Last week, SC10 was held in New Orleans, LA. SC is an annual international conference for High Performance Computing, Storage, Networking and Analysis, and this year’s festivities were the biggest and best attended yet. The meeting’s technology thrusts were Climate Simulation, Heterogeneous Computing, Data Intensive Computing. Read an interview about the meeting with general chair Barry Hess.
I could not attend, but I was particularly interested in papers on the last topic. I was fortunate to be a co-author on a paper entitled “Data Sharing Options for Scientific Workflows on Amazon EC2,” by Gideon Juve, Ewa Deelman, Karan Vahi, Gaurang Mehta, Bruce Berriman, Benjamin P. Berman, and Phil Maechling (download paper), which explored how the cost and performance of running scientific workflows on Amazon EC2 depended on the storage options offered by Amazon.
These were papers that caught my eye:
“Computing the Universe,” by Salman Habib, which explored how high end computing will be needed to interpret the latest cosmology survey results.
“Cyber-Infrastructure for the LSST Data Management System,” by Jeffery Kantor.
“High Performance Scientific Workflows,” by Adrian Jackson and Marcin Plociennik, which explored how workflow technologies can be used to study big science questions.
“Performance Tuning of Scientific Applications on HPC Systems,” by Michele Weiland, Jason Beech-Brandt, Charles Henriet, Eric Michele, Nils Smeds, Luigi Brochard.
“Provisioning Virtual Clusters in the Cloud using Wrangler,” by Gideon Juve and Ewa Deelman