c0t0d0s0.org

But at first: Dear Readers,

this article is not for you, it's a text i link in my social network profiles in order to explain some stuff to headhunters. So you can safely ignore it.

Kind regards
Joerg

Dear Headhunters,

always nice to hear from you. Because ... who knows, what you will offer to me. Perhaps a job flying around the world, talking about the deeper inner workings of technology i love to talk about and helping people using it (except the around the world part, it's the stuff i'm doing today). So i'm finding several of your mails in my Inbox, at Xing or LinkedIn and out of curiosity i don't delete them directly. While some are somewhat interesting, the kind of mails i get at and increasing number at the moment is just different from spam by a minuscule margin ... just a quantum leap above spam... in the real physical sense, not the in the way it's used incorrectly today. In the last weeks i got way too many request from you for positions that aren’t interesting to me at all. The sad thing about it: It's avoidable.

Please understand, that the role of the Principal Sales Consulant is technical position at Oracle, in my case a very technical one. I’m the guy technical presales ask when they have very technical questions and don’t want to go to engineering with their question or they want me to go to customer to explain for example inner workings of Solaris for example. I'm at least that good at that job, that someone thought it would be a good idea to put me in the Oracle Elite Engineering Exchange program. And i love to be at this interface between Engineering and Sales. This is something you could see by studying the profile at Xing or LinkedIn. To say this differently: I think there are really not many Sales Reps or Key Account Managers that can do performance tuning on Solaris. On the other side, tuning Solaris or explaining the inner workings of the process scheduling isn't a talent a good Sales Rep could use or need.

So just because you fid me with SELECT * FROM SocialNetwork WHERE POSITION LIKE '%Sales%' this shouldn’t lead you to the conclusion that i want to do a Sales Job. So … please don't offer positions for Sales Manager, Key Account Manager, Sales Rep or something like that to me. You can safe your time by skipping me at such positions. If this ever changes, i will change this blog entry. When you want to lure me away from Oracle, positions with "%Engineer%" or "%Consultant%" have at least a chance to get past the initial check.

However: The situation about this leads me to the conclusion that you did a pretty bad job at finding people for your clients as „I’m a technology guy“ is pretty much written on the wall in Helvetica bold at several thousand points and it really shouts that you just searched for above pattern and wrote all hits on the query without further checking. You should check if an offering looks even remotely interesting for a person, it's not the job of the people you are writing too. Or to use a different analogy: You should hunt heads as a sniper, not as a machine gunner.

Best regards
Joerg Mollenkamp

Sometimes i would like to be able to do the jedi mind trick. Like "This isn't the performance issue you are searching for". Because it would be easier to convince people in that case that they saw the effect, but not the cause. That said, I think the real root cause on what i will describe in this block of this problem is the separation of database admins and system admins. But that's my personal opinion.. Imagine from one day to another, your storage goes mad. The analytic tools of your storage show long latencies, the led of the storage just show massive use of the storage. As you didn't made any change your first assumption is "storage has a defect" or "i just hit a bug in the storage firmware". At the end he or she thinks "No major release changes. Perhaps a few minor configuration changes. And none of them has to do with storage".

(As the events are in german language, i will proceed in this blog entry in german language as well. Sorry.)

So ... eine Reihe von Terminen für die Solaris 11.1 Security Vorträge, die ich im Rahmen der "Oracle Business Breakfast"-Serie halten werde stehen nunmehr fest:

• 10.9. Hamburg
• 13.9. Potsdam
• 18.9. Düsseldorf
• 20.9. München

Die Anmeldelinks werde ich rumschicken, sobald diese verfügbar sind.

PS: Der Vortrag "Solaris 11.1 Security - Was gibt es neues?" im Rahmen der DOAG Konferenz in Nürnberg ist eine gekürzte und mit einem etwas anderen Schwerpunkt versehene Version des Businessbreakfast-Vortrages.

I'm playing around with Airmail for reading mails. While configuring this application, i saw the nuclear option to bring down any mail traffic in the company for days with "remove me from this discussion" mails or private comments/insults/gossip that never should have been send to all people:



I hope the developers of Airmail know that they have created a monster

A few days ago i wrote, that my talk about "Parasitic Hadoop - Wie man bereits vorhandene Server für Hadoop mitnutzen kann und dabei die eigentliche Produktionslast nicht in Mitleidenschaft zieht." was accepted by the FrOSCon program comitee. The program is public since a few days, so want to say that you can attend this lecture on Sunday, 28th August at 14:00. The language of the lecture is German.

