Introduction About Site Map

XML
RSS 2 Feed RSS 2 Feed
Navigation

Main Page | Blog Index

Archive for the ‘Personal’ Category

Windows High-Perfrmance Computing – Miserable Failure

Macs cluster

I may have just helped in intercepting a key decision and plan. It was the plan of our Division to convert a Linux HPC cluster into a Windows Compute Server, deploying it across approximately 32 nodes.

Not everybody was happy with this decision which made by just one junior staff, especially after I had voiced my own opinions. That guy, a regular Windows user, was rather frustrated. He wanted to convert a perfectly-operational Linux cluster into something else. He wanted to do so because the *nix system administrator who runs the Linux supercomputer has recently left. Without much clue about the alternative route, he went ahead and made a proposition to the management, contacting Microsoft in the process.

It worries me when such decisions are pushed forward by people who are not qualified to judge. The initiator does not know anything but Windows, let alone is able to draw a comparsion between different high-performance solutions (IBM have clearly stated that Linux is superior in that arena).

The person in question was/is bound to go to a training course on using Windows supercomputing, but an afternoon IT staff meeting will review the whole plan and possibly scrape it. I feel as though I have, to some extent, prevented a huge and expensive mistake from being made. That was my Linux advocacy for today and it was done locally, where it truly matters, rather than over cyberspace.

The Road to Completion

SO it’s official. I have submitted the Notice of Submission form, which currently indicates that I should submit my thesis strictly on schedule, i.e. at the end of September. A final draft is due by the end of August, so I will bury myself to comply with the schedule. I have no clear plans for the period that extends beyond September, but I will definitely take a vacation. I probably need one, but I always have my doubts, based on experience.

On the down side of things, my submission to the British Machine Vision Conference was rejected yesterday. Ho hum. The paper outlined a method for evaluating model such as the one shown below.

First mode of an appearance model of the brain

Appearance model of the brain, ±2.5 standards deviations shown

Meeting the European Free Software Foundation

Season of the playful penguins
Season of the playful penguins from Oyonale

I have just returned from a very nice evening/night out with the European Free Software Foundation. It was a very intersting experience, which I would have happily become a part of had it occurred more frequently.

This weekend was their annual General Assembly, so we went out for drink and dinner. I appeared to have been the only person among the group (except for perhaps one other person) who was not in the foundation or the comity. Despite this, I believe we hit it off. All in all, we got along very well as everyone was open and friendly. I managed to meet the people who coordinate participantion and contributions in different countries across Europe. I might get involved or join them in the future. We shall soon find out. The goals seem more political than technical.

Good News Come in Couples

Poor appearance model
The first mode of variation in a
poor appearance model of the brain

SOME “dreadful deadline”, which I mentioned in my previous post, is thankfully behind us, after a short extension. To make matters much better, I have just received a notification of acceptance for another conference, which saturates my once-modest publications list. Tomorrow I may find out that another paper was accepted (of which I am not the first author though).

All in all, such decent progress will make it easier for me to complete the Ph.D. fairly early. I will be presenting my work in July and fingers remain crossed for TMI. When one gets showered by many achievements at once, it is hard not to make a vanity blog post. I hope others feel happy for me as I have already informed my entire family.

Rowing Challenge Won

Game seat
Picture from the BBC

EXCITING news (for me at least) came out yesterday afternoon. I won the rowing competition (first stage among 10) with the time of 10:43.5 for 3000 meters. The second among the 10 contestants was almost 1 minute behind me, which considering my neglected endurance, makes this a high achievement.

Related items on exercise:

Returning to Top of Things

I finally caught up with work. Having returned from this long Easter vacation overwhelmed by workload, I was rather restless for days. I am still feeling sore due to lack of exercise. This comes at a very bad time as a contest is commencing rather soon and I must be in a better shape.

I can finally take the time to post a photo which I was sent a couple of days ago. The photo shows myself, my cousin Rachelle, and my cousin Michael at their house in Davie, FL.

Roy in Davie

Back Online, Re-gaining Pace

I returned from a long vacation yesterday morning. It will take me a while to catch up with the workload and then resume a stable blogging pace. For the mean time, here are a couple photos from my stay at my cousin’s house in North Miami Beach.

Garden view
A view on the yacht at sunset, taken from the garden

Family and Miami
I am the person at the bottom. Not a good
photo, but I hadn’t much choice.

The conference in Virginia, by the way, turned out very well (for me, at least). Our work was overwhelmed by attention and, as it turned out a couple of days ago, it is likely to reach the world’s most formidable (in the good sense of the word) journal on medical imaging. It is work in progress, which joined two other in-progress submissions. Prospects are positive and the future seems bright.

Retrieval statistics: 21 queries taking a total of 0.084 seconds • Please report low bandwidth using the feedback form
Original styles created by Ian Main (all acknowledgements) • PHP scripts and styles later modified by Roy Schestowitz • Help yourself to a GPL'd copy
|— Proudly powered by W o r d P r e s s — based on a heavily-hacked version 1.2.1 (Mingus) installation —|