Aurora, hosted by Argonne National Laboratory, is one of three planned Exascale-class systems in the United States. While the Intel-led system has seen a myriad of conceptual transformations (originally planned as a pre-exascale system) and setbacks (Intel’s delayed 7nm node), the avid researcher hasn’t stopped planning on using the massive system. Among those researchers is a team led by Amanda Randles, an assistant professor of biomedical sciences at Duke University, who is preparing to use Aurora to visualize how cancer spreads in the human body.
An example of HARVEY’s blood flow simulations. Image courtesy Joseph Insley.
The tool in question is called HARVEY, and it was previously used to track how microscopic blood cells flow through our bodies. The researchers are using the same skills to model microscopic cancer cells as well, hoping to harness Aurora’s newly discovered power to create a more accurate model of the circulatory system and understand how tumor cells move through that model. “By understanding the biological mechanisms behind metastatic cancer cells, we hope that our collaboration with HARVEY will ultimately help doctors and their patients fight cancer,” Randles said in an interview with Joan Koka of Argonne.
The team has a lot of time before Aurora arrives, but also a lot of hurdles to overcome before that point: For example, how to explain the big differences between Aurora’s computing speeds and hard drive writing speeds. The researchers, like many others faced with the same problem, try to do at least some of the data analysis inside the machine (rather than exporting and analyzing the data) to fix the problem.
“For example, if you identify a specific area of interest throughout the system, you can store data in that small area with a higher frequency or a higher level of detail. That way, you can still get more science out of the data while reducing the amount of data you actually need to write to disk, ”said Joseph Insley, head of the visualization and data analysis team at Argonne Leadership Computing Facility (ALCF). .
“When we tackle our new research on the process of metastasis and perform the complex simulations that are required, we will need even more computing power to process the huge data sets in real time. The Aurora system will help us meet that need, ”added Randles.
The ALCF has also prepared a unified library of frameworks for this process of data analysis and visualization on the machine (part of a DOE project called SENSEI) that the ALCF employees are working on to convert the HARVEY code base using the existing supercomputing power of the Prepare for ALCF.
“We used the ALCF’s theta supercomputer and early Aurora hardware to incorporate this library into the HARVEY code,” said Insley. “By enabling visualization and analysis of the data while it is still in memory, the science team can gain more insight from the data than they would otherwise.”
Randles’ team is one of 15 selected for the ALCF’s Aurora Early Science Program (ESP). Aurora is scheduled to ship in 2022.
Click here to read the coverage of Joan Koka from Argonne.