Supercomputing Teaching, Researchers, and More
Jul. 26th, 2025 03:51 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
The past couple of weeks have resulted in some rather pleasing work-related events. The first followed a meeting with representatives of the Pawsey Supercomputing Centre in Western Australia which has resulted in me acquiring a test account and becoming the responsible person for approving projects that are seeking to upgrade from UniMelb's Spartan to systems like Setonix, which is the most powerful in the country and number 59 in the world (this time last year it was 28 - the landscape moves quickly!). The architecture is somewhat different to what I am used to (HPE Cray OS, HPE Cray MPI), so I'll be doing some additional learning myself and some hand-holding for researchers as they navigate and migrate data and code to this system. At the other end of the scale for the newcomers on Spartan I conducted two introductory workshops last week with some fifty attenedees. Quite a few times now, I have met researchers several years after taking my introductory courses who express their significant gratitude that I led them on the path of using high performance computing to process their data with efficiency, and hopefully, there will be a few of that nature from this cohort.
I often make the point of how supercomputers are utterly critical for current economic development with an ROI of 7:1 almost all through positive externalities, and literally save millions of lives. Could you imagine COVID-19 vaccines without HPC systems to do the simulations? We'd still be living under lockdown. Well, on a smaller scale, I was very proud to organise a researcher presentation last week for Research Computing Services (RCS) and Melbourne Data Analytics Platform (MDAP) with Dr Debjyoti Karmakar who is developing a non-invasive fetal monitoring device to prevent perinatal asphyxia for high-risk pregnancies. Dr Deb's presentation was very well received, even if he had to take a brief emergency call in the midst of it! He has been making extensive use of both the Spartan supercomputer and our Mediaflux-based data storage system, which dovetailed quite nicely with a virtual meeting a few days later with several representative from Princeton University who also are interested in integrating the HPC systems with Mediaflux, which is not as simple as it should be, but that is the nature of our work.
In my own research, I am still making good progress on studies in the psychology and sociology of climate change denial and have recently made contact with a long-standing member of the UNFCCC Accreditation Panel of Experts and Methodology Panel of Experts who shares this interest. Over the next two months or so, I am hoping to elaborate from this member's own interest on the psychological need for "belonging" and "club membership" which leads to climate science denial being strongly correlated with ideology rather than scientific evidence, and to draw a stronger correlation between this and vested interests in political economy. Ideological positions are usually strongly associated with political economy, so it should not surprise me to find such a connection, and when it comes to the damage done, there are those who are responsible, and one thing we do know from institutional and individual analysis, very few people like to take responsibility.
I often make the point of how supercomputers are utterly critical for current economic development with an ROI of 7:1 almost all through positive externalities, and literally save millions of lives. Could you imagine COVID-19 vaccines without HPC systems to do the simulations? We'd still be living under lockdown. Well, on a smaller scale, I was very proud to organise a researcher presentation last week for Research Computing Services (RCS) and Melbourne Data Analytics Platform (MDAP) with Dr Debjyoti Karmakar who is developing a non-invasive fetal monitoring device to prevent perinatal asphyxia for high-risk pregnancies. Dr Deb's presentation was very well received, even if he had to take a brief emergency call in the midst of it! He has been making extensive use of both the Spartan supercomputer and our Mediaflux-based data storage system, which dovetailed quite nicely with a virtual meeting a few days later with several representative from Princeton University who also are interested in integrating the HPC systems with Mediaflux, which is not as simple as it should be, but that is the nature of our work.
In my own research, I am still making good progress on studies in the psychology and sociology of climate change denial and have recently made contact with a long-standing member of the UNFCCC Accreditation Panel of Experts and Methodology Panel of Experts who shares this interest. Over the next two months or so, I am hoping to elaborate from this member's own interest on the psychological need for "belonging" and "club membership" which leads to climate science denial being strongly correlated with ideology rather than scientific evidence, and to draw a stronger correlation between this and vested interests in political economy. Ideological positions are usually strongly associated with political economy, so it should not surprise me to find such a connection, and when it comes to the damage done, there are those who are responsible, and one thing we do know from institutional and individual analysis, very few people like to take responsibility.