Dask-cuDF
E890460
Dask-cuDF is a RAPIDS library that enables distributed, GPU-accelerated DataFrame processing by integrating cuDF with Dask for scalable data analytics.
Statements (50)
| Predicate | Object |
|---|---|
| instanceOf |
GPU-accelerated data processing framework
ⓘ
Python library ⓘ software library ⓘ |
| basedOn |
Dask
NERFINISHED
ⓘ
cuDF NERFINISHED ⓘ |
| compatibleWith | Pandas-like DataFrame API via cuDF ⓘ |
| developer | NVIDIA NERFINISHED ⓘ |
| documentation | https://docs.rapids.ai/api/dask-cudf/stable ⓘ |
| ecosystem | RAPIDS AI ecosystem NERFINISHED ⓘ |
| genre |
data analytics library
ⓘ
dataframe library ⓘ distributed computing framework ⓘ |
| homepage | https://rapids.ai ⓘ |
| integratesWith |
Dask
NERFINISHED
ⓘ
RAPIDS cuGraph NERFINISHED ⓘ RAPIDS cuML NERFINISHED ⓘ cuDF NERFINISHED ⓘ |
| license | Apache License 2.0 ⓘ |
| operatingSystem | Linux ⓘ |
| optimizedFor | NVIDIA GPU hardware NERFINISHED ⓘ |
| partOf | RAPIDS NERFINISHED ⓘ |
| programmingLanguage | Python ⓘ |
| purpose |
big data processing on GPUs
ⓘ
distributed GPU-accelerated DataFrame processing ⓘ scalable data analytics ⓘ |
| repository | https://github.com/rapidsai/cudf ⓘ |
| requires |
CUDA-capable GPU
ⓘ
Dask NERFINISHED ⓘ cuDF NERFINISHED ⓘ |
| supportsFeature |
lazy evaluation
ⓘ
multi-GPU scaling ⓘ multi-node distributed execution ⓘ out-of-core computation ⓘ parallel I/O ⓘ task scheduling via Dask ⓘ |
| supportsFormat |
CSV
ⓘ
JSON lines ⓘ ORC ⓘ Parquet NERFINISHED ⓘ |
| supportsLanguage | Python ⓘ |
| supportsOperation |
aggregation
ⓘ
filter ⓘ groupby ⓘ join ⓘ window operations ⓘ |
| typicalUseCase |
distributed feature engineering for machine learning
ⓘ
interactive analytics on large tabular datasets ⓘ large-scale ETL on GPUs ⓘ |
| uses |
CUDA
NERFINISHED
ⓘ
NVIDIA GPUs NERFINISHED ⓘ |
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.