scalability.org

This is more economic than HPC related, but it has a relationship to HPC as it turns out.

Cash for clunkers was designed to replace lower MPG cars with higher MPG cars, by offering effectively a discount for purchase. The government allocated $1B to the program. It started Monday. As of today, Friday, the money is gone.

Inventory was moved, it is no longer aging, Higher MPG cars are now on the roads.

I have one. I finally … finally got rid of my Jeep Grand Cherokee. 13 years I had that car. I really liked it. But it was showing signs of aggressively falling apart. So I traded in for a car that gets 13 MPG better gas mileage.

(more…)

Viewed 22400 times by 1659 viewers

Our 7-year business anniversary (from when we started) is in 2 days. We launched 1-August-2002, at the height of the internet bubble collapse. I took a golden handshake from my employer at the time and formed Scalable. Took the risk I had always been afraid of.

7 years ago.

Amazing.

Growing and profitable six out of the seven years since then.

Yes, we may have a 7th anniversary sale. Contact us if you want details, or if you want to take advantage of it early.

And … we had been thinking about a specific product for a while. A very interesting product. We have had 6-ish requests for exactly this in the last 2 weeks. Working out the details, you will hear stuff soon from us/resellers.

Viewed 21982 times by 1620 viewers

Here.

Viewed 20971 times by 1532 viewers

I have never had as many problems directly caused by one application, across so many machines, across so many distributions, as NetworkManager.

For those who don’t know, NetworkManager is your friendly helper application (mistakenly) adopted by distros to handle setting up networks.

This would be well and good if it, I dunno, actually worked?

I won’t recount my long painful history with it. Suffice it to say, everywhere I see it … everywhere … I immediately replace it with wicd. There is a very good reason why I do this. Wicd just works. Never any problem with it. Wicd does what NetworkManager says it does.

NetworkManager is also not supposed to show up on servers. Its a bad enough tool on its own … on a server? Its a disaster.

So, I found it strange that on the last update for the OS on our central server (running the server version of the distro), NetworkManager reared its ugly head. Connections were going up and down, in standard NetworkManager style. Enough to drive a person nuts.

Ugh. NetworkManager. Just say no.

Viewed 18883 times by 1545 viewers

In this past weeks HPCwire podcast, Chris Willard and Michael Feldman discuss many things.

The business side of HPC, the future of companies, etc. I agreed with everything they said (having said it here in these pages in the past).

That is, until the last minute. Thats where what they said doesn’t quite mesh with what we observe and are experiencing.

Specifically, they suggesting that in these tough times, end users are being more conservative, sticking with the large vendors, and eschewing the smaller vendors. They appear to base this analysis on the picking up of Onstor by LSI and Ibrix by HP.

With all due respect to Chris and Michael, I disagree with their view. We see something very different in the HPC storage market. We see users needing the same if not better performance, with much smaller budgets than in the past. Moreover, when we speak with these users, we hear how they wanted to buy X, but they just couldn’t afford it.

(more…)

Viewed 18388 times by 1588 viewers

I am seeing lots of deep soul searching in pundit circles, as well as head scratching on the part of customers, as various vendors writhe and contort in their death throes. Pundits regularly trash that which they neither grasp, nor prefer. Customers wonder what the right path going forward is. Vendors struggle to figure out what the market really wants, and to be able to offer that (all the while the marketing teams are spinning hard and fast).

(more…)

Viewed 17296 times by 1568 viewers

Been running into a bit of this recently. Its usually preceded or followed with some sort of performance requirement, that ‘X’ just can’t hit, or they would need so much of ‘X’ that it blows their budget up.

I find it interesting that the IT folks, the ones really worried about their futures due to budget cuts placing pressure on them to do more while spending less, really get our message, and grok what we do, and how we can help them achieve their mission goals while reducing their costs.

Even more to the point, with the sale of Sun approved, and the impending … er … changes … to the Oracle “hardware” product line, worries about bricking are driving their customers to seek alternatives. We provide an excellent alternative for high performance high density storage, lower priced, higher performance, and complete freedom from bricking. Those folks get it.

But we still do run into the (dwindling) few who insist on ‘X’ to “reduce risk”. Um. Yeah. Reduce risk.

Risk is that of bricking. Not of dealing with a small vendor. Large vendors blow up just as easily ( Sun and SGI to name a few). And your brand new shiny proprietary system is now a very expensive brick after your proprietary vendor falls over. I hate to use them as the example, but SiCortex is an example of this. Lots of new customers, right before they were turned off. But those customers are SOL for replacement parts. It sucks if you are a customer of such a proprietary system, and the vendor goes bust.

That is risk. If your machines are supportable even if the company goes away? Yeah, that is risk reduction. If you have to hoard spare parts and buy used machines to keep parts available? No … that is risk you are attempting to ameliorate.

We have had multiple groups call us up asking if we have any thumpers or thors turned in for JackRabbits, as they want to buy them off of us.

Which gives me an idea …

Viewed 18397 times by 1566 viewers

This was a long time coming. The previous management, prior to them sinking in april 2009, nor the management teams before that … going back at least 10 years, would never have done this.

Its a shame. It should have happened long long ago.

(more…)

Viewed 16942 times by 1544 viewers

For those benchmark pr0n viewers … here is Scalable Informatics Inc. JackRabbit JR5 unit with 48 drives.

Simple benchmark. How long does it take you to write 1TB.

How about 524 seconds?

(more…)

Viewed 15115 times by 1449 viewers

It wasn’t enough to impact results, but it was enough to cause questioning my results (and sanity).

Part of the IO operation is having N processes write to 1 file. To make this happen correctly, each process has to compute their offset into the file, and start operations from there.

There is a seek involved. Now if I am smart, it won’t be

lseek(file_descriptor,0,SEEK_SET);

Nope … that would be wrong (the zero).

Having the line like this would mean that 128 GB sized files … might not look like 256 GB sized files.

[root@jr5 ~]# ls -alF /data/file
-rwx------ 1 root root 274877906944 Jul 21 20:43 /data/file*

grrrr….

Viewed 13728 times by 1389 viewers