Recently on the ZFS mailing list http://wiki.illumos.org/display/illumos/illumos+Mailing+Lists, there was some discussion about how ZFS distributes data across disks. I thought I might show some work I've done to better understand this.
I am currently working on an online course that is aimed at Linux system administrators and developers who want to use SmartOS.
A ZFS record size is variable: by default it's 128K, but write 2KB of data, then only 2KB writes to disk. What does record size actually enforce?
This week, Mr. Bruning investigates memory leaks.
Carlos Cardenas has been working on porting a program from Linux that is CPU bound. He found some hotspots using DTrace. He asked someone what else he could do to maximize performance and was asked in turn if the program was cache friendly. Carlos went on to ask if I could provide some insight into the cpc provider in DTrace. I'll do that, concentrating on examining cache utilization.
For the past few weeks, I have been studying performance of an application that uses MongoDB, Node.js, and jmeter. The application was written by a third party, and they set up the application so that each component runs in its own Smart Machine.
I have been thinking about this one for a while, because every now and again, mostly for testing purposes I find that I really wish I could reset a kstat. I have been able to figure out how to reset certain kstat(s) with MDB, but wondering if perhaps you have one method that makes it relatively easy to find and reset any kstat.
Mr. Bruning answers the question, "what happens when the memory being used by a zone exceeds the cap?"
Max Bruning answers a question from Twitter about a DTrace script.
Weekly column from Mister Max Bruning @MrBruning answering your burning SmartOS, DTrace, and mdb questions.
First column from Mister Max Bruning @MrBruning answering your burning questions on SmartOS, DTrace, and mdb.