Intel Announces 2023‑2025 Xeon Processor Roadmap: Emerald Rapids, Granite Rapids, Sierra Forest, and Clearwater Forest
Intel unveiled its 2023‑2025 Xeon roadmap, detailing the upcoming Emerald Rapids, Granite Rapids, Sierra Forest, and Clearwater Forest CPUs—highlighting core counts, cache sizes, process technologies, performance targets for HPC, AI, and data‑center workloads, and also previewing future GPU, AI accelerator, and FPGA products.
Intel officially released its Xeon processor roadmap for 2023‑2025, outlining a series of new P‑Core and E‑Core products that will succeed the current generation.
2023 – Emerald Rapids
Emerald Rapids, based on Intel 7 process, is slated for Q4 2023 and targets higher performance‑per‑watt for HPC, AI, and data‑center workloads. It will feature up to 64 cores (64C/128T) with a maximum 320 MB L3 cache, improving IPC by 5‑10% over Golden Cove and offering configurations compatible with the existing Eagle Stream platform.
Intel Emerald Rapids‑SP (64‑core SKU): 320 MB L3 + 128 MB L2 = 448 MB
AMD EPYC Genoa (64‑core SKU): 384 MB L3 + 96 MB L2 = 480 MB
Intel Sapphire Rapids‑SP (60‑core SKU): 112.5 MB L3 + 120 MB L2 = 232.5 MB
2024 – Granite Rapids and Sierra Forest
In 2024 Intel will launch Granite Rapids and Sierra Forest using the Intel 3 process and Redwood Cove cores. Granite Rapids‑SP will increase core density, support DDR5‑8800 MCR memory, and deliver up to 1.5 TB/s platform bandwidth, with a demo of DDR5‑8000 MT/s on a dual‑socket board.
Sierra Forest, the first E‑Core Xeon, is expected in early 2024 with up to 144 cores on Intel 3, aimed at cloud‑optimized workloads.
2025 – Clearwater Forest
Clearwater Forest will replace the current line in 2025 as the second‑generation E‑Core Xeon family, built on the Intel 18A process and optimized RibbonFET architecture, promising higher core counts and further performance gains.
Additional Roadmap Highlights
Intel also updated its future GPU, AI accelerator, and FPGA plans, introducing the Melville Sound data‑center GPU Flex series, Falcon Shores accelerators (succeeding Rialto Bridge), next‑generation Habana Gaudi AI accelerators, and new eASIC and AGILEX FPGA families.
The roadmap demonstrates Intel’s commitment to continuous innovation in performance, efficiency, and scalability across high‑performance computing, artificial intelligence, and data‑center markets.
Architects' Tech Alliance
Sharing project experiences, insights into cutting-edge architectures, focusing on cloud computing, microservices, big data, hyper-convergence, storage, data protection, artificial intelligence, industry practices and solutions.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.