Communications
From Modelado Foundation
QUESTIONS | XPRESS | TG X-Stack | DEGAS | D-TEC | DynAX | X-TUNE | GVR | CORVETTE | SLEEC | PIPER |
---|---|---|---|---|---|---|---|---|---|---|
PI | Ron Brightwell | Shekhar Borkar | Katherine Yelick | Daniel Quinlan | Guang Gao | Mary Hall | Andrew Chien | Koushik Sen | Milind Kulkarni | Martin Schulz |
What are the communication "primitives" that you expect to emphasize within your project? (e.g. two-sided vs one-sided, collectives, topologies, groups) Do we need to define extensions to the traditional application level interfaces which now emphasize only data transfers and collective operations? Do we need atomics, remote invocation interfaces, or should these be provided ad-hoc by clients? | (EXPRESS) | (TG) | (DEGAS) | (D-TEC) | (DynAX) | (X-TUNE) | (GVR) | (CORVETTE) | SLEEC | Communication will be out of band and need to be isolated, emphasis on streaming communication |
Traditional communication libraries (e.g. MPI or GASNet) have been developed without tight integration with the computation "model". What is your strategy for integrating communication and computation to address the needs of non SPMD execution? | (EXPRESS) | (TG) | (DEGAS) | (D-TEC) | (DynAX) | (X-TUNE) | (GVR) | (CORVETTE) | N/A | N/A |
What type of optimizations should be transparently provided by a communication layer and what should be delegated to compilers or application developers?
What is the primary performance metric for your runtime? |
(EXPRESS) | (TG) | (DEGAS) | (D-TEC) | (DynAX) | (X-TUNE) | (GVR) | (CORVETTE) | N/A | unclear |
What is your strategy towards resilient communication libraries? | (EXPRESS) | (TG) | (DEGAS) | (D-TEC) | (DynAX) | (X-TUNE) | (GVR) | (CORVETTE) | N/A | ability to drop and reroute around failed processes |
What and how can a communication layer help in power and energy optimizations? | (EXPRESS) | (TG) | (DEGAS) | (D-TEC) | (DynAX) | (X-TUNE) | (GVR) | (CORVETTE) | N/A | N/A |
Congestion management and flow control mechanisms are of particular concern at very large scale. How much can we rely on "vendor" mechanisms and how much do we need to address in higher level layers? | (EXPRESS) | (TG) | (DEGAS) | (D-TEC) | (DynAX) | (X-TUNE) | (GVR) | (CORVETTE) | N/A | N/A |