Releases: oneapi-src/oneCCL
Releases · oneapi-src/oneCCL
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.7.1
Intel® oneAPI Collective Communications Library 2021.7.1 has been updated to include functional and security updates. Users should update to the latest version as it becomes available.
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.7
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.6
- Added support for Intel® ITT (Intel® Instrumentation and Tracing Technology) profiling
- Added WSL2 (Windows Subsystem for Linux 2) support
- Moved dependency check for Level Zero from compile-time to runtime
- Bug fixes
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.5.2
What's New:
- Bug fixes
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.5
What's New:
- Added support for output SYCL event to track status of CCL operation
- Added OFI/verbs provider with dmabuf support into package
- Bug fixes
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.4
What's New:
- Memory binding of worker threads is now supported
- NIC filtering by name is now supported for OFI-based multi-NIC
- IPv6 is now supported for key-value store (KVS)
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.3
What's New:
- Added OFI-based multi-NIC support
- Added OFI/psm3 provider support
- Patches & security updates
- Fixed bugs
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.2.1
What's New:
- Bug fixes
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.2
What's New:
- Added float16 datatype support.
- Added ip-port hint for customization of KVS creation.
- Optimized communicator creation phase.
- Optimized multi-GPU collectives for single-node case.
- Bug fixes
Intel(R) oneAPI Collective Communications Library (oneCCL) 2021.1
Intel® oneAPI Collective Communications Library (oneCCL):
- Enables efficient implementations of collectives used for deep learning training (allgatherv, allreduce, alltoall(v), broadcast, reduce, reduce_scatter)
- Provides C++ API and interoperability with DPC++
- Deep Learning Optimizations include:
- Asynchronous progress for compute communication overlap
- Dedication of cores to ensure optimal network use
- Message prioritization, persistence, and out-of-order execution
- Collectives in low-precision data types (int[8,16,32,64], fp[32,64], bf16)
- Linux* OS support only