MySQL Forums
Forum List  »  Utilities

Re: Fabric project status
Posted by: Rick James
Date: October 07, 2016 12:54PM

(I don't work for Oracle; the opinions are my own.)

For HA ("High Availability"), Fabric was having trouble avoiding a Single Point of Failure (the "Fabric node"). Multi-master provides a solid solution to this.

You are read heavy? Hang lots of slaves of the 3 masters. Have you actually determined that you need more than one read server, and cannot stretch things via index improvements, summary tables, selective caching, etc? Summary tables often provide 10x improvement in DW applications.

Options: ReplyQuote


Subject
Views
Written By
Posted
1082
October 02, 2016 03:52AM
415
October 03, 2016 09:09AM
438
October 05, 2016 12:44PM
448
October 06, 2016 06:51PM
376
November 23, 2016 11:30AM
321
November 30, 2016 01:04PM
Re: Fabric project status
363
October 07, 2016 12:54PM
313
November 23, 2016 12:20PM


Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.