Originally published in Arab News, September 7, 2020
KAUST Ph.D. graduate, Dr Tariq Alturkestani, won the best paper award at the prestigious annual international conference Euro-Par 2020. The paper was selected out of 158 papers that were submitted by candidates from all over the world.
The submission was based on Alturkestani's doctoral thesis project which focuses on overlapping I/O and compute in large-scale scientific computation using multilayered buffering mechanisms.
"Ph.D. is about pushing limits. It takes years of designing, implementations, engineering and hard work," said Tariq Alturkestani, Ph.D. graduate at the KAUST Extreme Computing Research Center (ECRC). "To be recognized by the research committee in a prestigious conference at the finish line of my Ph.D. journey is an honor and an inspiration for me to push the limits further."
Alturkestani's work utilizes new emerging storage technology such as node-local and remote-shared Solid-State Drive (SSD), Non-volatile Memory (NVMe) and High Bandwidth Memory (HBM).
The project re-evaluates the computer-based method Reverse Time Migration (RTM) used by geoscientists to explore seismic data for oil and gas explorations. This method requires storing huge amount of data before calling it back in stages and correlating with later available data to create the desired output image. Making RTM run faster translates into an increase in capacity for seismic simulations and eventually obtaining subsurface images to characterize the presence of hydrocarbons with higher fidelity.
Read the full article