'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
(Image credit: Intel)
- Intel's Diamond Rapids processors will feature up to 256 performance cores
- The upcoming Xeon architecture promises memory bandwidth reaching 1.6 TB/s
- Intel's GPU D-series accelerators will support up to 480GB LPDDR5X memory
Intel is advancing its high-performance computing technologies through new Xeon processors, dedicated AI accelerators, and developments in optical connectivity and quantum computing.
These investments are part of the company's strategy to support increasingly demanding scientific applications while allowing research institutions to retain existing software and computing infrastructure.
Jeff McVeigh, Intel's vice president and general manager of Data Center Strategic Programs, says the company's approach prioritizes cooperation between different hardware and software technologies rather than dependence on a single computing architecture.
Diamond Rapids and GPU D-series address computational demands
Intel's forthcoming Diamond Rapids server processors will increase computing capacity through additional processing cores, expanded cache resources, and substantially higher memory throughput.
The architecture will support configurations containing up to 256 performance cores, accompanied by memory bandwidth reaching 1.6 TB/s and last-level cache capacity of 1.28 GB.
Additional capabilities include Intel Advanced Performance Extensions, AVX 10.2, and improved Advanced Matrix Extensions, providing hardware acceleration for demanding scientific calculations and artificial intelligence workloads.
Intel says Diamond Rapids will extend the capabilities of its Xeon 6 processors, which support intensive simulation, mathematical modeling, and other computationally demanding applications.
Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!
The company is also developing its GPU D-series for research organizations requiring substantial accelerator memory alongside artificial intelligence processing and double-precision mathematical computing capabilities.
These accelerators will accommodate up to 480GB of LPDDR5X memory, with supported numerical formats extending from FP4 and MXFP4 to FP64.
Their conventional PCIe connectivity and air-cooled configuration are intended to simplify installation within existing computing facilities without requiring specialized cooling infrastructure.
Greater onboard memory capacity could reduce repeated transfers between processors and accelerators, particularly when scientific applications handle extensive datasets or computationally demanding artificial intelligence models.
Software compatibility, optical networking and quantum computing
Beyond processor development, Intel is maintaining investments in software technologies that allow research institutions to retain existing applications while adopting newer computing architectures.
Its software portfolio includes Intel MPI, Fortran compilers, optimization tools, and oneAPI, which support parallel applications across different hardware configurations.
McVeigh argues that introducing faster processors becomes less useful when developers must abandon established applications or rewrite substantial amounts of existing software.
Intel therefore intends to prepare software support alongside new hardware releases, allowing commonly used artificial intelligence frameworks and inference environments to operate immediately.
This approach also supports hybrid scientific workflows in which conventional processors coordinate applications while accelerators undertake computationally intensive artificial intelligence operations.
Intel is additionally pursuing silicon photonics to address communication limitations associated with moving information across increasingly large supercomputing installations.
The technology uses optical signals instead of conventional electrical transmission, potentially reducing energy consumption and transmission delays as connections extend between processors, memory, and accelerators.
Quantum computing represents another research direction, with Intel proposing to acquire Classiq to combine automated quantum algorithm development with its silicon spin-qubit technology.
Classiq's software platform is designed to simplify quantum programming, potentially allowing developers familiar with conventional supercomputing environments to explore quantum applications without directly managing complex hardware operations.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.
Efosa has been writing about technology for over 7 years, initially driven by curiosity but now fueled by a strong passion for the field. He holds both a Master's and a PhD in sciences, which provided him with a solid foundation in analytical thinking.
What's Your Reaction?
Like
0
Dislike
0
Love
0
Funny
0
Wow
0
Sad
0
Angry
0
Comments (0)