Intel Corp. today announced a new version of its oneAPI toolkit that enables developers to build applications that can run on multiple types of processors.
Many companies’ technology environments contain multiple types of chips. For example, an organization might run a database on a central processor and an artificial intelligence application on a graphics processor card. Some companies also use more specialized chips such as field programmable gate arrays.
Applications that run on one type of chip are not necessarily compatible with another type of chip. Porting an application to a new processor architecture often requires developers to make significant code changes. The more chips your workload needs to support, the more code changes are required.
Intel’s oneAPI Toolkit makes it easy to build applications that can run on multiple types of chips. According to the company, oneAPI reduces the amount of code that must be changed when porting an application from one processor architecture to another. As a result, developers can complete software projects faster.
The latest version of oneAPI announced by Intel today features support for some of the company’s upcoming data center processors. It also improves compatibility with silicon from competing chip makers.
According to Intel, the new oneAPI release will support future support. sapphire rapids Server CPU series.Company’s recently introduced Also supports the Xeon Max series. The latter product line includes CPUs optimized for high-performance computing systems, especially supercomputers.
Developers can also use oneAPI to build applications for Intel’s upcoming line of data center graphics cards. The centerpiece of the product line is the Max series, chips with over 100 billion transistors. It consists of 47 chiplets (computing modules) optimized to run machine learning software.
Some applications require the ability to run on GPUs from multiple suppliers instead of just one. To facilitate development of such workloads, Intel is extending oneAPI with a collection of new software plugins. The plugin was developed by Codeplay Software, an Edinburgh-based company developed by Intel. got Earlier this year.
Developers write oneAPI applications using a programming language called SYCL. Code written in a language can run on multiple types of chips. According to Intel, the new oneAPI plugin will make it easier for him to write SYCL code that can run on graphics processing units from Nvidia Corp. and Advanced Micro Devices Inc.
Nvidia provides a software toolkit called CUDA with their GPUs. Enterprises use toolkits to optimize the performance of GPU-powered applications.
By default, CUDA-powered applications built for Nvidia chips are not compatible with other graphics cards. As a result, porting such an application to another chip manufacturer’s GPU can require significant work. To make this task easier, Intel has updated oneAPI with features that make it easy to convert CUDA-powered software to SYCL code that supports multiple types of graphics cards.
Intel is simplifying development of GPU-based applications in other ways as well. As part of today’s update, the company added the ability to automatically run a roofline analysis of your workload. Roofline analysis is a performance evaluation method that helps developers find ways to speed up their software.
Timothy Williams, Deputy Director of Computational Sciences at Argonne National Laboratory, said: “Leadership-class computational science emphasizes the benefits of code portability from its multi-vendor, multi-architecture programming standards.”
In addition to the new GPU features, the latest release of oneAPI also includes several other enhancements. Intel has enhanced support for OpenMP, an open source tool used by developers to build applications that can run on multiple processor cores simultaneously. Intel also enhanced his oneMKL component of oneAPI, which makes it easier to build applications that can perform complex math operations.