Nachdem der Entschluss nunmehr fest steht, das ich nach Oldenburg zurückziehe, schaue ich mich ja verstärkt nach Wohnungen um. Was Wunder auch, will ja nicht unter einer Brücke leben Irgendwie stehe ich da aber vor einem Luxusproblem: Es scheint so zu sein, als waeren die Wohnungen in den von mir präferierten Gegenden durchschnittlich ein ganzes Stück grösser als meine Wohnung in Hamburg oder die an meinem Hauptwohnsitz in der Nähe von Oldenburg. Auf dem Weg zur letzteren entstehen übrigens immer die "Am A... der Welt"-Photos. Um sehr regelmässig nach Hamburg zu pendeln ist das da ein wenig zu sehr am A... der Welt. Also wird es direkt nach Oldenburg in die Nähe der Innenstadt gehen, wenn sich da dann was Akzeptables finden lässt....

Nungut, die Wohnung scheinen wirklich größer zu sein. 80 oder 100 qm? Teurer ist das nicht wirklich, aber was soll ich damit? Ich hab schon Probleme meine jetzige Wohnung in Ordnung zu halten und selbst auf meiner 20qm Studi-Wohnheimsbutze war das damals ein Herausforderungen. Andererseits, der Krempel aus den zwei Wohnungen will auch zusammengeführt werden. Und vielleicht klappt das ja mit dem Ordnung halten, wenn ich nicht ständig zwischen Oldenburg und Hamburg hinunherfahre und ich mich an beiden Orten nicht um jeweils einen Haushalt kümmern muss. Sondern meine eingeschränkten Aufräumfähigkeiten auf eine Wohnung konzentrieren kann. Und so ein richtig grosses,helles Arbeitszimmer haette auch was ...

The only way to wake up your audience after 20 sildes with statistics and pie charts



(inspired ... eehm ... stolen from presentation zen)

I still search for a small unit for doing presentations. Those small and beautiful UMPC are much to expensive for this task. I start to think about buying an Eee PC as soon they are available in germany. This tiny devices got some good reviews. As i started to use Keynote for preparing the presentation and Acrobat Reader for finally presenting it, this may be a solution ...

Looks like an interesting press announcement tomorrow. Let´s keep the finger crossed and hope that they will announce that they ditch AIX

I´ve ranted yesterday about the marketing habits of IBM. But kudos to IBM, they have excellent documentation, they even document the weaknesses of their technologies. So you find very intersting documents about their technology. Not that they hadn´t widthdrawn redbooks in the past like the one that documented the overhead of mpars, but mostly it´s a good source of realistic information. And thanks to the RedBooks it´s easy to clear some of fuss around some IBM technologies. And so, you find some interesting facts for example in the Redbook. And in this article i want to talk about WPARS. Many people think, that this is a cheap rip off of Solaris Zones. But many of them think, that the Live WPARS mobility is a cool feature, too. But when you really read public documentation much of the fuss appears a little bit overblown. Let´s have a look in the Redbook 247431: Introduction to Workload Partition Management in IBM AIX":

• The application in a WPAR has to write into a filesystem.

  Page 32: All files that needs to be written by the application must be hosted on an NFS filesystem

  That means: No large applications that needs direct raw access to disk like Oracle on WPARS and the speed of you application is limited to NFS.
• WPARs works with checkpoint/restore. That means: When you want to migrate your application, you have to freeze the WPAR. Then you have to checkpoint it. The checkpoint state file will be written via NFS to the shared storage. When you restart the WPAR on a different system, the process loads the checkpoint state file from the NFS server, and starts the WPAR. The application doesn´t run while to take or revive the snapshot:

  The chkptwpar command captures a snapshot of all tasks executing within one WPAR. It first interrupts all processes so they reach a quiescence point, then stores a copy of the processes context in a state file.

  Still sound unproblematic to you? The problem is in a detail. Think about an application that allocates 8 GB of memory (your java application for example), now take into consideration that most datacenters run on Gigabit Ethernet. Let´s assume 80 MB/s per Gigabit Ethernet interface via NFS. Thus the writing the state file would take 100 seconds. Reading it on the target system: Another 100 seconds. IBM states:

  The only visible effect for a user of the application is a slightly longer response time while the application is migrating.

  Well, i don´t know your users, but three and a half minute application downtime isn´t a slightly longer response time.

Okay, perhaps you can live with all this stuff, but still believe this, then you should read the manual for the Workload Partition Manager. The parts about Migration compatibility beginning arround page 27 is especially interesting: There are compatibility modes like "inbound/outbound compatile". An example:

Outbound Compatible
Compatibility testing shows that a WPAR can be relocated from the departure system to the arrival system, but it cannot be relocated back from the arrival system to the departure system.

