New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

"WITH_INFO" review across the collective API #84

Closed
tonyskjellum opened this Issue Mar 7, 2018 · 4 comments

Comments

@tonyskjellum

tonyskjellum commented Mar 7, 2018

Problem

Make sure all the functions have variants "WITH_INFO" for which this makes sense.

Proposal

Check the collective API, starting with the GRAPH/CART APIs below, to see where WITH_INFO is missing but needed; add such APIs. In some cases, we might deprecate the prior meaning of the info argument instead (TBD per API).

Changes to the Text

MPI_GRAPH_CREATE_WITH_INFO
The "non-dist" graph constructor should be deprecated in favour of the newer (and better) "dist" version(s). Every call to MPI_GRAPH_CREATE can be legally and 'simply' replaced with a call to MPI_DIST_GRAPH_CREATE.

MPI_GRAPH_CREATE requires that each process passes the full (global) communication graph to the call. This limits the scalability of this constructor. With the distributed graph interface, the communication graph is specified in a fully distributed fashion.

In the absence of a proposal to deprecate MPI_GRAPH_CREATE (and thereby obviate the need to maintain/improve it), adding an MPI_INFO argument is one of the advisable changes/improvements.

MPI_CART_CREATE_WITH_INFO
The proposed cartesian topology "with info" constructor seems like a good addition but should probably be part of a more general change to add an MPI_INFO argument to all object constructors that don't already have one. For example, MPI_COMM_SPLIT_WITH_INFO makes a lot of sense too.

Impact on Implementations

Incremental APIs will need to be added; these are analogous to existing APIs and mechanisms to manage WITH_INFO already exist in other parts of the API.

Impact on Users

Describe the changes that will impact users here.

References

Issue #80 must be cross-correlated; See also Issue #85.

@tonyskjellum tonyskjellum self-assigned this Mar 7, 2018

@dholmes-epcc-ed-ac-uk dholmes-epcc-ed-ac-uk self-assigned this Mar 7, 2018

@hjelmn hjelmn self-assigned this Mar 7, 2018

@tonyskjellum tonyskjellum changed the title from "WITH_INFO" review across the collective API; specifically GRAPH and CART in "Topology" Chapter to "WITH_INFO" review across the collective API; specifically GRAPH and CART in "Process Topologies" Chapter Mar 8, 2018

@tonyskjellum

This comment has been minimized.

tonyskjellum commented May 28, 2018

We have plenty of work for Austin, and we need to understand the plans and outcomes of Ticket #80 to pursue this ticket efficiently. So I am going to mark its goal as for Barcelona, not Austin. Let's discuss in the WG in Austin.

@tonyskjellum

This comment has been minimized.

tonyskjellum commented Jun 14, 2018

We reviewed this in the working group time in Austin. We will discuss this further in the plenary in Austin ahead of presenting this as a reading for the Barcelona meeting.

@tonyskjellum tonyskjellum changed the title from "WITH_INFO" review across the collective API; specifically GRAPH and CART in "Process Topologies" Chapter to "WITH_INFO" review across the collective API in all Chapters Sep 24, 2018

@tonyskjellum tonyskjellum changed the title from "WITH_INFO" review across the collective API in all Chapters to "WITH_INFO" review across the collective API in "Topologies Chapter" Sep 24, 2018

@tonyskjellum

This comment has been minimized.

tonyskjellum commented Sep 24, 2018

Ticket #85 is apparently a superset of this ticket.

@tonyskjellum tonyskjellum changed the title from "WITH_INFO" review across the collective API in "Topologies Chapter" to "WITH_INFO" review across the collective API Sep 26, 2018

@tonyskjellum

This comment has been minimized.

tonyskjellum commented Sep 26, 2018

Closing this; see Ticket #85.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment