MySQL High Availability, 2nd Edition

Chapter 7. Data Sharding

Joel finished reviewing his server logs and noted a few issues with a couple of queries. He made notes in his engineering notebook to watch these queries so he could learn whether they are simply long running or queries that need to be refactored to be more efficient. He was just noting the username for each query when his boss pinged him on the company’s Jabber channel.

“I wish I’d never suggested we adopt instant messaging,” he thought. Joel typed the normal response, “pong,” and waited for his boss to fire off another microtask. While it was nice not to get ambushed in his own office, Joel knew his boss well enough that sooner or later Mr. Summerson would give up using Jabber and return to his usual drive-by tasking routine.

“J, I need u to wrt a WP on shrdin. Do we need? Hlp new acct?”

Joel took a moment to decipher his boss’s propensity to think instant messaging was the same as texting, where typing is more difficult and some services charge by the letter. “OK, so he’s discovered sharding. Now, what is a WP?” Joel typed his response, “Sure, I’ll get right on it. Do you mean white paper?”

“Ack” appeared a moment later.

Joel minimized his Jabber window, opened his browser, entered “sharding mysql white paper pdf” in the search box, and pressed Enter. “If it’s out there, I’ll find it.”

In the previous chapter, you learned how to scale reads by attaching slaves to a master and directing reads to the slaves while writes go to the master. As the load increases, ...