← All stories
● Covered by 1 source Β· 1 reportMedium impact

Amazon EC2 G7 Instances Launch with NVIDIA RTX PRO 4500 GPUs

Aggregated by BrevFeed cloud Β· updated 4d ago
πŸ”– Save

Amazon has launched EC2 G7 instances featuring NVIDIA RTX PRO 4500 GPUs, enhancing AI inference and graphics performance. This introduces significant improvements over previous generations, with up to 4.6x AI inference performance and 700 Gbps networking throughput, impacting workloads across AI, rendering, and analytics sectors.

Key points

Launch Announcement

Amazon has announced the general availability of its EC2 G7 instances, which feature NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs. This makes AWS the first major cloud provider to incorporate these GPUs, aimed at delivering high performance for AI inference, graphics, and data analytics workloads.

Performance Improvements

The G7 instances achieve up to 4.6 times the AI inference performance and 2.1 times the graphics performance compared to their predecessors, the G6 instances. The enhanced specifications include 32 GB GPU memory per unit, 5th Gen Tensor Cores, and 4th Gen RT Cores, facilitating improved performance for a variety of applications.

Networking and Storage Enhancements

These new instances feature 700 Gbps EFA-enabled networking throughput, which is seven times faster than G6 instances. This high bandwidth is essential for low-latency and high-performance connectivity required by GPU-accelerated applications. Additionally, G7 instances offer up to 7.6 TB of NVMe SSD storage to minimize data transfer and optimize performance.

Video Processing Capabilities

G7 instances incorporate advanced video encoding and decoding engines, allowing for better support of high-resolution video workflows. This includes 1.5 times the number of concurrent video streams compared to G6 instances, enhancing capabilities for video transcoding and streaming applications.

Specifications Summary

The EC2 G7 instances support configurations of up to eight NVIDIA GPUs and up to 256 GB of total GPU memory. They are available in seven sizes, accommodating up to 192 vCPUs and 768 GiB of system memory, making them versatile for various intensive applications.

✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors β€” check the original sources. How BrevFeed works β†’

Reporting from

Amazon has launched EC2 G7 instances featuring NVIDIA RTX PRO 4500 GPUs, enhancing AI inference and graphics performance. This introduces significant improvements over previous generations, with up to 4.6x AI inference performance and 700 Gbps networking throughput, impacting workloads across AI, rendering, and analytics sectors.