1235 · 2 comments · Opened Sep 24, 2023
Conda-forge packaging dependencies Open

Difficulty 3/5 1-2 days Newbie friendliness 20/100

rapidsai/dask-cuda#1208 · 4 comments · Opened Jul 6, 2023
Use KvikIO when available even when GDS isn't available Open

Difficulty 2/5 1-3 hours Newbie friendliness 55/100

rapidsai/dask-cuda#1206 · Opened Jun 28, 2023
Heteogenous CUDA Cluster Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

rapidsai/dask-cuda#1161 · 6 comments · Opened Apr 14, 2023
[FEA] Expose ability to configure cudf's copy on write in LocalCUDACluster Open

Difficulty 3/5 1-2 days Newbie friendliness 38/100

rapidsai/dask-cuda#1150 · Opened Mar 31, 2023
Failing Out-of-Core Merge Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

rapidsai/dask-cuda#1143 · Opened Mar 27, 2023
test_cuda_visible_devices_and_memory_limit_and_nthreads spews benign (?) warnings on systems with fewer than eight GPUs Open

rapidsai/dask-cuda#1127 · 3 comments · 1 assignee · Opened Feb 23, 2023
Make all `LocalCluster` tests async Open

Difficulty 4/5 3-5 days Newbie friendliness 28/100

rapidsai/dask-cuda#1087 · 1 comment · Opened Jan 18, 2023
LocalCUDACluster doesn't come up cleanly after client.restart Open

rapidsai/dask-cuda#1075 · 6 comments · 1 assignee · Opened Jan 6, 2023
Raise error when setting communication protocols if only 1 GPU is used or if hardware is uncapable Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

rapidsai/dask-cuda#1066 · 1 comment · Opened Dec 14, 2022
Optional Python dependencies Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

rapidsai/dask-cuda#1054 · 7 comments · Opened Nov 23, 2022
Running large inputs results in "Worker process still alive after 3.1999992370605472 seconds, killing" Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

rapidsai/dask-cuda#1051 · 5 comments · Opened Nov 23, 2022
CI fails due to Jupyter errors Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

rapidsai/dask-cuda#1040 · Opened Nov 15, 2022
Use `profile = "black"` to simplify `isort` Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

rapidsai/dask-cuda#1039 · Opened Nov 14, 2022
Program crashes when running program repeatedly Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#1037 · 14 comments · Opened Nov 11, 2022
Configuration Profiles Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

rapidsai/dask-cuda#1020 · 8 comments · Opened Oct 18, 2022
Memory Fragmentation Meta Issue Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

rapidsai/dask-cuda#1000 · Opened Sep 28, 2022
[Bug] Proxy tests fail for matmul/imatmul Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 35/100

rapidsai/dask-cuda#995 · 1 comment · Opened Sep 26, 2022
Switch on `Jit-Unspill` by default Open

inactive-30d

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#990 · 5 comments · Opened Sep 20, 2022
Benchmark Spilling Open

inactive-30d

Difficulty 4/5 3-5 days Newbie friendliness 30/100

rapidsai/dask-cuda#988 · 1 comment · Opened Sep 12, 2022
which best practices are already handled for you by CUDAWorker? Open

Difficulty 3/5 1-2 days Newbie friendliness 40/100

rapidsai/dask-cuda#983 · 3 comments · Opened Aug 29, 2022
Setting `CUPY_ACCELERATORS` Open

inactive-30d

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#972 · 5 comments · 1 reaction · Opened Aug 8, 2022
Get rid of setting `CUDA_VISIBLE_DEVICES` as an environment variable to `Nanny` Open

rapidsai/dask-cuda#954 · 17 comments · 1 assignee · Opened Jul 21, 2022
Opinionated subclasses of other cluster managers Open

inactive-30d

Difficulty 5/5 Over a week Newbie friendliness 20/100

rapidsai/dask-cuda#942 · 3 comments · Opened Jul 1, 2022
Performance regression in cuDF merge benchmark Open

inactive-30d

Difficulty 4/5 3-5 days Newbie friendliness 30/100

rapidsai/dask-cuda#935 · 7 comments · Opened Jun 16, 2022
Problems using preload to get scheduler address Open

inactive-30d

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#907 · 11 comments · Opened May 9, 2022
[FEA] Expose ability to turn off/down logging Open

inactive-30d inactive-90d

Difficulty 4/5 3-5 days Newbie friendliness 35/100

rapidsai/dask-cuda#875 · 3 comments · Opened Mar 18, 2022
Consider logging spilling Open

inactive-30d inactive-90d

rapidsai/dask-cuda#873 · 3 comments · 1 reaction · 1 assignee · Opened Mar 15, 2022
Multiple Distributed IO threads causes hang with UCX Open

inactive-30d inactive-90d

Difficulty 5/5 Over a week Newbie friendliness 20/100

rapidsai/dask-cuda#863 · 2 comments · Opened Feb 22, 2022

rapidsai/dask-cuda

Get new issues in your inbox