Presentation
Fast and Generic Concurrent Message-Passing
SessionDoctoral Showcase II
Author
Advisor
Event Type
Doctoral Showcase
W
TUT
TP
EX
EXH
Compilers
Data Analytics
Data Management
GPUs
MPI
OS and Runtime Systems
Doctoral Showcase
TimeTuesday, November 13th2:15pm - 2:30pm
LocationD221
DescriptionCommunication hardware and software have a significant impact on the performance of clusters and supercomputers. Message-passing model and the Message-Passing Interface (MPI) is a widely used model of communications in the High-Performance
Computing (HPC) community. However, MPI has recently faced new challenges due to the emergence of many-core architecture and of programming models with dynamic task parallelism, assuming a large number of concurrent threads. These applications come from important classes of applications such as graph and data analytics.
In this thesis, we studied MPI under the new assumptions. We identified several factors in the standard which were inherently problematic for scalability and performance. Next, we analyzed graph, threading and data-flow frameworks to understand the communication. We then proposed a communication engine (LCI) targetting these frameworks. Our thesis benefits MPI by developing several patches in production MPI. Furthermore, LCI presents a simple and ground-up design which benefits various frameworks of study.
Computing (HPC) community. However, MPI has recently faced new challenges due to the emergence of many-core architecture and of programming models with dynamic task parallelism, assuming a large number of concurrent threads. These applications come from important classes of applications such as graph and data analytics.
In this thesis, we studied MPI under the new assumptions. We identified several factors in the standard which were inherently problematic for scalability and performance. Next, we analyzed graph, threading and data-flow frameworks to understand the communication. We then proposed a communication engine (LCI) targetting these frameworks. Our thesis benefits MPI by developing several patches in production MPI. Furthermore, LCI presents a simple and ground-up design which benefits various frameworks of study.
Archive
