ProOnE: a general-purpose protocol onload engine for multi- and many-core architectures

Lecture Notes in Computer Science

Taboada

2018

Non-blocking collectives have been proposed so as to allow communications to be overlapped with computation in order to amortize the cost of MPI collective operations. To obtain a good overlap ratio, communications and computation have to run in parallel. To achieve this, different hardware and software techniques exists. Dedicated some cores to run progress threads is one of them. However, some CPUs provide Simultaneous Multi-Threading, which is the ability for a core to have multiple hardware threads running simultaneously, sharing the same arithmetic units. Our idea is to use them to run progress threads to avoid dedicated cores allocation. We have run benchmarks on Haswell processors, using its Hyper-Threading capability, and get good results for both performance and overlap only when inter-node communications are used by MPI processes. However, we also show that enabling Simultaneous Multi-Threading for intra-communications leads to bad performances due to cache effects.

show abstract

Section: Related Workmentioning

confidence: 99%

Progress Thread Placement for Overlapping MPI Non-blocking Collectives Using Simultaneous Multi-threading

Lecture Notes in Computer Science

Taboada

2018

show abstract

“…The topic of communication progression has already been studied for some aspects in the literature. Several strategies do exist for background progression of point-to-point communications, such as offloading the communication to hardware (Rashti and Afsahi, 2008; Sur et al, 2006) and let the hardware do the progression, use of a thread (Hoefler and Lumsdaine, 2008a; Ruhela et al, 2018) or process (Lai et al, 2009) dedicated to communication progression, and opportunistic scheduling of communication tasks (Denis, 2015; Si et al, 2014).…”

Section: Related Workmentioning

confidence: 99%

Study on progress threads placement and dedicated cores for overlapping MPI nonblocking collectives on manycore processor

The International Journal of High Performance Computing Applica

Jeannot

et al. 2019

To amortize the cost of MPI collective operations, nonblocking collectives have been proposed so as to allow communications to be overlapped with computation. Unfortunately, collective communications are more CPU-hungry than point-to-point communications and running them in a communication thread on a dedicated CPU core makes them slow. On the other hand, running collective communications on the application cores leads to no overlap. In this article, we propose placement algorithms for progress threads that do not degrade performance when running on cores dedicated to communications to get communication/computation overlap. We first show that even simple collective operations, such as those based on a chain topology, are not straightforward to make progress in background on a dedicated core. Then, we propose an algorithm for tree-based collective operations that splits the tree between communication cores and application cores. To get the best of both worlds, the algorithm runs the short but heavy part of the tree on application cores, and the long but narrow part of the tree on one or several communication cores, so as to get a trade-off between overlap and absolute performance. We provide a model to study and predict its behavior and to tune its parameters. We implemented both algorithms in the multiprocessor computing framework, which is a thread-based MPI implementation. We have run benchmarks on manycore processors such as the KNL and Skylake and get good results for both performance and overlap.

show abstract

“…Several strategies do exist for background progression of point-to-point communications, such as offloading the communication to hardware [13,11] and let the hardware do the progression; use of a thread [4] or process [7] dedicated to communication progression; opportunistic scheduling of communication tasks [3,12].…”

Section: Related Workmentioning

confidence: 99%

Dynamic Placement of Progress Thread for Overlapping MPI Non-blocking Collectives on Manycore Processor

Euro-Par 2018: Parallel Processing

Jeannot

et al. 2018