scalability.org

First, and most important … my 19th wedding anniversary. Woot! I’m a lucky guy, and I know it!

Second, and very important for the day job front, we’ve been in business 8 years … self bootstrapped, profitable, and growing. This is not to say we don’t need capital, but we are a business first, and making a loss is not something we can sustain for very long, as we fund our operations from our cash flow.

I posted previously about Chris Mellor’s article on Bluearc … it seems that many of the high end storage vendors (apart from EMC and Netapp) are having trouble doing what we do.

I should also point out that, much like last year, this year is absolutely record setting for us. Of that I am happy. But we are fighting the standard battles of smaller companies … getting capital to fund various things is very tough right now. I am learning (more than I ever knew before) about international finance bits … and how to deal with large overseas purchases. There is a whole industry around this. Its interesting … fascinating … and hopefully workable.

Viewed 278 times by 125 viewers

Matt Asay has an interesting article at The Register. In it, he argues that Microsoft needs to adapt to the world that has evolved around it, and do something drastic. This article references a Wall Street Journal article/post on the state of Microsoft and the lack of motion of its share price over the last decade.

In the quoted WSJ article, Matt points to a paragraph that I’ll repeat here:

Even disregarding Microsoft’s bubble valuation when Mr. Ballmer took over in 2000, the stock has been the proverbial dead money for a decade … At bottom, this is a corporate governance problem. Manifestly, the solution is not to let management keep stepping up to the plate with shareholder money and promising home runs that never materialize…

That is, hold them accountable for failure, and more to the point, question the business objectives and assumptions behind various efforts.

One obvious one is HPC. How, exactly, is Microsoft planning on turning its HPC offering into a billion dollar per year revenue source, rather than a loss maker?

There are other obvious ones, such as the Kin, WinCE, etc.

I’ve argued in the past that they are trying too hard to be all things to all people. Going after Google in search, and failing … again, and again, and … Building an Xbox game console, and not making much money in that division. And so on.

Matt argues that they should go long on open source. I agree, but in a different way than I think he opines in this article.

Read the rest of this entry »

Viewed 316 times by 133 viewers

See here. These are in testing, so please report any bugs/errors. mpiBLAST is of course is one of the original cluster-accelerated BLAST implementations, being developed by Wu Feng’s group at VT.

IMO there is a strong need for applications like this, as well as mpihmmer and others. As data set sizes continue to grow at exponential paces, we need tools that can scale to the need. mpiBLAST is definitely in this set of tools … being an enabling technology to perform analyses at scale, that might not be possible without it.

Viewed 377 times by 144 viewers

At The Register, Chris Mellor has an interesting article on Bluearc. In it he notes that they just raised a new series of capital

BlueArc, the hardware-accelerated NAS array supplier startup, has pocketed another $20m in a seventh funding round, taking total funding to around $225m.

Seven rounds. Total capital input is $225M USD. For a VC to be really interested, they need to see some serious multiplicative effects of this investment. Assume that they can exit at 10x valuation … assume that for $20M they sold 50-ish percent of the company. VC’s typically want in the 33-50% region, and the money is expensive. That means that their post money valuation is of the order of $40M or so. So they need to exit around $400M and up to be interesting to the folks who just put in money.

Several things fascinate me about this. First, if we changed “Bluearc” for another set of vendors, we’d have a very similar story.

Second, as Chris points out

The company was started up in 1998 and, 12 years later, still isn’t able to stand on its own feet and make profits.

Yeah, this is an issue. But not simply for them. Other folks in this market have had many many tranches, and aren’t at a self sustaining profitability yet.

Thats whats interesting about this. Does the market support very high cost per gigabyte and high cost per IOP storage anymore? It may have in the 90s and early part of the last decade … but times have changed.

Read the rest of this entry »

Viewed 403 times by 159 viewers

For a while, we’ve used Ubuntu 8.04 as the baseline distribution for DeltaV. In the earlier days, it was easier to get some aspects of the load working, as we had a modern kernel and userspace to work from.

Ubuntu 10.04 has come out, and I am not sure I like it as much. It has some good features, but Canonical has been pushing Ubuntu into some not so great directions as of late, IMO. Not directly relevant for DeltaV, but the nVidia/nouveau driver bits is an example of what I consider ill-advised. Grub2 (which sort of … kind of … works) is another, directly relevant issue.

Basically, I realized our default load was causing us something of a larger set of problems on OS load than I had intended. It was harder to automate than we liked, and I couldn’t share much of the programmatic configuration code between the JackRabbit and DeltaV lines. That is, we wound up spending a great deal more time setting up the DeltaV, which is a simpler system in general.

So we decided to simplify this.

Read the rest of this entry »

Viewed 7393 times by 716 viewers

… died.

Display randomly quits. Just weeks after the warranty expired.

Ugh.

Spec’s for new one: 8+ GB RAM, quad core Intel, Nvidia graphics. So far, Dell has a 4500 workstation that looks good, and HP has a multimedia laptop that looks good. Anyone else I should look at?

Need to run Linux, Windows 7. Mostly Linux. 64 bit.

Long battery life (3+ hours) would be nice, this is what I have today. 1920×1200 display (what I have now) is also nice.

Are Acer/Asus/Alienware good units? Brand name doesn’t matter. Performance, memory do matter.

Viewed 8065 times by 775 viewers

(with apologies to Queen)

Looks like Oracle/IBM have parted ways on Solaris.

None of this bodes well for Solaris market share. If Oracle wants a private OS to run for Oracle’s apps, to compel people to buy its hardware/OS to run, then, well, it might pursue a strategy like this.

Or maybe IBM demanded onerous terms.

Or …

Ok, we don’t know. But that agreement is coming to an end. Which suggests that Oracle isn’t interested in saving it.

What does this say about OpenSolaris? I’d say that the deadline given by the OpenSolaris board a few weeks ago is going to go by without a peep from Oracle.

[update] Looks like this whole exercise was a negotiating ploy by Oracle. Proliants will offer Solaris (not OpenSolaris). Same with Dell.

I think the writing on the wall for OpenSolaris is really easy to read now.

Viewed 8010 times by 786 viewers

Results from our baseline tests of our Delta-V unit is showing a sustained write speed north of 850 MB/s, and a sustained read speed north of 1 GB/s. I compare these numbers to some of our competitiors systems, and note that these are a bit higher than what we have seen reported from them in realistic configurations.

These systems are slated to be iSCSI targets for the customer who bought them. I’ll try to get some iSCSI testing going soon as well. I’d like to see what we can hit performance-wise.

Viewed 8253 times by 783 viewers

spent the weekend working on our DeltaV 4 unit, tuning it a bit. Previous write numbers were a bit lower than I liked.

So we adjusted some of the configuration a little. This is what resulted (old write numbers were in the 450MB/s region for this test)

Read the rest of this entry »

Viewed 8796 times by 807 viewers

… why are we still programming them with serial languages? And more to the point, why are these language compilers so terrible at converting serial code to parallel code? No, seriously … I know there are several constraints on the semantics of the serial language code processing. Debugging and exceptions for one … you wouldn’t want to signal a floating point exception in code that had nothing to do with the FPE in the first place.

But this may be more due to thinking about machines as big serial processing engines, rather than a hierarchically organized collections of parallel and asymmetric processing elements. Most programming languages encourage these thought processes. Parallelism is either an explicit bolt on system, or an intrinsic ‘directive’ driven system.

Neither of these models works well at expressing a parallel algorithm.

Read the rest of this entry »

Viewed 10000 times by 1191 viewers