MENU

The drivers behind Intel’s Scottish software deal

Feature articles |
By Nick Flaherty

Intel is to buy Scottish compiler developer Codeplay Software in a move to open up its software stack for high performance computing, RISC-V and FPGAs.

The deal aims to boost the open credentials of Intel’s OneAPI, an area where it has struggled to convince customers of its intentions.

Codeplay is a leading developer in open standards, and the move was flagged back in February with a focus on the SYCL. This is a royalty free open standards programming model developed by the Khronos Group to support fully heterogeneous data parallelism using C++ across all kinds of processors, from CPUs to GPU and AI accelerators.

SYCL scales from edge compute devices and automotive ADAS up to exascale supercomputers as well as FPGAs. The end developer typically has thousands or millions of lines of code with software creation increasing each year. This code needs to be ported quickly and achieve highest performance.

This is a key element in Intel’s battle with Nvidia as it launches its GPUs for the data centre, as code developed with Nvidia tools is only licensed for Nvidia silicon, says Intel.

Related articles

Since 2002, Codeplay has raised £3m and enabled over 20 complex processor systems providing software developers with robust and stable development tools. It supplies the NERSC at Lawrence Berkeley National Laboratory as well as a collaboration of the Argonne National Laboratory with Oak Ridge National Laboratory with oneAPI SYCL compilers for both AMD and Nvidia GPUs, highlighting an increasing focus by Intel.

“Intel is further advancing its support of the oneAPI ecosystem. We have signed an agreement to acquire Codeplay Software, a global leader in cross-architecture open standards-based developer technologies,” said Joe Curley, Senior Director – HPC Platform and Ecosystem at Intel. “Codeplay is globally recognized for its expertise and leadership in SYCL, the Khronos open standard programming models used in oneAPI, and its significant contributions to the industry ranging from open ecosystem activities like SYCL and OpenCL to RISC-V, automotive software safety, and medical imaging.

Open standards

Intel and Codeplay have worked together to define and extend open standard programming models that work across multiple platforms and devices. Codeplay is highly active on Khronos standards committees, and currently chair the Khronos SYCL Working Group and some ISO C++ standards, acting as the OpenCL Spec editor, and significantly contributing to numerous other standards.

It develops and maintains compilers and runtimes for CPUs, GPUs, DSPs and many other advanced processor architectures to achieve high processing performance for AI and HPC.

Intel sees the deal as enabling Codeplay to extend the delivery of SYCL into cross-architecture and multi-vendor products, based on open standards and the open source libraries and tools. The company will operate as a subsidiary of Intel to ensure that it keeps connections with other companies.

“Codeplay will continue to work in partnership with organizations across the industry to enable open standard software on the latest cutting edge processors,” said Andrew Richards, CEO of Codeplay.

“Our engineers have worked closely with engineers from Intel alongside other key organizations to bring the SYCL standard to the level of maturity it now has. Intel has chosen SYCL to be at the heart of the oneAPI initiative, a cross-industry, standards-based way to develop for heterogeneous architectures,” he said.

“This puts Codeplay in a strong position to work across the industry to bring SYCL and other open standards to both processor vendors and teams of software developers supporting the stated strategy of both parties. Codeplay is therefore at the heart of Intel’s strategy to democratize oneAPI and SYCL, ensuring that all processors support open standards,” he said.

Intel’s FPGA business (formerly Altera) also benefits. James Reinders returned to Intel last year  and has written about the role of SYCL with Michael Wong of Codeplay, the current SYCL committee chair.

SYCL

The underlying programming model of SYCL, SPMD, has shown to be usable across many architectures. SPMD is how most programmers using Nvidia CUDA/OpenCL/SYCL think: writing code from the perspective of operating on one work item and expecting it to run concurrently on most hardware such that multiple work-items fill vector hardware lanes. SYCL offers a large degree of portability across vendors as well as architectures.

“The licensing for CUDA tools and libraries, from Nvidia, specifically states they must be used to “develop applications only for use in systems with Nvidia GPUs.” Even “open source” from Nvidia includes licensing language restricting key parts in the same manner,” says Reinders.

SYCL grew out of a desire to bring the benefits of OpenCL’s open, multivendor, multiarchitecture approach by providing a standard C++ interface for heterogenous parallel architectures. SYCL implementations often use OpenCL for their implementations, but also have the flexibility to use other backends under the hood as of SYCL2020.

SYCL allows standard C++ to address programming for full heterogeneous computing built on top of ISO C++. This includes device enumeration (info), defining work (kernels), submitting and coordinating work across devices (queue), and managing remote memories.

Codeplay has been instrumental in SYCL from the very beginning an d Intel’s interest in SYCL grew after entering both the FPGA market and announcing the Intel Xe architecture to include GPUs for compute.

sycl.tech/ www.codeplay.com; www.intel.com

Other articles on eeNews Europe

 


Share:

Linked Articles
eeNews Europe
10s