2008 Year in Review

Posted on December 30, 2008

Here it is, big post 1,000. I’m fairly proud of that given that the vast bulk of all my blog entries are technical and not just brainless linkdumps. There is still a lot to blog about and I’ve still written a great many entries that ended with “more to come…”, never the less its a good milestone.

Looking back at 2008, we’ve had a very good a productive year in OpenSolaris land. COMSTAR arrived, Crossbow arrived, ZFS is getting stronger all the time, we got a new iSCSI Target, the first and second release in the 6 month cycle of Indiana went out on schedule, and Solaris 10 is now more or less on par with Nevada. Technically there is a lot to be proud of and excited about.

On the non-technical side we had another OpenSolaris Developers Summit and the first annual OpenSolaris Storage Summit. Ian Murdock gave a keynote at CommunityOne and there was a heavy emphasis on OpenSolaris at JavaOne. We did several good conferences this year, although not as many as in years prior. We had a dominant year at SNIA’s Developers Conference, helping solidify Sun’s role in the future of storage development.

On the Sun side, the mighty FISHworks released to the world and the response to the resulting offerings has been tremendous thus far and sets a new standard in storage particularly in the realm of the Sun-created buzzword “OpenStorage”. Business for Sun is poor but there are several areas of growth and although I think the MySQL acquisition was a massive blunder it may all pan out in the end.

On the OpenSolaris governance side, its been a sad year. Rather than moving forward the OGB decided to rehash old ground and fall right back into the same pitfalls. An all Sun OGB proved to be less effective than a mixed OGB. OpenSolaris governance in general is more closed off and insular than ever, but thats indirectly what Simon Phipps and others were shooting for.

The Silicon Valley OpenSolaris Users Group fell into significant decline over previous years, but tends to be a valley trend as technologies loose their initial buzz and become more established… the Silicon Valley Linux Users Group felt the same kind of declines, although not as sharply.

As we look to 2009, I think the word is “established”. OpenSolaris is here, Nevada is strong, we’ve proven that its not going to disappear. We now need to set the tone for the future by definitively establishing the future of Solaris 11 (or lack of one), upgrade path from Solaris 10 (if there is one beyond HP-UX like Update-forever), and wrapping extension technologies like xVM, Sun Cluster, and others around OpenSolaris. In general, customers are still largely unclear on where this is all ultimately going and what it means to them. If you have big SPARC box like Sun Fire E2900 in production running Sun Cluster, what does the future hold? S10 till you retire it? OpenSolaris makes a lot of sense to new adopting customers, but then a lot of them are running it on non-Sun hardware (Dell, HP, and Supermicro are popular)… how do we monetize them in a compelling way? And how do we continue to ramp Sun support of Nevada? To date most experiences with Sun Support over post-S10 releases are horrible as a lot of Sun’s Support organization simply doesn’t know it well enough.

So, the pave stones are on the ground, they now need to be shifted into a resting position so we can start walking people across the path. Its time to unify offerings and improve Sun’s sales, marketing and support around it.

Here’s to 2009!