Posted On: Jan 29, 2021
HAQM Elastic File System (HAQM EFS) now allows you to drive up to 3x higher read throughput on your file system. For example, bursting mode file systems now provide 300MB/s of bursting read throughput, or 300MB/s per TiB of data stored in HAQM EFS standard, whichever is higher. If you have configured 1 GB/s of provisioned throughput, your application can now drive up to 3 GB/s for read operations (and continue to drive 1GB/s for write operations) without any application or configuration changes.
HAQM EFS provides simple, scalable, serverless file storage for use with AWS Cloud services and on-premises resources. The new increased read throughput improves performance of most applications running on HAQM EFS, especially read-heavy workloads like web serving and machine learning (ML), which are commonly deployed in containers using HAQM Elastic Container Service (HAQM ECS) and HAQM Elastic Kubernetes Service (HAQM EKS), and functions using AWS Lambda. This increased read throughput allows real-time applications like ML inference to handle 3x more concurrent requests, and ML model training jobs to complete up to 3x faster.
The throughput increase is automatically enabled for all file systems at no additional cost in all AWS Regions including the AWS GovCloud (US) Regions, the AWS China (Beijing) Region, operated by Sinnet, and the AWS China (Ningxia) Region, operated by NWCD. For more information about HAQM EFS performance, see the HAQM EFS documentation.