The first time i´ve heard of it, i thought this is a joke, but it´s documented in IBM´s own documentation. It´s perfectly possible, that you can migrate a system to another system, but not back. When i understand it correctly, all systems have to patched to exactly the same level. When the patchlevel is lower on the arrival machine than on the departure machine, you can´t migrate to the system. When the patchlevel ist higher on the arrival machine than one the depature system, you can´t migrate back to your old system. And the libc has to be same on both systems in any case. Whoa .. what a bullshit ...

When i look at the WPARS solution, it looks like a really bad kludge. It´s good for the bullet point list wars.I´m sure that IBM sales reps will run around with WPARS and tell "Sun cannot do that" and maybe some customers who doesn´t like Sun, will use it as an argument to their management, to explain why they didn´t bought Sun.

But honestly, I don´t think, that you really can use it for anything practical.

PS: I´m not an AIX expert. When i´ve got something wrong, please correct me!

Interesting article about IBMs step back to "in order architectures":

What do you get out of it performance wise? If you look at the specfp2006, Power5+ at 2.2GHz reached 14.9 specfp2006, while Power6 reaches 22.3 specfp2006 at 4.7GHz - but of course, in an adapted Power5 machine. In summary, just under half additional speed at well over twice the clock.

The Power6 was something like the bogey man in the RISC world, but after all it´s a good evolution of their Power architecture. But´s by far not the "kill-all-other-proc" IBM and some analysts wanted to make us believe.

@last
@home

Everybody who thinks that a job with much travel is cool never sat in a McDonalds at 10 pm to wait for the last train towards home. At least the flight to Zürich was postponed to next week (no, not a customer in switzerland, there are locations in germany, you reach better from a foreign airport). But instead of this i´m scheduled to fly to San Francisco at the 19th.

Vielleicht sollten die Menschen in Firmen, die sich Geschäftsprozesse einfallen lassen für ein Jahr ein Praktikum in der Chipentwicklung machen. Denn die Herausforderungen sind ja recht ähnlich. Angenommen man muss durch verschiedene Stufen der Bearbeitung durch einzelne Abteilungen, dann ist das nix anderes als ein klassisches superskalares Design, nur das die einzelnen Piplinestufen hier anders heissen. Hat natuerlich den nachteil, wenn in irgendeiner Pipelinestufe jemand sagt: "Taugt nix", dann muss die ganze Pipeline geflushed werden und noch mal angefangen werden. Als einigermassen flexibler Mitarbeiter ist man sowieso dauernd im "out-of-order"-modus. Ständig zwischen den Jobs hin und her schalten.

Was Approvals angeht, die Manager geben muessen, ist dann eine gut funktionierende Branch Prediction recht sinnvoll, damit man sich moeglichst überfluessige Arbeit spart, wenn sich der Chef dann kurz vor Ende doch für "Geben wir nicht ab" entscheidet. Gleichsam möchte jeder VB die Ausführung mit "speculative execution", so wie es sie im Rock-Prozessor gibt: "Macht euch schon mal die Arbeit, wir gucken dann mal ob es überhaupt grundlegend genehmigungsfähig ist".

Vielleicht sollte man sich im Designprozess von Geschäftsprozessen die Lehren des Niagara-Prozessors vor augen halten. Hoher Durchsatz durch hohe Parallelität von unabhängigen Einheiten statt der Zentralisierung, die mehr dazu geeignet ist, einen Task möglichst schnell durchzuführen.

Bei so viel Ähnlichkeiten wäre es wirklich sinnvoll, wenn jeder Manager und Prozessverantwortliche zumindestens einen Crashkurs in moderner Prozessortechnologie erhält.

Nun ... ein Wochenende Ruhe ... viel nachgedacht. Es stehen eine ganze Reihe Frontbegradigungen in der nächsten Zeit an. Ich habe eine Reihe von Entschlüssen gefasst. Der wichtigste Punkt auf meiner Liste: Lernen "Nein" zu sagen. Gemäß dem Rat einer Person, an deren Meinung mir viel liegt. Alles was mir nicht wirklich wichtig ist, rauswerfen.

Ich habe zwar noch einige Dinge ausgeklammert, aber da muss ich einfach mal gucken, wie es sich dort weiterentwickelt. Vor allen Dingen werde ich die nächste Wochen erstmal zur gedanklichen Positionsbestimmung benutzen.