First products, the Flex Series 170 & 140 Break Cover | Panda Anku

Intel has officially announced its Data Center GPU Flex series graphics cards, which are a rebrand of the Arctic Sound M series. As part of its Flex line, Intel is introducing two new graphics cards within the family, the Flex Series 170 and the Flex Series 140.

Intel’s Data Center GPU Flex series is the reincarnation of the arctic sound and comes in two flavors

Press release: The Intel Data Center GPU Flex Series (formerly codenamed Arctic Sound-M) helps customers break free from the limitations of isolated and proprietary environments and reduces the need for data centers to use separate, discrete solutions. Intel offers customers a single graphics processing (GPU) solution designed to flexibly handle a variety of workloads without compromising performance or quality. It also helps lower and optimize the total cost of ownership for various cloud workloads such as media delivery, cloud gaming, AI, Metaverse, and other emerging visual cloud use cases.

“We are in the midst of a pixel explosion driven by more consumers, more applications and higher resolutions. Today’s data center infrastructure is under intense pressure to compute, encode, decode, move, store and display visual information. The Intel Flex Series GPU is a breakthrough design that uniquely meets today’s computing needs while providing flexibility and scalability for tomorrow’s immersive experiences.”

–Jeff McVeigh, Intel vice president and general manager of Super Compute Group

what they offer: The Flex series GPUs meet quality, density and latency requirements. Featuring the industry’s first hardware-based AV1 encoder in a data center GPU, the Flex Series GPU delivers five times the media transcoding throughput performance and twice the decoding throughput performance at half the performance of competing solutions. It offers bandwidth improvement of more than 30 percent for significant total cost of ownership (TCO) savings, as well as broad support for popular media tools, APIs, frameworks and the latest codecs.

Powered by Intel’s Xe-HPG architecture and supported by an extensive ecosystem of hardware vendors and software developers, these GPUs provide flexible scaling of AI inference workloads from media analytics to smart cities to medical imaging between CPUs and GPUs without tying developers to proprietary software.

Why it matters: Media processing, media delivery, visual AI inference, cloud gaming, and desktop virtualization are proliferating in data centers. As a result, the GPU silicon market segment in the data center for Visual Cloud is expected to grow to $15 billion by 2026. This rapid rise stems from an industry largely shaped by reliance on proprietary, licensed coding models such as CUDA for GPU programming.

The Flex Series GPU solution stack overcomes these limitations while offering significant advantages over alternative solutions by improving flexibility, scalability, and power consumption. It contributes to a lower total cost of ownership for solution providers by supporting more subscribers with fewer servers.

The software stack is enabled by oneAPI – the productive path for accelerated computing – freeing developers from the economic and technical burdens of proprietary programming models. It is an open alternative to locking in proprietary languages, enabling the full power of hardware with a complete, proven set of tools that complement existing languages ​​and parallel models. This allows users to develop open, portable code that takes full advantage of different combinations across Intel CPUs and GPUs.

About performance: From the start, the Intel Data Center GPU Flex Series offers media processing and delivery capabilities, as well as Android cloud gaming.

Media processing and delivery: The video processing demands of video conferencing, streaming, and social media have transformed the computing resource requirements of the data center.

  • The Flex Series GPU media architecture with up to four Xe Media Engines focuses on the highest stream density requirements for the modern data center without compromising on quality:
    • Delivers 5x media transcoding throughput and 2x decoding throughput at half performance (Intel Flex Series 140 GPU vs Nvidia A10).
    • Delivers up to 36 streams of 1080p60 transcode throughput per card.
    • Delivers eight streams of 4K60 transcode throughput per card.
    • When scaled to 10 cards in a 4U server configuration, it can support up to 360 streams with HEVC-HEVC 1080p60 transcoding throughput.
  • The Flex Series GPU’s hardware-based, open-source AV1 encoder offers bandwidth savings of more than 30%. The Flex Series GPU’s media engine also supports and enhances performance for a wide range of industry codecs, including HEVC, AVC, and VP9.
  • Leveraging the Intel® Deep Link Hyper Encode feature, the Intel® Data Center GPU Flex Series 140 can meet the industry-standard one-second latency requirement with two devices on a single card while still providing 8K60 real-time transcoding. This feature is available for AV1 and HEVC HDR formats.

Android Cloud Gaming: The explosive growth of cloud gaming has created the need for more flexible solutions that maximize concurrent game streams for users. Engineered for high gaming quality, the Flex Series GPU delivers an unparalleled gaming experience across multiple devices. It is validated for almost 90 of the most popular Google Play Android games.

  • A single Flex Series 170 GPU can achieve up to 68 streams at 720p30, while a single Flex Series 140 GPU can achieve up to 46 streams at 720p30 (measured on select game titles).
  • When scaled with six Flex Series 140 GPU cards, up to 216 streams at 720p30 can be achieved.

About the open software stack: Developers can access a comprehensive software stack that combines open source components and tools to effectively realize Flex Series GPU capabilities for visual cloud workloads, including support for oneAPI and OpenVINO. Intel’s oneAPI tools allow developers to deploy accelerated applications and services, including the Intel oneAPI Video Processing Library (oneVPL), Intel VTune Profiler, and others.

About Availability: Systems with Flex Series GPUs will be available from global vendors such as Dell Technologies, HPE, H3C, Inspur, Lenovo and Supermicro. Solutions using the Flex Series GPU will be expanded in the coming months, starting with media delivery and Android cloud gaming workloads. This is followed by Windows cloud gaming, AI and VDI workloads.

Leave a Comment