HomeCloud ComputingAmazon Elastic File System Replace – Sub-Millisecond Learn Latency

Amazon Elastic File System Replace – Sub-Millisecond Learn Latency

Amazon Elastic File System (Amazon EFS) was introduced in early 2015 and have become typically accessible in 2016. We launched EFS to be able to make it simpler so that you can construct functions that want shared entry to file knowledge. EFS is (and all the time has been) easy and serverless: you merely create a file system, connect it to any variety of EC2 cases, Lambda features, or containers, and go about your work. EFS is very sturdy and scalable, and provides you a powerful read-after-write consistency mannequin.

For the reason that 2016 launch we’ve got added many new options and capabilities together with encryption knowledge at relaxation and in transit, an Rare Entry storage class, and a number of other different decrease price storage courses. We’ve additionally labored to enhance efficiency, delivering a 400% enhance in learn operations per second, a 100% enhance in per-client throughput, after which an extra tripling of learn throughput.

Our clients use EFS file methods to help many alternative functions and use circumstances together with residence directories, construct farms, content material administration (WordPress and Drupal), DevOps (Git, GitLab, Jenkins, and Artifactory), and machine studying inference, to call a couple of of every.

Sub-Millisecond Learn Latency
Sooner is all the time higher, and right this moment I’m thrilled to have the ability to let you know that your latency-sensitive EFS workloads can now run about twice as quick as earlier than!

Up till right this moment, EFS latency for learn operations (each knowledge and metadata) was usually within the low single-digit milliseconds. Efficient right this moment, new and present EFS file methods now present common latency as little as 600 microseconds for almost all of learn operations on knowledge and metadata.

This efficiency enhance applies to One Zone and Normal Basic Function EFS file methods. New or outdated, you’ll nonetheless get the identical availability, sturdiness, scalability, and robust read-after-write consistency that you’ve got come to count on from EFS, at no extra price and with no configuration adjustments.

We “flipped the swap” and enabled this efficiency enhance for all present EFS Basic Function mode file methods over the course of the previous few weeks, so chances are you’ll have already got observed the development. After all, any new file methods that you simply create can even profit.

Study Extra
To study extra in regards to the efficiency traits of EFS, learn Amazon EFS Efficiency.


PS – Our multi-year roadmap comprises a bunch of short-term and long-term efficiency enhancements, so keep tuned for extra excellent news!



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments