Along with having an awesome title for a Solaris-oriented blog, Joerg Moellenkamp over at c0t0d0s0.com has been covering Sun’s open-source ZFS super-filesystem for quite a while. Last week he posted about the Adjustable Replacement Cache (ARC) technology, and today’s post covers synchronous vs. asynchronous writes. But Joerg has been writing about this stuff for a while – check out his piece on MAID (Massive Arrays of Idle Disk) from back in August, for example.
We in the storage and server fields (not to mention Apple fans) have been watching ZFS with keen interest for a while. It holds great promise as the next-generation UNIX filesystem foundation, if all of the legal and licensing issues can get worked out. I personally spent 15 minutes during each of my storage virtualization seminars last year touting all of the great things ZFS might do for us once it’s rolled out. We’re starting to see it appearing in storage arrays like those from greenBytes now, and Apple will enable it in Snow Leopard Server later this year. Exciting stuff!