Looking inside Sun's results

I do not own any significant Sun gear at all, other than in my archive storage system, so I think I’m pretty unbiased when I say that one thing I really appreciate about them is the fact that so much of the company blogs. It’s handy. You might like their stuff, you might not, but the bottom line is that the company is pretty accessible.

Marc Hamilton’s blog has a few lines this week on Sun’s earnings, and in particular their server performance

Sun logoBeing responsible for a large portion of Sun’s server sales, I naturally gravitate to the page 6 Billings Disclosure By Category chart where nine quarters of billings for Solaris SPARC Enterprise Servers, Solaris SPARC CMT Servers, and x64 Servers are detailed. Lets start with our x64 servers, which customers use to run applications on Solaris, Linux, and Windows. Q209 our x64 server billings grew 11% year over year. In fact, if you read the footnote, X64 Servers includes X64 Rack, X64 Blades and X64 Netra products and excludes products categorized under Open Storage and note that Open Storage, which includes products like our x4540 storage server, grew at 21%, the growth was even higher. Not bad when some of our competitors saw greater than 30% decline last quarter in their x64 servers. So why are Sun x64 sales growing when others using the same Intel and AMD CPUs declining? In fact, in Q2, we saw the second highest growth of any quarter going back to Q4FY07.

Marc politely complains in his post about the kind of coverage that results often get

The information content in the presentation is much richer than the many press headlines which seem to be more impacted by hourly stock swings than detailed results.

This fair. On the one hand an article’s headline is supposed to summarize, in a few words, the nut of the story. If you only have a few words to say what’s happening with Sun’s earnings, then some of those words should point to the fact that they lost money. On the other hand, I think that real news organizations owe it to their readers to dig in a little bit and see if there is more to the story than what lies on the surface (note that I don’t put insideHPC in this group most of the time — this site usually summarizes what others have said; but HPCwire is in that other boat, and does a good job of digging in).

Early on I decided I wanted to avoid the likelihood of introducing biases in earnings headlines based on my own personal likes and dislikes for particular companies. So an insideHPC earnings headline always follows the same format:

Company announces Q*; posts {loss, profit}

This does make for the kind of headline that Marc is compaining about, but on the plus side it summarizes the nut of the story and prevents me from introducing bias into news about a company’s financial performance. I don’t mind being biased about other things, but financial news seems more important (somehow).