An analysis of VI Architecture primitives in support of parallel and distributed communication

نویسندگان

  • Andrew Begel
  • Philip Buonadonna
  • David E. Culler
  • David Gay
چکیده

We present the results of a detailed study of the Virtual Interface (VI) paradigm as a communication foundation for a distributed computing environment. Using Active Messages and the Split-C global memory model, we analyze the inherent costs of using VI primitives to implement these highlevel communication abstractions. We demonstrate a minimum mapping cost (i.e. the host processing required to map one abstraction to a lower abstraction) of 5.4 μsec for both Active Messages and Split-C using 4-way 550 MHz Pentium III SMPs and the Myrinet network. We break down this cost to use of individual VI primitives in supporting flow control, buffer management and event processing and identify the completion queue as the source of the highest overhead. Bulk transfer performance plateaus at 44 Mbytes/sec for both implementations due to the addition of fragmentation requirements. Based on this analysis, we present the implications for the VI successor, Infiniband.

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

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Communication Optimizations for Irregular Scientiic Computations on Distributed Memory Architectures 3

This paper describes a number of optimizations that can be used to support the eecient execution of irregular problems on distributed memory parallel machines. These primitives (1) coordinate inter-processor data movement, (2) manage the storage of, and access to, copies of oo-processor data, (3) minimize interprocessor communication requirements and (4) support a shared name space. We present ...

متن کامل

Process algebraic modeling of authentication protocols for analysis of parallel multi-session executions

Many security protocols have the aim of authenticating one agent acting as initiator to another agent acting as responder and vice versa. Sometimes, the authentication fails because of executing several parallel sessions of a protocol, and because an agent may play both the initiator and responder role in parallel sessions. We take advantage of the notion of transition systems to specify authen...

متن کامل

Parleda: a Library for Parallel Processing in Computational Geometry Applications

ParLeda is a software library that provides the basic primitives needed for parallel implementation of computational geometry applications. It can also be used in implementing a parallel application that uses geometric data structures. The parallel model that we use is based on a new heterogeneous parallel model named HBSP, which is based on BSP and is introduced here. ParLeda uses two main lib...

متن کامل

Communication Primitives for Unstructured Finite Element Simulations on Data Parallel Architectures Communication Primitives for Unstructured Finite Element Simulations on Data Parallel Architectures

EEcient data motion is critical for high performance computing on distributed memory architectures. The value of some techniques for eecient data motion is illustrated by identifying generic communication primitives. Further, the eeciency of these primitives is demonstrated on three diier-ent applications using the nite element method for unstructured grids and sparse solvers with diierent comm...

متن کامل

Architectural Support and Mechanisms for Object Caching in Dynamic Multithreaded Computations

High-level parallel programming models supporting dynamic fine-grained threads in a global object space are becoming increasingly popular for expressing irregular applications based on sophisticated adaptive algorithms and pointer-based data structures. However, implementing these multithreaded computations on scalable parallel machines poses significant challenges, particularly with respect to...

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

عنوان ژورنال:
  • Concurrency and Computation: Practice and Experience

دوره 14  شماره 

صفحات  -

تاریخ انتشار 2002