3 citations · 4 across the 11 of their papers we have counts for
11 papers
Co-Design and Evaluation of a CPU-Free MPI GPU Communication Abstraction and Implementation
Patrick G. Bridges, Derek Schafer, Jack Lange +7
Removing the CPU from the communication fast path is essential to efficient GPU-based ML and HPC application performance. However, existing GPU communication APIs either continue t…
Leveraging Caliper and Benchpark to Analyze MPI Communication Patterns: Insights from AMG2023, Kripke, and Laghos
Grace Nansamba, Evelyn Namugwanya, David Boehme +6
We introduce ``communication regions'' into the widely used Caliper HPC profiling tool. A communication region is an annotation enabling capture of metrics about the data being com…
Understanding GPU Triggering APIs for MPI+X Communication
Patrick G. Bridges, Anthony Skjellum, Evan D. Suggs +2
GPU-enhanced architectures are now dominant in HPC systems, but message-passing communication involving GPUs with MPI has proven to be both complex and expensive, motivating new ap…
MPI Implementation Profiling for Better Application Performance
Riley Shipley, Garrett Hooten, David Boehme +3
While application profiling has been a mainstay in the HPC community for years, profiling of MPI and other communication middleware has not received the same degree of exploration.…
Implementation-Oblivious Transparent Checkpoint-Restart for MPI
Yao Xu, Leonid Belyaev, Twinkle Jain +3
This work presents experience with traditional use cases of checkpointing on a novel platform. A single codebase (MANA) transparently checkpoints production workloads for major ava…
MPI Advance : Open-Source Message Passing Optimizations
Amanda Bienz, Derek Schafer, Anthony Skjellum
The large variety of production implementations of the message passing interface (MPI) each provide unique and varying underlying algorithms. Each emerging supercomputer supports o…
A More Scalable Sparse Dynamic Data Exchange
Andrew Geyko, Gerald Collom, Derek Schafer +2
Parallel architectures are continually increasing in performance and scale, while underlying algorithmic infrastructure often fail to take full advantage of available compute power…
Collective-Optimized FFTs
Evelyn Namugwanya, Amanda Bienz, Derek Schafer +1
This paper measures the impact of the various alltoallv methods. Results are analyzed within Beatnik, a Z-model solver that is bottlenecked by HeFFTe and representative of applicat…
Checkpoint-Restart Libraries Must Become More Fault Tolerant
Anthony Skjellum, Derek Schafer
Production MPI codes need checkpoint-restart (CPR) support. Clearly, checkpoint-restart libraries must be fault tolerant lest they open up a window of vulnerability for failures wi…
MPIs Language Bindings are Holding MPI Back
Martin Ruefenacht, Derek Schafer, Anthony Skjellum +1
Over the past two decades, C++ has been adopted as a major HPC language (displacing C to a large extent, andFortran to some degree as well). Idiomatic C++ is clearly how C++ is bei…
Extending the Message Passing Interface (MPI) with User-Level Schedules
Derek Schafer, Sheikh Ghafoor, Daniel Holmes +2
Composability is one of seven reasons for the long-standing and continuing success of MPI. Extending MPI by composing its operations with user-level operations provides useful inte…