thinking sysadmin

There’s been a lot of noise from the storage industry about flash recently – in particular, noise from EMC and Sun, both of whom recently announced storage products using flash, EMC in January and Sun earlier this month. Below are my thoughts on what EMC and Sun are doing, as well as what NetApp might do. Since I see a fair amount of visitors from all three companies here, if I’ve got something about your employer wrong, please correct me in the comments.
Read the rest of this entry »

• Jonathan’s Blog: Anything But a Flash in the Pan – Jonathan Schwartz on Sun’s soon-to-be-released (late this year) flash drives. Quote: “ZFS will transparently incorporate Flash into the storage hierarchy of a running system, using the microprocessor cache for the most performance sensitive tasks, DRAM for the next, then Flash, then disk (then ultimately tape).” Speaking of tiered storage, I wonder how flash drives would work within a SAM-QFS implementation – and if SAM-QFS is destined to wither in the shadow of ZFS: Sure it’s open source now, but where is it going these days?

• Performance Isolation of a Misbehaving Virtual Machine with Xen, VMware and Solaris Containers – A paper comparing the ability of VMware (Workstation), Xen and Solaris Containers to isolate virtual machines from each other with regards to performance. Quick summary: VMware and Xen go one-two in protecting one virtual machine from another, and Solaris Zones lag far behind, a definite trade-off to consider if you’re inclined to use zones for their higher density. I would expect similar results to what was reported with Solaris if the authors had also looked at OpenVZ and FreeBSD Jails – in fact, (anecdotally) I’ve seen generally the same behavior with OpenVZ and Jails, although OpenVZ does give you a lot of knobs to adjust (if performance isolation is a concern for you with OpenVZ, you may want to look at all the options it gives you before making decisions about its suitability). It would have been nice to see more details about the Solaris configuration used in the paper, and to have had the work repeated with a recent OpenSolaris build – substantial work appears to be being done in this area (for example, Improved Resource Management and Zones Integration). Seen at blog.scottlowe.org.

I missed this: Apparently the OpenSolaris 2008.05 AMI on EC2 has reached a capacity limit. So, while I got Sun’s approval this morning for their OpenSolaris/SXCE.79 on EC2 beta, I don’t get to use OpenSolaris until they add more capacity; SXCE.79 is a nice consolation prize, though.

Generally, I’m a fan of Sun Microsystems. For the most part, I like their hardware and their software – and their best products show real innovation and willingness to take risks. I’m also a fan of Amazon’s EC2 product, so the announcement that Sun would be officially bundling OpenSolaris for EC2 was great news. Unfortunately, it seems that after all the hullabaloo, Sun doesn’t really want to make it that easy for you to actually use OpenSolaris on EC2, by managing access to it like a control freak would.
Read the rest of this entry »