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
/** @brief Provides programmatic access to the computer being run on. **/structRuntime { /// Initializes the runtimeRuntime(); /// Gets the type-erased MPI Communicator std::any mpi_comm(); /// Returns the number of clusters (can have multiple clusters for inter-communicators) size_type size(); /// Gets the Cluster object associated with the runtime Cluster operator[](size_type i); /// Not sure what the arguments are, but should partition the runtime into smaller runtimes Runtime partition(); /// Tears down the runtime, releasing resources~Runtime(); };
Key points:
We want MPI to be fully encapsulated
Initialization/tear-down should follow RAII
Basically a container of Cluster objects
Questions:
Does it make sense to have a single output file or something here? Or should users just capture stdout/err from all nodes?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
As far as an API goes I'm thinking:
Key points:
ClusterobjectsQuestions:
Beta Was this translation helpful? Give feedback.
All reactions