2. The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.
26. What Facebook is saying I previously tested InnoDB on an 8-core server to determine how many IOPs it can do for a simple IO-bound workload. The limits were ~12k disk reads/second for MySQL 5.0 and ~18k reads/second for MySQL 5.1. I just repeated the tests using a 16-core server and the results are much better. I can get 20,000 to 30,000 disk reads/second using InnoDB 5.1. InnoDB 5.5 appears to be capable of 50,000 to 70,000 disk reads/second courtesy of support for multiple buffer pools (innodb_buffer_pool_instances). Mark Callaghan’s Blog, Facebook – June 21, 2010 http://www.facebook.com/note.php?note_id=403975340932
27. What Facebook is saying http://www.facebook.com/note.php?note_id=403975340932
49. The preceding is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.
At 1024 Connections: - 188% performance gain for MySQL 5.5 over 5.1.40 (InnoDB 1.0) - 200% performance gain for MySQL 5.5 over 5.1.40 (InnoDB built in)
At 1024 Connections: 164% performance gain for MySQL 5.5 over 5.1.40 (InnoDB 1.0) - 365% performance gain for MySQL 5.5 over 5.1.40 (InnoDB built in)
Give examples of concrete applications looking at different 9’s Real life application Order management, ecommerce.
In-Memory Database snapshot to disk non-indexed columns to disk Shared-Nothing Storage Synchronous Fast-Automatic Fail Over High Performance NDB API available No Special Hardware Requirements*