Closed hadim closed 1 year ago
No hard recommendation at the moment, but it should be larger than the previous version. There are a few operations (particularly submit()
) that will take too long if there are lots of molecules, but that is actively being worked on. Other than that, we will just have to try larger datasets and see if/when it falls over.
(In the future, operations like submit()
will be handled server-side outside of the client request by an internal job process, if that makes sense)
Ok thanks for the informations!
In the SPICE dataset they split the entire dataset into subset and then chunks.
I am wondering what is recommended with the new qcfractal server?