Skip to main content
Ctrl+K
NVIDIA legate (25.03) - Home NVIDIA legate (25.03) - Home

NVIDIA legate (25.03)

NVIDIA legate (25.03) - Home NVIDIA legate (25.03) - Home

NVIDIA legate (25.03)

Table of Contents

Contents:

  • Overview
  • User Guide
    • Installing pre-built packages
      • Networking with Legate Wheels
      • Installation of the Legate MPI wrapper
    • Building from source
    • Usage
    • Using with Jupyter
    • GASNet-based Installation
    • Links to resources
  • Contributing Guide
  • Frequently Asked Questions
  • API Reference
    • Python API Reference
      • Routines
        • legate.core.get_legate_runtime
        • legate.core.get_machine
        • legate.core.track_provenance
      • Classes
        • Legate Runtime
        • Operations
        • Store
        • Machine and Resource Scoping
        • Shape
      • Python Tasks
        • legate.core.task.task
        • legate.core.task.InputStore
        • legate.core.task.OutputStore
        • legate.core.task.ReductionStore
        • legate.core.task.InputArray
        • legate.core.task.OutputArray
        • legate.core.task.ReductionArray
        • legate.core.task.PyTask
        • legate.core.task.VariantInvoker
      • Settings
      • I/O
        • legate.io.hdf5.from_file
        • legate.io.hdf5.kerchunk_read
        • Using GPUDirect Storage to perform HDF5 file I/O
        • legate.core.experimental.io.file_handle.from_file
        • legate.core.experimental.io.file_handle.to_file
        • legate.core.experimental.io.file_handle.FileHandle
    • C++ API Reference
      • Data
      • Partitioning
      • Runtime
      • Mapping
      • Types
      • Typedefs
      • Task
      • Util
      • IO
      • Legate.STL Reference
  • Versions
  • Changelog
    • 2503
    • 2501
    • 2411
    • 2406
  • Third-party notices

Experimental Features:

  • Legate.STL
    • User guide
      • Installation
      • Usage
    • Legate.STL Reference
  • API Reference
  • Python API Reference
  • I/O

I/O#

Utilities for serializing and deserializing Legate stores.

HDF5#

API#

from_file(path, unicode dataset_name)

Read an HDF5 array from disk using the native hdf5 API.

kerchunk_read(filepath, dataset_name)

Read an HDF5 array from disk using Kerchunk and KvikIO.

Examples#

  • Using GPUDirect Storage to perform HDF5 file I/O

KVikIO#

from_file(path, Type array_type)

to_file(path, LogicalArray array)

FileHandle(file[, flags])

File handle for distributed IO.

previous

Settings

next

legate.io.hdf5.from_file

On this page
  • HDF5
    • API
    • Examples
  • KVikIO
NVIDIA NVIDIA
Privacy Policy | Manage My Privacy | Do Not Sell or Share My Data | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2021-2025, NVIDIA.