simple grid scheduling algo

grids are becoming increasingly popular but after you remove some fancy protocols and security measures they are still composed of distributed machines. If you have a fairly ‘dirty’ grid, ie., very heterogeneous and not endowed with MPI etc, when performing data decomposition, some thought must be given as to how data is divided. Suppose you would like to divide M same tasks amongst N processors. If (taking into account network latency etc) one can order the time it takes to perform a single task as a₁ < a₂ < .. < a_i < .. < a_N, then a naive division of labor for each node would be:

But this overlooks possible data collision and latency involved in loading buffers at the manager, so take this into account with an extra term, giving the tasks for a single node as:

where b is less than one, and since the total tasks handled by N procs must be M, this implies the relationship btwn r and b:

r may be determined from experiment, and b follows.

Complementarity

Here’s an extract from an old essay of mine, from way the heck back in 1995, submitted in PH237 at UQ, document link follows:

Bohr’s well documented opposition to Einstein’s corpuscular theory of light abated in lieu of the Bothe- Geiger experiments, in which the particle nature of radiative phenomena manifested itself. These results were in blatant contradiction with the Bohr-Kramer-Slater interpretation of the interaction between atomic systems. The B-K-S paper assumed that the radiative aspects of atomic transitions were solely describable in terms of the “wave picture”. It was this descriptive contrast which prompted Bohr to find
a harmonious relationship between the particle-wave nature of the radiative aspects of quantum interactions.

Essay

Press for Image Proc Work

In an unusual twist, a few news sources are reporting on the aforementioned image proc work at Penn State: New Scientist , Naked Scientists and L’Atelier

I’m flattered and more than a little surprised

ED: also showed up in ACM tech news

JCDL image extraction paper/talk

Here’s the paper abstract, links to paper/talk follow:

“Most search engines index the textual content of documents in digital libraries. However, scholarly articles frequently report important findings in figures for visual impact and the contents of these figures are not indexed. These contents are often invaluable to the researcher in various fields, for the purposes of direct comparison with their own work. Therefore, searching for figures and extracting figure data are important problems. To the best of our knowledge, there exists no tool to automatically extract data from figures in digital documents. If we can extract data from these images automatically and store them in a database, an end-user can query and combine data from multiple digital documents simultaneously and efficiently. We propose a framework based on image analysis and machine learning to extract information from 2-D plot images and store them in a database. The proposed algorithm identifies a 2-D plot and extracts the axis labels, legend and the data points from the 2-D plot. We also segregate overlapping shapes that correspond to different data points. We demonstrate performance of individual algorithms, using a combination of generated and real-life images.”

paper
talk

cyberinfrastructure part 1

Gridded, multi-scale problem solving environments (PSE’s) are designed to reduce the burden often associated with information processing, and also to provide a collaborative environment in which data, tools and resources are shared by the group. Traditional barriers such as geography and limited resources are dissolved, collaborations are readily formed and research goals may be accomplished more easily using this so-called cyberinfrastructure.

Attached are a cyber lecture series delivered to Chemists by this hapless Physicist. Web 2.0, languages (Java/PHP/HTML etc etc), collaboratory design/models, Machine Learning, visualization and more.

lecture notes

ranging device

Attached are the main elements for a vehicle ranger which I designed, built and tested (it worked). The unit was contained in acrylic which I fashioned and shaped in our kitchen apartment while doing PhD at W&M (note: acrylic doesn’t like 400 deg F for more than 5 minutes ). Control is via a PICmicro 16x, also included is the assembly code for programming the PIC using MPLAB. There should be links somewhere on this site to homebrew PIC programmers, you also need a UV light source to clear the PIC 16Fx if trying out different codes. Best to source the stepper motor and larger things from allelectronics in CA, I use digikey for smaller items.

If nothing else this project will ram home the difficulty in calculating the distance to a moving object using less than 100USD worth of parts Nice description of stepper motor types here

CAD drawings, assembly code & sketches

honours thesis

I completed this at UQ physics in 1999, here’s the abstract and the document itself. The device I constructed was originally intended for use with the scramjet project, although in my time was tested (successfully) on a small shock tunnel in the basement of Hawken. The shadowgraph images produced were spectacular although they seem to be MIA right now

---

Time resolved imaging in compressible flows using a Cranz-Schardin Camera

A Cranz-Schardin camera arrangement was devised and constructed to obtain time resolved shadowgraphs of the flow over a cylinder. Detached shock waves are clearly visible as well as some three dimensional effects. The camera system uses a high power AlInGaP light emitting diode (LED), the model HLMP-DG08 by Hewlett Packard. The diode has a narrow viewing angle (6^o), high luminous intensity (6500mcd @ 20mA), with peak emission at 626nm and a FWHM¹ of 17nm. Within the camera system, the LED is pulsed rapidly at high currents, serving as both shutter and light source. These pulses are separated temporally and spatially, back-lighting the object which is subsequently imaged onto the film plane. Generation of pulsed input to the sources is provided by CMOS² circuitry coupled with an analogue driver stage for each channel, producing high current gain. The maximum output of the analogue driver stages was in the vicinity of 5 amps, producing sufficient light intensity in the LED’s to saturate a CCD³ camera. The light emitting diode is advantageous over lasers in this application for a number of reasons. Besides the obvious reduction in cost, LED’s are responsive (tau_res = 20ns) may be switched rapidly and have a fairly good signal to noise ratio for such a small, inexpensive device. The camera overall promises to be an invaluable diagnostic tool, at least in situations where competing luminosity can be reduced.

1. Full Width (at) Half Maximum (intensity).
2. Complementary Metal Oxide Semiconductor.
3. Charge Coupled Device.

thesis

Hello Web 2.0

Allow me to introduce myself.  My name is Bill Brouwer, I finished a PhD in physics at the College of William and Mary in 2005. While my training is in physics, I’ve been privileged to work in a number of fields now, including cyberinfrastructure.

My first real cyber experience was related to a zilog z-80, part of a larger 100 pound “desktop” system which in 1978 retailed for 10k or more. By the time my sticky mits came into contact with the machine in 1987, the introduction of MS-DOS and the associated meteoric rise of the PC in the intervening years had  killed off all but middling interest in the venerable zilog, at least in mid 1980’s Brisbane. Thus, sans viable support, the ailing machine became the play thing of an eleven year old.

At the same juncture a disabled yet highly gifted friend of the family was called  to assist in the salvation of the zilog.  Morris ultimately coached me in electronics, and encouraged me to buy a Radioshack digital multimeter which I still use today. I am also an audiophile and tube fanatic, amongst other things, thanks to him. With great alacrity, he recited makes of chips within the zilog as well as their functions and his diagnosis based on the symptoms. Morris made short work of the beast, as he did with most things. During this time and forever thereafter I found myself immersed in an exquisite world, a marriage of physics, engineering and great practicality.

It was also Morris who introduced me to dial-up and BBS or Web 0.0 if you like, and so for all these reasons and more, this blog is dedicted to the comsumate genuis that was and I hope still is Morris.  And to my muse, my remarkable wife Rebecca.