rapidsai/dask-cuda

Issues

• Add cli option to enable pytorch to use same memory pool as rapids.

  #1281 opened 2 months ago by VibhuJawa
  2

• DASK Deployment using SLURM with GPUs

  #1381 opened 4 months ago by AquifersBSIM
  7

• [CI] add wheel tests in CI

  #1344 opened 7 months ago by jameslamb
  4

• Allow enabling CUDF_SPILL via an environment variable

  #1374 opened 4 months ago by ayushdg
  0

• Slurm Cluster CPU Affinity

  #1358 opened 5 months ago by ilan-gold
  10

• Allow deregistering cuDF from Dask's spilling mechanism

  #1363 opened 5 months ago by pentschev
  0

• Add option to use `cudf` spilling with `dask cuda`

  #1280 opened 5 months ago by VibhuJawa
  2

• Explicit-comms shuffle produces different partitioning than `"tasks"`

  #1355 opened 6 months ago by rjzamora
  0

• CPU Memory Usage for Tasks with CPU-GPU Transfer

  #1351 opened 6 months ago by ilan-gold
  28

• cudaErrorDevicesUnavailable: CUDA-capable device(s) is/are busy or unavailable

  #1342 opened 7 months ago by trivialfis
  2

• Error when installing latest version (24.6.0) of dask-cuda

  #1348 opened 7 months ago by j-alchimowic
  2

• Pin newer pandas version

  #1336 opened 8 months ago by secrettoad
  2

• Support "dataframe.query-planning" config in ``dask.dataframe``

  #1311 opened 10 months ago by rjzamora
  0

• Broken link in explicit_comms docs

  #1328 opened 9 months ago by yoldemir
  1

• `test_dataframe_shuffle`: (CuPy's?) CUB emits `warning: cuda_runtime_api.h [jitify] File not found`

  #1317 opened 10 months ago by jakirkham
  3

• `dask.dataframe` `DeprecationWarning` turns into error (no `dask.dataframe`)

  #1319 opened 10 months ago by jakirkham
  3

• Conda-forge packaging dependencies

  #1208 opened a year ago by ericpre
  4

• Support Pandas version 2

  #1306 opened a year ago by siddharthab
  10

• Dask-CUDA 23.12.0 requires `rapids-dask-dependency` on PyPI

  #1297 opened a year ago by jakirkham
  2

• New version bug

  #1288 opened a year ago by raphael7777777
  3

• `dask cuda` CLI doesn't work with click 8.0.x

  #1271 opened a year ago by jacobtomlinson
  0

• Failing tests on `distributed>2023.9.2`

  #1265 opened a year ago by pentschev
  3

• Dask LocalCudaCluster compute error when `threads_per_worker` not equal to 1

  #1262 opened a year ago by kanglcn
  5

• Specifying Worker Listen Port

  #1253 opened a year ago by otavioon
  2

• UCX serialization errors after `dumps_task` removal from Distributed

  #1246 opened a year ago by pentschev
  4

• [Dask-CUDA] CUDA 12 Conda Packages for ARM

  #1232 opened a year ago by jarmak-nv
  0

• Dask-Cuda running out of memory with Cupy Sobel

  #1250 opened a year ago by a1234jehad
  8

• Add a cluster/worker option to log cuDF spilling statistics

  #1254 opened a year ago by charlesbluca
  1

• [BUG] Explicit-comms shuffle does not obey `"_partitions"` column

  #1239 opened a year ago by rjzamora
  1

• `FutureWarning`s in `LocalCluster` not being raised as errors

  #1242 opened a year ago by pentschev
  0

• CI failures and `ConnectionPool` errors

  #1236 opened a year ago by pentschev
  5

• Workers keep running regardless of retire all workers and close the client

  #1235 opened a year ago by LUOXIAO92
  2

• [BUG] dask scheduler not working: dask_cuda/cli.py AttributeError: 'function' object has no attribute 'command'

  #1179 opened a year ago by pyotr777
  10

• `dask cuda` command raises errors with the latest alpha release

  #1229 opened a year ago by vijayaditya
  3

• Dashboard for cuDF spilling

  #1226 opened a year ago by pentschev
  1

• Tests failing after Distributed upstream changes

  #1224 opened a year ago by pentschev
  3

• Cannot import `dask_cuda` with `distributed:main`

  #1219 opened a year ago by rjzamora
  0

• Pooling memory (DRAM) from multiple GPUs

  #1216 opened a year ago by Dillon214
  4

• Remove versioneer

  #1203 opened a year ago by bdice
  0

• Use KvikIO when available even when GDS isn't available

  #1206 opened a year ago by madsbk
  0

• Stop Using Versioneer

  #1202 opened 2 years ago by quasiben
  1

• Error serializing CuPy==12.0.0 array

  #1174 opened 2 years ago by pentschev
  6

• Make `dask_cuda.cli runnable` as a script with `python -m`

  #1180 opened 2 years ago by hmacdope
  0

• Problem in LocalCUDACluster creating worker

  #1173 opened 2 years ago by MRTater
  6

• errors while use ucx with {rmm_managed_memory: True} option

  #1165 opened 2 years ago by LUOXIAO92
  3

• dask_cuda memory error when calculating dask.array correlation

  #1162 opened 2 years ago by Rridley7
  10

• Heteogenous CUDA Cluster

  #1161 opened 2 years ago by jackyko1991
  6

• Is the use of `__main__` necessary for `LocalCUDACluster`?

  #1152 opened 2 years ago by trivialfis
  3

• Can't start LocalCUDACluster w/ ucx

  #1148 opened 2 years ago by randerzander
  22

• [FEA] Expose ability to configure cudf's copy on write in LocalCUDACluster

  #1150 opened 2 years ago by randerzander
  0