c0t0d0s0.org

I had an interesting discussion with a blog reader and customer via icq (oh wonderful new world of business). It started with a question about the J4500 and it ended in the the discussion, if MAID is really a viable way for storing data of if it´s just one of this hypes working their way through the storage business every year.

MAID? What is MAID? MAIDs are Massive Arrays of Idle Disk. The idea is to save power by switching of disks in the case they aren´t in use. You can buy such components from companies like Copan.

There is a new resident in my home office. I´m jumpstarting fangorn at the moment. fangorn is a Sun Ultra 60 workstation (dual-proc UltraSPARC II system with 450 MHz on each proc and 1,5 GB memory). The Ultra60 line had it´s GA in 1998. This was shortly after my first contact with Sun systems. I purchased it for a few euros from a colleague to rescue it from his cellar and to have a multi-proc SPARC system for some testing. I think, fangorn is a good name for such an venerable system.

Sun and Fujitsu have started to deliver systems with the 4 core SPARC64 VII. But Fujitsu already talked about the next generation at the HotChips conference: An 8 core processor in development and they hope to talk about this a little more at the HotChips next year. 8 core SPARC64 would lead to a 1024 threads M9000 ... 4 Teraflops in a single operating system image.

Relatively short after the crash there were rumours about deployed thrust reversers at one of the engines of the Spanair aircraft and shortly after this you were able to find a photo (login needed) of this engine at the usual discussion areas. Today two spanish newspaper reported according to Aviationweek and the german newssite sueddeutsche.de (based on dpa) that the thrust reverser of one engine were found in the deployed position.

Albeit this would explain some of the reported reactions of the plane, it´s much to early to draw any conclusions. The question is now: Were the reversers in this position before the crash or were they moved in this position by the crash. We have to wait for the final accident report ...

PS: Nevertheless this evoked memories to the LaudaAir 004 crash in me: Mid-air deployment of the reversers in 24,700 feet. Nobody in the aircraft survived the crash.

I´ve uploaded another early pdf of the tutorial. I´ve added some text and the first to figures. Kerberos is just one of the topics where you should explain only the configuration or you have to explain it almost completly. You can´t explain just a few things and still hope your explanation makes sense.

Kris zeigt in Mein privates Datawarehouse - Sparen mit MySQL wie man mit der CSV-Export-Funktion des Sparkassen-Homebankings, Mysql und einigen Gedanken sich ein eigenes Data Warehouse bauen kann, um die Geldsenken im eigenen Budget zu finden.

Reichen 4 GB fuer eine Seele? Hängt vielleicht von den darin befindlichen Abgründen ab. Und wie gut lässt sich eine Seele komprimieren? Und wie oft muesste man Diffs sichern?

In the Webshop of Dell you can choose Solaris x86 as an option. This is an example from the Dell PowerEdge R905:


But it costs $231 more than the subscription at Sun

The ARC has approved the PSARC/2008/427 iSCSI Boot case.

This project is to enable Solaris to boot off iSCSI luns via regular network adapters. Different approaches, iBFT/OBP, are adopted to implement this feature on x86/sparc platforms.

For the readers unaware of iBFT: This is a part of the ACPI 3.0 specification to enable a kernel to gather additional data needed to boot via iSCSI.

Jonathan Rauch writes in Caring for Your Introvert:

Do you know someone who needs hours alone every day? Who loves quiet conversations about feelings or ideas, and can give a dynamite presentation to a big audience, but seems awkward in groups and maladroit at small talk? Who has to be dragged to parties and then needs the rest of the day to recuperate? Who growls or scowls or grunts or winces when accosted with pleasantries by people who are just trying to be nice?

I´m feeling observed. I can do a lessig-style presentation in front of 200 people without any problems but in a bar i´m really quiet, i hate senseless smalltalking like hell. A text catching my thoughts at the moment.

An dieser Stelle möchte ich meine Glückwünsche an meine Schwester Solveig und meinen Bruder Jens richten. Solveig hat heute ihren Bachelor in Frühpädagogik mit hervorragender Benotung geschaft. Jens hat am Donnerstag sein Diplom als Wirtschaftsingenieur mit ebenfalls sehr guter Benotung erlangt . Euer grosser Bruder ist stolz auf euch

Hothardware.com reports about some internals of Intels upcoming Nehalem architecture. Sounds interesting but i´m sure this turbo mode stuff will lead to many single thread performance benchmarks without practical meaning. Sound a little bit like this benchmark where a well known other chip company deactivated all cores except one, but not the respective cache. Thus they were able to execture large portions of the benchmark in the processor cache. Downclocking all cores except one and upclocking this one sounds like a similiar "benchmark-special".

I should cook tea the whole day. I had an idea what we could do with all this threads in a 4 socket Victoria Falls system while waiting for the water to boil: Let´s build a poor mans fault tolerant system out of it. It´s an unrefined idea ...

Let´s assume you have to control something: For example a valve in a chemical plant. It´s an important valve, thus you can´t afford a fault (for example by cosmic rays flipping a bit). Thus it´s a common method to compute something multiple times and compare the result. Two computations isn´t enough. When the results differ, you can´t tell what result is the right one. Or better: What´s the result with the highes probability to be correct. The practice suggests to compute it three times and to compare it afterwarts. The result with at least two votes of three in this quorum wins.

You could implement every decision making instance in an LDOM. Thus you could even implement different operating system patch levels in each of the systems.

And this fits with the four procs of a fully blown Victoria Falls system . Three procs for computations and one procs for the comparator. To ensure that every process runs on the same core it was tested, you could bind it to a processor.

This would reduce the thread count to 64 effective processors, but you have implemented a poor-mans fault-tolerant system (poor man because of no hardware lockstepping, it would be just application lockstepping) and well, you have more than enough hardware threads in a Victoria Falls system ...

I will think a little bit more about it in the next few days....