Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userGPU-aware Communication with UCX in Parallel Programming Models: Charm++, MPI, and Python
82
0
0.0
(
0
)
Added by
Jaemin Choi
Publication date
2021
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
As an increasing number of leadership-class systems embrace GPU accelerators in the race towards exascale, efficient communication of GPU data is becoming one of the most critical components of high-performance computing. For developers of parallel programming models, implementing support for GPU-aware communication using native APIs for GPUs such as CUDA can be a daunting task as it requires considerable effort with little guarantee of performance. In this work, we demonstrate the capability of the Unified Communication X (UCX) framework to compose a GPU-aware communication layer that serves multiple parallel programming models of the Charm++ ecosystem: Charm++, Adaptive MPI (AMPI), and Charm4py. We demonstrate the performance impact of our designs with microbenchmarks adapted from the OSU benchmark suite, obtaining improvements in latency of up to 10.2x, 11.7x, and 17.4x in Charm++, AMPI, and Charm4py, respectively. We also observe increases in bandwidth of up to 9.6x in Charm++, 10x in AMPI, and 10.5x in Charm4py. We show the potential impact of our designs on real-world applications by evaluating a proxy application for the Jacobi iterative method, improving the communication performance by up to 12.4x in Charm++, 12.8x in AMPI, and 19.7x in Charm4py.
rate research
Read More
High-level programming languages such as Python are increasingly used to provide intuitive interfaces to libraries written in lower-level languages and for assembling applications from various components. This migration towards orchestration rather than implementation, coupled with the growing need for parallel computing (e.g., due to big data and the end of Moores law), necessitates rethinking how parallelism is expressed in programs. Here, we present Parsl, a parallel scripting library that augments Python with simple, scalable, and flexible constructs for encoding parallelism. These constructs allow Parsl to construct a dynamic dependency graph of components that it can then execute efficiently on one or many processors. Parsl is designed for scalability, with an extensible set of executors tailored to different use cases, such as low-latency, high-throughput, or extreme-scale execution. We show, via experiments on the Blue Waters supercomputer, that Parsl executors can allow Python scripts to execute components with as little as 5 ms of overhead, scale to more than 250 000 workers across more than 8000 nodes, and process upward of 1200 tasks per second. Other Parsl features simplify the construction and execution of composite programs by supporting elastic provisioning and scaling of infrastructure, fault-tolerant execution, and integrated wide-area data management. We show that these capabilities satisfy the needs of many-task, interactive, online, and machine learning applications in fields such as biology, cosmology, and materials science.
In this paper we present the Task-Aware MPI library (TAMPI) that integrates both blocking and non-blocking MPI primitives with task-based programming models. The TAMPI library leverages two new runtime APIs to improve both programmability and performance of hybrid applications. The first API allows to pause and resume the execution of a task depending on external events. This API is used to improve the interoperability between blocking MPI communication primitives and tasks. When an MPI operation executed inside a task blocks, the task running is paused so that the runtime system can schedule a new task on the core that became idle. Once the blocked MPI operation is completed, the paused task is put again on the runtime systems ready queue, so eventually it will be scheduled again and its execution will be resumed. The second API defers the release of dependencies associated with a task completion until some external events are fulfilled. This API is composed only of two functions, one to bind external events to a running task and another function to notify about the completion of external events previously bound. TAMPI leverages this API to bind non-blocking MPI operations with tasks, deferring the release of their task dependencies until both task execution and all its bound MPI operations are completed. Our experiments reveal that the enhanced features of TAMPI not only simplify the development of hybrid MPI+OpenMP applications that use blocking or non-blocking MPI primitives but they also naturally overlap computation and communication phases, which improves application performance and scalability by removing artificial dependencies across communication tasks.
Due to the increasing size of HPC machines, the fault presence is becoming an eventuality that applications must face. Natively, MPI provides no support for the execution past the detection of a fault, and this is becoming more and more constraining. With the introduction of ULFM (User Level Fault Mitigation library), it has been provided with a possible way to overtake a fault during the application execution at the cost of code modifications. ULFM is intrusive in the application and requires also a deep understanding of its recovery procedures. In this paper we propose Legio, a framework that lowers the complexity of introducing resiliency in an embarrassingly parallel MPI application. By hiding ULFM behind the MPI calls, the library is capable to expose resiliency features to the application in a transparent manner thus removing any integration effort. Upon fault, the failed nodes are discarded and the execution continues only with the non-failed ones. A hierarchical implementation of the solution has been also proposed to reduce the overhead of the repair process when scaling towards a large number of nodes. We evaluated our solutions on the Marconi100 cluster at CINECA, showing that the overhead introduced by the library is negligible and it does not limit the scalability properties of MPI. Moreover, we also integrated the solution in real-world applications to further prove its robustness by injecting faults.
Analytic, first-principles performance modeling of distributed-memory parallel codes is notoriously imprecise. Even for applications with extremely regular and homogeneous compute-communicate phases, simply adding communication time to computation time does often not yield a satisfactory prediction of parallel runtime due to deviations from the expected simple lockstep pattern caused by system noise, variations in communication time, and inherent load imbalance. In this paper, we highlight the specific cases of provoked and spontaneous desynchronization of memory-bound, bulk-synchronous pure MPI and hybrid MPI+OpenMP programs. Using simple microbenchmarks we observe that although desynchronization can introduce increased waiting time per process, it does not necessarily cause lower resource utilization but can lead to an increase in available bandwidth per core. In case of significant communication overhead, even natural noise can shove the system into a state of automatic overlap of communication and computation, improving the overall time to solution. The saturation point, i.e., the number of processes per memory domain required to achieve full memory bandwidth, is pivotal in the dynamics of this process and the emerging stable wave pattern. We also demonstrate how hybrid MPI-OpenMP programming can prevent desirable desynchronization by eliminating the bandwidth bottleneck among processes. A Chebyshev filter diagonalization application is used to demonstrate some of the observed effects in a realistic setting.
We investigate a parallelization strategy for dense matrix factorization (DMF) algorithms, using OpenMP, that departs from the legacy (or conventional) solution, which simply extracts concurrency from a multithreaded version of BLAS. This approach is also different from the more sophisticated runtime-assisted implementations, which decompose the operation into tasks and identify dependencies via directives and runtime support. Instead, our strategy attains high performance by explicitly embedding a static look-ahead technique into the DMF code, in order to overcome the performance bottleneck of the panel factorization, and realizing the trailing update via a cache-aware multi-threaded implementation of the BLAS. Although the parallel algorithms are specified with a highlevel of abstraction, the actual implementation can be easily derived from them, paving the road to deriving a high performance implementation of a considerable fraction of LAPACK functionality on any multicore platform with an OpenMP-like runtime.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
