You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This issue is meant as a place-holder, so that we don't forget our discussion on how to handle multiple time-chunks in XBraid.
There is a branch that implements multiple time-chunks (called timechunks) that is integrated into Stefanie's SU2 code. This branch takes in a numchunks integer from the user, and then breaks the time-domain in numchunks evenly spaced sub-intervals. XBraid is then run on each subinterval.
Rob and I thought that there might be a more general way of doing this, namely having the user specify something like a maximum number of time-points allowed. Then, the time-domain would be split up this way.
We in particular liked this approach, because the reason for doing time-chunks is to save memory, and number of time-points is more natural way to "count" memory usage. One could even weight each time-point, similar to the load balancing strategy.
Additionally, a max-num-time-points parameter could be integrated into FRefine.
The text was updated successfully, but these errors were encountered:
This issue is meant as a place-holder, so that we don't forget our discussion on how to handle multiple time-chunks in XBraid.
The text was updated successfully, but these errors were encountered: