Video processing on HPC


Currently, all users of a special microscope that generates terabytes of data must use a Windows workstation for data processing. Bob adapted this PC workflow to Harvard’s HPC environment; this now provides flexibility & higher throughput for the lab’s image processing needs.

“To help us start tackling this challenge, we greatly benefited from the help of Bob Freeman from the ACI-REF program, who helped us transition from using a workstation to parallelizing and speeding image processing using cluster computing at Harvard. This has transformed our ability to process and think about the data, as well as radically changed the way we plan and design experiments given these new computational power for downstream analyses. Furthermore, the computational pipelines designed by Bob are not only applicable to our immunological questions, but also to virtually any field of research using light sheet imaging – from neuroscience to development among others. Lastly, this work – between ACI-REF, research computing, the Harvard center for biological imaging, and several labs across the FAS center for systems biology and molecular and cellular biology departments – has enabled new collaborations across departments, innovative data analysis solutions, and educated my group, including undergraduates, to the incredible power of cluster computing to accelerate and transform our research.” –Nicolas Chevrier, Harvard University Bauer Fellow, on Harvard ACI-REF Bob Freeman