en.Wedoany.com Reported - AWS has announced the launch of Amazon Elastic Compute Cloud (Amazon EC2) G7 instances, powered by NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, providing acceleration for AI inference, graphics processing, and analytical workloads. AWS is the first major cloud provider to support this GPU.
G7 instances are accelerated by the NVIDIA RTX PRO 4500 Blackwell Server Edition GPU in conjunction with custom 6th Gen Intel Xeon Scalable Processors. Compared to G6 instances, G7 instances deliver up to 4.6x higher AI inference performance and up to 2.1x higher graphics performance. These instances also provide faster performance for GPU-accelerated analytics on Amazon EMR on Amazon Elastic Kubernetes Service (Amazon EKS). Use cases include AI inference, graphics rendering, video transcoding and analysis, spatial computing, virtual desktop infrastructure (VDI), and data analytics.
G7 instances offer significant advancements in GPU memory. The NVIDIA RTX PRO 4500 Blackwell Server Edition GPU features 32 GB of memory, combined with 5th Gen Tensor Cores and 4th Gen RT Cores, providing 1.33x the GPU memory capacity and 2.45x the GPU memory bandwidth compared to G6 instances. In terms of networking and storage, G7 instances deliver up to 700 Gbps of EFA network throughput, a 7x improvement over G6 instances, enabling low-latency, high-bandwidth connectivity for AI inference and graphics-intensive applications. Local NVMe SSD storage supports up to 7.6 TB, reducing data transfer overhead and improving throughput. For video encoding, the 9th Gen NVENC and 6th Gen NVDEC engines support 4:2:2 encoding and decoding, with a 1.5x increase in concurrent video streams compared to G6 instances.
G7 instances can be configured with up to 8 NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, providing a total GPU memory of up to 256 GB (32 GB per GPU). The instances feature custom Intel Xeon Scalable Processors and are available in 7 sizes, supporting up to 192 vCPUs, up to 700 Gbps network bandwidth, up to 768 GiB of system memory, and up to 7.6 TB of local NVMe SSD storage.
In terms of technical features, G7 instances support NVIDIA GPUDirect P2P for multi-GPU configurations, as well as NVIDIA GPUDirect RDMA for EFA and Amazon FSx for Lustre, enabling low-latency GPU-to-GPU communication for multi-GPU and multi-node workloads.
G7 instances can be quickly deployed using AWS Deep Learning AMIs (DLAMI) or NVIDIA Workstation AMIs, which come pre-installed with GPU drivers for AI inference and graphics workloads. On Amazon EKS, users can leverage automated tools provided by EKS to build EKS AMIs based on NVIDIA driver version R595. The instances are compatible with operating systems such as Amazon Linux, Ubuntu, RHEL, and Windows Server, and support industry-standard graphics libraries including DirectX, Vulkan, and OpenGL. Amazon EC2 G7 instances are currently available in two AWS Regions: US East (Ohio) and US West (Oregon).
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
