Posted On: Aug 30, 2016
Today we are adding three new features to Amazon Aurora aimed at making Aurora even more performant on a wide range of workloads commonly run by AWS customers. Amazon Aurora is a MySQL-compatible relational database engine that combines the speed and availability of high-end commercial databases with the simplicity and cost-effectiveness of open source databases.
These new features include Parallel Read Ahead, Faster Indexing, and NUMA Awareness. Using Parallel Read Ahead, Aurora performs a parallel pre-fetch of additional relevant pages during a table scan, taking advantage of the replicated storage architecture. This results in range selects, full table scans, the ‘Alter Table’ operation, and index generation performing 5x faster than before. Faster Indexing makes adding indexes and rebuilding tables up to 4x faster. NUMA-Aware scheduling makes CPU-bound operations that make heavy use of the query cache and the buffer cache now run up to 10% faster in large DB instances like db.r3.8xlarge.
You can take advantage of all these new features today by upgrading your instance to version 1.7 of Amazon Aurora from th AWS Management Console. Read more about these new features in the AWS Blog. To learn more about Amazon Aurora please visit our product page.