'Intel’s commitment to the HPC community is grounded in a pragmatic reality': Intel affirms its backing for the next generation of high-powered computing to push the boundaries of AI
Intel is developing next-generation Xeon processors, AI accelerators, optical networking, and quantum computing technologies while maintaining software compatibility across scientific computing environments.
Intel's commitment to advancing high-performance computing (HPC) is driven by a pragmatic approach that focuses on supporting the evolving needs of AI and scientific research. The company's upcoming Diamond Rapids processors will boast up to 256 performance cores, a memory bandwidth of 1.6 TB/s, and substantial last-level cache capacity of 1.28 GB.
These enhancements will bolster the capabilities of Intel's Xeon 6 processors, which are designed for intensive simulation, mathematical modeling, and other data-intensive applications.
In addition to the processor advancements, Intel is developing its GPU D-series accelerators to cater to research organizations requiring large accelerator memory, alongside artificial intelligence processing and double-precision mathematical computing. These accelerators will support up to 480GB of LPDDR5X memory, ensuring seamless handling of extensive datasets and complex AI models.
The inclusion of conventional PCIe connectivity and air-cooled configurations simplifies installation within existing computing facilities, eliminating the need for specialized cooling infrastructure.
Intel's strategy extends beyond hardware development to encompass software technologies that enable research institutions to retain their existing applications while transitioning to newer computing architectures. The company's software portfolio includes Intel MPI, Fortran compilers, optimization tools, and oneAPI, which facilitate parallel applications across various hardware configurations.
By maintaining compatibility with popular software frameworks and inference environments, Intel ensures that researchers can immediately leverage their established applications without the need for extensive software rewrites.
Beyond hardware and software enhancements, Intel is also investing in optical connectivity and quantum computing. The development of silicon photonics aims to address communication limitations in large supercomputing installations by utilizing optical signals instead of traditional electrical transmission. This technology promises reduced energy consumption and transmission delays, facilitating the connection of processors, memory, and accelerators over greater distances.
Intel's pursuit of quantum computing through the acquisition of Classiq further demonstrates its commitment to exploring new frontiers in high-performance computing. Classiq's software platform simplifies quantum programming, enabling developers with conventional supercomputing experience to delve into quantum applications without directly managing complex hardware operations.
Written by urgent.news from TechRadar's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.