Lightweight threading with MPI using Persistent Communications Semantics
نویسندگان
چکیده
Multi-threaded performance in MPI is of concern for future systems, particularly at Exascale, where massive concurrency will be necessary to leverage the full power of systems. While MPI provides generalized solutions and additional proposals like endpoints expand this general model, examining common use cases that have good solutions that may not be universally applicable is a viable additional approach. This paper details a new conceptual concurrency support mechanism for MPI that is applicable to a (large) subset of MPI applications. This approach is expected to provide very low overhead while still allowing for optimizations in the MPI library that are not currently possible.
منابع مشابه
Practical Formal Verification of MPI and Thread Programs Tutorial Notes
This document is being prepared to serve as a reference for the tutorial with the same title that will be offered during EuroPVM/MPI 2009 in Espoo, September 7, 2009. In this tutorial, we will mostly focus on the topic of formal dynamic verification [1] of MPI programs using our tool ISP. There will also be a short session on shared memory program verification using our tool Inspect. Section 4 ...
متن کاملThe Chapel Tasking Layer Over Qthreads
This paper describes the applicability of the third-party qthread lightweight threading library for implementing the tasking layer for Chapel applications on conventional multisocket multicore computing platforms. A collection of Chapel benchmark codes were used to demonstrate the correctness of the qthread implementation and the performance gain provided by using an optimized threading/tasking...
متن کاملAn Operating System Support to Low-Overhead Communications in NOW Clusters
This paper describes an Operating System approach to the problem of delivering low latency high bandwidth communications for PC clusters running a public domain OS like Linux and connected by standard, oo-the-shelf networks like Fast-Ethernet. The PARMA 2 project has the main goal of designing the new lightweight protocol suite PRP, in order to drastically reduce the software overhead introduce...
متن کاملImplications of a PIM Architectural Model for MPI
Memory may be the only system component that is more commoditized than a microprocessor. To simultaneously exploit this and address the impending memory wall, processing in memory (PIM) research efforts are considering ways to move processing into memory without significantly increasing the cost of the memory. As such, PIM devices may become the basis for future commodity clusters. Although the...
متن کاملScalable Data-Privatization Threading for Hybrid MPI/OpenMP Parallelization of Molecular Dynamics
Calculation of the Coulomb potential in the molecular dynamics code ddcMD has been parallelized based on a hybrid MPI/OpenMP scheme. The explicit pair kernel of the particleparticle/particle-mesh algorithm is multi-threaded using OpenMP, while communication between multicore nodes is handled by MPI. We have designed a load balancing spanning forest (LBSF) partitioning algorithm, which combines:...
متن کامل