MTIO - A Multi-Threaded Parallel I/O System

نویسندگان

  • Sachin More
  • Alok N. Choudhary
  • Ian T. Foster
  • Ming Q. Xu
چکیده

This paper presents the design and evaluation of a multithreaded runtime library for parallel I/O. We extend the multi-threading concept to separate the compute and I/O tasks in two separate threads of control. Multi-threading in our design permits a) asynchronous I/O even if the underlying file system does not support asynchronous I/O; b) copy avoidance from the I/O thread to the compute thread by sharing address space; and c) a capability to perform collective I/O asynchronously without blocking the compute threads. Further, this paper presents techniques for collective I/O which maximize load balance and concurrency while reducing communication overhead in an integrated fashion. Performance results on IBM SP2 for various data distributions and access patterns are presented. The results show that there is a tradeoff between the amount of concurrency in I/O and the buffer size designated for I/O; and there is an optimal buffer size beyond which benefits of larger requests diminish due to large communication overheads.

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

Modeling Multi-thre(aded Architectures in PAMELA for Real-time

In this paper we present a method t o explore the design space of multi-threaded archatectures using the PAMELA [3] modeling language. The domain of applications we consider i s digital signal processing (DSP), where high performance i s derived by exploiting both f ine grain and coarse grain parallelism in the application. The modeling scheme takes a n unified view of both f ine grain and coar...

متن کامل

Case Study Resolving the Challenges of CAD Format Conversion Parallelism Brings CAD Exchanger* Software Dramatic Gains in Performance and User Satisfaction, Plus a Competitive Advantage

" CAD Exchanger* is broadly using multi-threaded algorithms to increase performance on multi-core systems, " said Roman Lygin of CADEX, Ltd. " This is the key advantage over our competitors. " Benchmarks show how it outperforms earlier editions in significant ways: • Some heavyweight computational algorithms, such as blended surface approximation, were accelerated by 15X over single-thread mode...

متن کامل

Object{oriented Distributed and Parallel I/o Streams

Writing programs for parallel and distributed computing environments can be signiicantly more complex than writing programs for their sequential counterparts. These complexities mainly arise from the additional synchronization and communication requirements imposed by such environments. These requirements also make debugging and maintaining such programs sig-niicantly more complicated. The prob...

متن کامل

On the Scalability of Data Reduction Techniques in Current and Upcoming HPC Systems from an Application Perspective

We implement and benchmark parallel I/O methods for the fully-manycore driven particle-in-cell code PIConGPU. Identifying throughput and overall I/O size as a major challenge for applications on today’s and future HPC systems, we present a scaling law characterizing performance bottlenecks in state-of-the-art approaches for data reduction. Consequently, we propose, implement and verify multi-th...

متن کامل

A Study of a Continuation-based Fine-grain Multithreaded Operating System CEFOS

CEFOS is an operating system based on a continuationbased zero-wait thread model derived from a data-flow computing model. A program consists of zero-wait threads, each of which runs to completion without suspension once started. Synchronization between zero-wait threads is autonomously performed in a dataflow manner according to their continuation relations. Handler routines for asynchronous e...

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:

دوره   شماره 

صفحات  -

تاریخ انتشار 1997