The PetscSF Scalable Communication Layer

نویسندگان

چکیده

PetscSF, the communication component of Portable, Extensible Toolkit for Scientific Computation (PETSc), is designed to provide PETSc's infrastructure suitable exascale computers that utilize GPUs and other accelerators. PetscSF provides a simple application programming interface (API) managing common patterns in scientific computations by using star-forest graph representation. supports several implementations based on MPI NVSHMEM, whose selection characteristics or target architecture. An efficient portable model network intra-node essential implementing large-scale applications. The Message Passing Interface, which has been de facto standard distributed memory systems, developed into large complex API does not yet high performance emerging heterogeneous CPU-GPU-based systems. In this article, we discuss design how it can overcome some difficulties working directly with GPUs, demonstrate its performance, scalability, novel features.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Scalable Tools Communication Infrastructure

The Scalable Tools Communication Infrastructure (STCI) is an open source collaborative effort intended to provide high-performance, scalable, resilient, and portable communications and process control services for a wide variety of user and system tools. STCI is aimed specifically at tools for ultrascale computing and uses a component architecture to simplify tailoring the infrastructure to a w...

متن کامل

A Protocol for Scalable Application Layer Multicast

We describe a new application-layer multicast protocol that is speciically designed to scale to large groups. Our scheme is based upon a hierarchical clustering of the application-layer multicast peers and can be used to produce a number of diierent data delivery trees with speciic properties. On average, group members using our protocol maintain only a constant amount of state about other grou...

متن کامل

Scalable Distributed Two-Layer Data Structures (SD2DS)

Scalability and fault tolerance are important features of modern applications designed for the distributed, loosely-coupled computer systems. In the paper, two-layer scalable structures for storing data in a distributed RAM of a multicomputer (SD2DS) are introduced. A data unit of SD2DS (a component) is split into a header and a body. The header identifies the body and contains its address in a...

متن کامل

Shape Approximation Through Recursive Scalable Layer Generation

This paper presents an e cient recursive algorithm for generating operationally optimal intra mode scalable layer decompositions of object contours. The problem is posed in terms of minimizing the shape distortion at full reconstruction subject to the total (for all scalable layers) bit budget constraint. Based on the chosen vertex-based representation, we solve the problem of determining the n...

متن کامل

Layer thickness in congestion-controlled scalable video

We address the problem of the proper choice of the thickness of pre-encoded video layers in congestion-controlled streaming applications. While congestion control permits to distribute the network resources in a fair manner among the different video sessions, it generally imposes an adaptation of the streaming rate when the playback delay is constrained. This can be achieved by adding or droppi...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Transactions on Parallel and Distributed Systems

سال: 2022

ISSN: ['1045-9219', '1558-2183', '2161-9883']

DOI: https://doi.org/10.1109/tpds.2021.3084070