c0t0d0s0.org

Bart Smaalders writes inPerformance Anti-Patterns about the best methods to slow down your application. Really an interesting read.

Some of these anti-patterns have their roots in hardware issues, some are the result of poor development or management practices, and some are just common mistakes—but we’ve seen them all repeatedly. In this article we discuss these mistakes: what causes them, how to find them, and how best to avoid them.

The European Commision awarded a huge contract to Sun:

The awarded amount of the market is €95,995,965.67 (approximately $141 million) for all the participating institutions and for its whole period of validity. Of this amount, The European Commission on its own awarded €81,430,488.29 (approximately $120 million).

A nice photo collection of small, big and epic failures - the Failblog

I assume, we will give more informations about Rock at the ISSCC´08 on Monday. In the advance program of the event you will find two sessions about our upcoming processor:

A Third-Generation 65nm 16-Core 32-Thread Plus 32-Scout-Thread CMT SPARC Processor at 01:30 pm

A 2.3GHz 396mm2 16-core 32-thread processor with deep out-of-order retirement and transactional memory is fabricated in 65nm CMOS. The logical and physical design of this high-end microprocessor enable high throughput, high single-thread performance, mainframe-class reliability, availability and serviceability (RAS), hardware transactional memory, and linear scalability.

Implementation of a Third-Generation 16-Core 32-Thread CMT SPARC
Processor at 02:00 pm

A 65nm third-generation chip-multithreading SPARC® processor operates at 2.3GHz and is targeted for high-performance servers. It is optimized for both single- and multi-threaded applications. Circuit innovations in memory arrays, register files, and floating-point hardware boost performance and circuit robustness with minimum area overhead.

(für Frau D.)

Network Appliannce works really hard on the title "major annoyment": NetApp posts a benchmark for an EMC Clariion device. This will spark some heated discussions in the next few days

I´ve read a review for the new Samsung 1TB. The review itself isn´t such an interesting piece of information. It´s a little fact, i didn´t thought of before albeit it´s obvious:

An attempt at reformatting the Spinpoint F1 still required 37 hours, but switching to a Windows XP MCE system saw the usual three and a half hours of formatting time emerge.

I assume, i´m really spoiled by a filesystem that comes back within seconds after adding it to a pool instead of waiting 3,5 hours. But this leads to an interesting observation: With the increasing capacity of harddisk and disks in the size of RAID arrays only a few years old, you need different concepts to integrate the storage in your system. Think about a 2 TB harddisk! Waiting 7 hours before you can use it?

IBM announced new Power6 based systems today. The p520 Express and the p550 Express. This announcement is somewhat hidden in the PowerVM announcement.Well, now legions of IBM sales reps will go to customers and tell them: "We have Power6 in smaller systems. Not only in p570". But you have to take this comment with a really large amount of salt.

When you look into the data sheets you find some interesting facts. You won´t get 4.7 Ghz p6 with the new systems. The p520 is a 4.2 Ghz only system and the p550 is only available with 3.5 or 4.2 Ghz. So still no mass-rollout of the 4.7 GHz p6 silicon. IBM must still have yield problems with the high-end version.

More interesting is the fact, that the p520 doesn´t have the 32 MB L3 Cache of the larger boxes. I wrote about the effect of having less cache on the Power6 chip some weeks ago when the new blade was announced. Like with the blade: You get the p6 proc but not the p6 performance.

Diesmal in der Berliner Umschau:



Oder auch Capital(Dank an Alex):

Sebastien posted a walk through a really nice feature that found it´s way to the Opensolaris build 83: vanity names for network interfaces.

Now you can give any name you want to a network interface. For example: You have four e1000g interfaces in your system. Without this feature they would be named e1000g0 to e1000g4 by Solaris. Now you name them eth0 to eth1 with some simple commands:

   dladm rename-link e1000g0 eth2

But why should you use the cryptic linux naming scheme? We can do better than that!

   dladm rename-link e1000g0 internet0
   dladm rename-link e1000g1 dmz0
   dladm rename-link e1000g2 private0

After all this offerings from Oracle, Sun and so on, IBM had something to do about virtualisation: And IBM does what they do everytime in this case ... bundling old products into something new. For example their new PowerVM offering. It´s just the virtualisation of the Power processors bundled with some code from Transitive to execute x86 code on a Power CPU. I´m not really convinced about such a bundle. To do virtualisation for x86 on Power looks like a marketing stunt for me. Why should i buy an expensive Power system to virtualize Linux x86 systems? Even at 2000$ socket for VMware it´s vastly cheaper to buy a x86 server in the first place. And this doesn´t take support from ISV into consideration. But at the end: You can say everything about IBM, but you can´t say that they have a bad marketing department.

PS: By the way ... LDOMS or Containers on Solaris goes for $0 ... beat this, IBM