ASRock preps AMD GPUs for AI inference and multi-GPU systems
ASRock introduces AMD GPUs in Creator series for AI and multi-GPU systems. Featuring 16-pin power connectors, vapor chamber cooling, and practical design, they target generative AI and workstation tasks effectively.
Read original article
ASRock is preparing AMD GPUs for AI inference and multi-GPU systems with its Creator series. These GPUs feature a dual-slot, blower-type design and utilize 16-pin power connectors, a departure from the previous 8-pin PCIe connectors. The transition to 16-pin connectors aligns with newer high-wattage power supplies that support multi-GPU setups. The Creator series graphics cards are tailored for generative AI workloads and workstation environments, offering a balance of performance and convenience. Despite the focus on functionality over aesthetics, ASRock's decision to adopt the 16-pin power connectors aims to streamline cable management and cater to evolving power supply standards. The GPUs feature a vapor chamber cooling system and omit RGB lighting effects, emphasizing practicality for multi-GPU configurations. While ASRock's move showcases advancements in AMD GPU technology, the market's preference between AMD and Nvidia offerings remains a point of interest, with AMD addressing its shortcomings through firmware documentation and open-sourcing software. Overall, ASRock's Creator series presents a competitive option for users seeking efficient GPU solutions for AI and multi-GPU setups.
Related
Intel's Gaudi 3 will cost half the price of Nvidia's H100
Intel's Gaudi 3 AI processor is priced at $15,650, half of Nvidia's H100. Intel aims to compete in the AI market dominated by Nvidia, facing challenges from cloud providers' custom AI processors.
Testing AMD's Giant MI300X
AMD introduces Radeon Instinct MI300X to challenge NVIDIA in GPU compute market. MI300X features chiplet setup, Infinity Cache, CDNA 3 architecture, competitive performance against NVIDIA's H100, and excels in local memory bandwidth tests.
AMD MI300X performance compared with Nvidia H100
The AMD MI300X AI GPU outperforms Nvidia's H100 in cache, latency, and inference benchmarks. It excels in caching performance, compute throughput, but AI inference performance varies. Real-world performance and ecosystem support are essential.
GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity
Major companies like AMD, Intel, and Nvidia are considering supporting Panmnesia's CXL IP for GPU memory expansion using PCIe-attached memory or SSDs. Panmnesia's low-latency solution outperforms traditional methods, showing promise for AI/HPC applications. Adoption by key players remains uncertain.
GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity
Major companies like AMD, Intel, and Nvidia are considering supporting Panmnesia's CXL IP technology for GPU memory expansion using PCIe-attached memory or SSDs. The low-latency CXL IP addresses increasing memory needs for AI training datasets, offering improved GPU performance for AI and HPC applications. Adoption by GPU manufacturers is uncertain, potentially shaping future GPU memory expansion technologies.
1 commentsRelated
Intel's Gaudi 3 will cost half the price of Nvidia's H100
Intel's Gaudi 3 AI processor is priced at $15,650, half of Nvidia's H100. Intel aims to compete in the AI market dominated by Nvidia, facing challenges from cloud providers' custom AI processors.
Testing AMD's Giant MI300X
AMD introduces Radeon Instinct MI300X to challenge NVIDIA in GPU compute market. MI300X features chiplet setup, Infinity Cache, CDNA 3 architecture, competitive performance against NVIDIA's H100, and excels in local memory bandwidth tests.
AMD MI300X performance compared with Nvidia H100
The AMD MI300X AI GPU outperforms Nvidia's H100 in cache, latency, and inference benchmarks. It excels in caching performance, compute throughput, but AI inference performance varies. Real-world performance and ecosystem support are essential.
GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity
Major companies like AMD, Intel, and Nvidia are considering supporting Panmnesia's CXL IP for GPU memory expansion using PCIe-attached memory or SSDs. Panmnesia's low-latency solution outperforms traditional methods, showing promise for AI/HPC applications. Adoption by key players remains uncertain.
GPUs can now use PCIe-attached memory or SSDs to boost VRAM capacity
Major companies like AMD, Intel, and Nvidia are considering supporting Panmnesia's CXL IP technology for GPU memory expansion using PCIe-attached memory or SSDs. The low-latency CXL IP addresses increasing memory needs for AI training datasets, offering improved GPU performance for AI and HPC applications. Adoption by GPU manufacturers is uncertain, potentially shaping future GPU memory expansion technologies.