lev_lafayette's blog

Whilst it is a loose metric, our little cluster, "Spartan", at the University of Melbourne ran its 1 millionth job today after almost exactly a year since launch.

The researcher in question is doing their PhD in biochemistry. The project is a childhood asthma study:

"The nasopharynx is a source of microbes associated with acute respiratory illness. Respiratory infection and/ or the asymptomatic colonisation with certain microbes during childhood predispose individuals to the development of asthma.

Using data generated from 16S rRNA sequencing and metagenomic sequencing of nasopharyn samples, we aim to identify which specific microbes and interactions are important in the development of asthma."

Moments like this is why I do HPC.

Congratulations to the rest of the team and to the user community.

After using Duolingo for over a year and accumulating almost 100,000 points I thought it would do the right thing and pay for the Plus service. It was exactly the right time as I would be travelling overseas and the ability to do lessons offline and have them sync later seemed ideal.

For the first few days it seemed to be operating fine; I had downloaded the German tree and was working my way through it. Then I downloaded the French tree, and several problems started to emerge.

Gaussian is a well-known computational chemistry package, and sometimes subject to debate over its license (e.g., the terms state researchers who develop competing software packages are not permitted to use the software, compare performance etc). Whilst I have some strong opinions about such a license, this will be elaborated at another time. The purpose here is to illustrate the use of heredocs with Slurm.

High performance computing and cloud computing have traditionally been seen as separate solutions to separate problems, dealing with issues of performance and flexibility respectively. In a diverse research environment however, both sets of compute requirements can occur. In addition to the administrative benefits in combining both requirements into a single unified system, opportunities are provided for incremental expansion.

Advanced Computing : An International Journal was a publication that I considering writing for. However it is almost certainly a predatory open-access journal, that seeks a "publication charge", without even performing the minimal standards of editorial checking.

I can just tolerate the fact that the most recent issue has numerous spelling and grammatical errors as the I believe that English is not the first language of the authors. It should have been caught by the editors, but we'll let that slide for a far greater crime - that of widespread plagiarism.

The fact that the editors clearly didn't even check for this is an inexcusable oversight.

I opened this correspondence to the editors in the hope that others will find it prior to submitting or even considering submission to the journal in question. I also hope the editors take the opportunity to dramatically improve their editorial standards.

Multicore World is a small conference held annually in New Zealand hosted by Open Parallel. What it lacks in numbers however it makes up in quality of the presenters. The 2017 conference included a typically impressive array of speakers dealing with some of the most difficult issues facing computational science, and included several important announcements in the fields of supercomputing, the Internet of Things, and manufacting issues.

A visit and presentation was conducted at Auckland University of Technology on February 27th, 2017. AUT has a small HPC research laboratory and like the rest of New Zealand makes us of the NESI national facilities, of which the "Pan" system hosted at the University of Auckland is local.

I have recently spent a few days in the company of Nyriad, a New Zealand IT company specialisng in GPU software. I wish to make a point of a few observations of the company because they are an example of both a startup company that uses agile project management, two terms much maligned and subject to justified cynicism, and does it right. Because I have seen so many colleagues burned by companies and organisations which profess such values and do not do it right, I hope the following observations will be useful for future organisations.

In recent weeks a project has been established by University of Melbourne marine research led by Dr. Eric Treml and Nyriad, an startup NZ company specialising in GPU software, in optimising code for marine population samples. Nyriad's main mission however is aimed to resolve one of the biggest and growing issues in computation, the growing gap between data computation and data i/o. The technical solution, led by Alex St.

It is difficult to describe the annual Multicore World conference with brevity. For the past six years it has operated out of New Zealand, the brain-child of Nicolas Erdody of Open Parallel, and for five of those six years your 'blogger has had the honour of MC for much of the proceedings. It is not a big conference by any stretch of the imagination, typically attracting around seventy participants.