The Wayback Machine - https://web.archive.org/web/20160409114134/http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/EBSPerformance.html
Menu
Sign In to the Console
Amazon Elastic Compute Cloud
User Guide for Linux Instances

Amazon EBS Volume Performance on Linux Instances

Several factors can affect the performance of Amazon EBS volumes, such as instance configuration, I/O characteristics, workload demand, and storage configuration. After you learn the basics of working with EBS volumes, it's a good idea to look at the I/O performance you require and at your options for increasing EBS performance to meet those requirements.

For information about volume performance on Windows instances, see Amazon EBS Volume Performance on Windows Instances in the Amazon EC2 User Guide for Microsoft Windows Instances.

Contents

Amazon EBS Performance Tips

  • When you consider the performance requirements for your EBS storage application, it is important to start with an EC2 configuration that is optimized for EBS and that can handle the bandwidth that your application storage system requires. For more information, see Amazon EC2 Instance Configuration.

  • When you measure the performance of your EBS volumes, especially with General Purpose SSD and Provisioned IOPS SSD volumes, it is important to understand the units of measure involved and how performance is calculated. For more information, see I/O Characteristics.

  • There is a relationship between the maximum performance of your EBS volumes, the amount of I/O you are driving to them, and the amount of time it takes for each transaction to complete. Each of these factors (performance, I/O, and time) affects the others, and different applications are more sensitive to one factor or another. For more information, see Workload Demand.

  • There is a significant increase in latency when you first access each block of data on a new EBS volume that was restored from a snapshot (General Purpose SSD, Provisioned IOPS SSD, or Magnetic). You can avoid this performance hit by accessing each block in advance; this process is called initialization (formerly known as pre-warming). For more information, see Initializing Amazon EBS Volumes.

  • Some instance types can drive more I/O throughput than what you can provision for a single Amazon EBS volume. You can join multiple General Purpose SSD or Provisioned IOPS SSD volumes together in a RAID 0 configuration to use the available bandwidth for these instances. You can also provide redundancy for your volumes with a RAID 1 (mirrored) configuration. For more information, see RAID Configuration on Linux.

  • You can benchmark your storage and compute configuration to make sure you achieve the level of performance you expect to see before taking your application live. For more information, see Benchmark Volumes.

  • Amazon Web Services provides performance metrics for EBS that you can analyze and view with Amazon CloudWatch and status checks that you can use to monitor the health of your volumes. For more information, see Monitoring the Status of Your Volumes.