Posted On: Nov 24, 2020

Customers can now deploy compute-intensive applications such as machine learning inferencing, multimedia processing, scientific simulations, HPC, and financial modeling that leverage Advanced Vector Extensions 2 (AVX2) to meet their performance requirements on AWS Lambda.

AVX2 allows CPU to perform a higher number of integer and floating-point operations per clock cycle, thus speeding up CPU intensive workloads with parallelizable operations, allowing customers to process data faster, improve throughput, schedule data-intensive jobs, and deliver performant end-user experiences. Starting today, customers can either compile their existing workloads or deploy new ones to target this instruction set at no additional cost. To learn more on how to build AVX2 compatible applications on AWS Lambda, read the Lambda Developer Guide and the AWS Compute Blog.

Support for AVX2 is available in all regions where AWS Lambda is available, with the exception of the AWS Regions in China. For more information on availability, please see the AWS Region table.