view article

Figure 3
Large arrays are partitioned along the axis of rotation using a two-tier partitioning scheme. Larger, tier-1, partitions are per GPU. Smaller, tier-2, partitions (orange and green rectangles) are asynchronously streamed into GPU to overlap data transfers and computations.

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775
Follow J. Synchrotron Rad.
Sign up for e-alerts
Follow J. Synchrotron Rad. on Twitter
Follow us on facebook
Sign up for RSS feeds