New Amazon Elastic Compute Cloud (Amazon EC2) instances powered by specialized 4th Gen Intel Xeon Scalable processors are now generally available, according to an announcement made today by Amazon Web Services (AWS). This introduction is the most recent in a long line of 4th Gen Xeon-powered instances that power critical workloads like AI, databases, networking, and corporate applications while offering the lowest total cost of ownership (TCO) and the greatest number of integrated accelerators of any CPU.
“Intel collaborated closely with AWS to provide its cloud clients our feature-rich 4th Gen Xeon processors; many of them have benefited from its performance and worth for months in private and public preview. We are pleased to offer the same practical value to cloud customers all over the world today.
Through AWS’s extensive worldwide reach, these new Amazon EC2 instances—the M7i and M7i—bring 4th Gen Xeon accelerator engines to the public. Customers with increasing need for AI workloads now have a much-needed option in the form of built-in accelerators like Intel Advanced Matrix Extensions (Intel® AMX). Inference performance metrics for large language models (LLMs) with fewer than 20 billion parameters can also be reached by 4th Gen Xeon with AMX, making LLMs cost-effective and viable for use with general-purpose infrastructure.
In general, Intel and AWS provide clients with a range of product choices that best match their workload, performance, and cost requirements, as well as the simplicity and flexibility to meet future demands for digital infrastructure.
The following AWS Regions are home to the M7i-flex and M7i instances: US East (Ohio), US East (N. Virginia), US West (Oregon), and Europe (Ireland).
The M7i-flex cases provide
For the bulk of general-purpose workloads, this is the simplest approach for clients to obtain price-performance advantages.
Web and application servers, virtual desktops, batch processing, micro services, databases, and business applications are just a few examples of designs for smoothly running general-purpose workloads.
Better price performance by up to 19%1 as compared to M6i instances.
What the M7i instances provide
Price performance advantages for important workloads including video streaming, gaming servers, big databases and application servers, and CPU-based machine learning.
additional instance sizes (up to 192 vCPUs and 768 GiB RAM).
The performance of workloads including databases, encryption, and compression, as well as queue management, is optimized thanks to new built-in accelerators that make it possible to offload and accelerate data processes effectively.
[…] later this year. The company is constantly testing fresh Macs that will be upgraded with the new processor. According to the latest reports, Apple is testing a new version of the Mac mini with the new M3 […]
[…] customers in effectively optimizing and managing their Databricks workspaces, Intel has created Cloud Optimization Modules for Databricks. These modules offer pre-configured resources and optimizations […]
[…] to IBM Power Virtual Server when a North American retailer wanted to move their applications to the cloud and benefit from a more adaptable and responsive […]
[…] isolated recovery environment in the cloud works in conjunction with disaster recovery in a number of ways, […]
[…] Cloud HPC Toolkit to AI and ML […]
[…] which facilitate the deployment, management, and security of applications across various cloud environments, stand out among these […]