On Gang Scheduling and Demand Paging

نویسندگان

  • Dan C. Marinescu
  • Kuei Yu Wang
چکیده

Uniprocessor system schedulers hide the high latency of a page fault, by performing a context switch from the process or thread experiencing the fault to another process ready to run. Gang scheduling attempts to schedule all the processes in a process group at the same time. Processes in a process group do not experience page faults at precisely the same time, and process group context switching is in general fairly expensive therefore one cannot hide the latency of a page fault by means of a group context switch. This paper examlnes several scheduling strategies for systems which support both gang scheduling and demand paging and reports some measurements of the paging activity of SPMD programs on the Intel Paragon running OSF/1 under Mach. ·Work supported in part by NSF grants CCR-9119388, BIR-9301210 and MCR 9527131, by a grant from the Intel Corporation and by a grant from CNPq Brasil.

برای دانلود 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

Paging Tradeoffs in Distributed - Shared - Memory

Massively parallel processors have begun using commodity operating systems that support demand-paged virtual memory. To evaluate the utility of virtual memory, we measured the behavior of seven shared-memory parallel application programs on a simulated distributed-shared-memory machine. Our results (1) connrm the importance of gang CPU scheduling, (2) show that a page-faulting processor should ...

متن کامل

Gang Scheduling with Memory Considerations

A major problem with time slicing on parallel machines is memory pressure, as the resulting paging activity damages the synchronism among a job's processes. An alternative is to impose admission controls, and only admit jobs that fit into the available memory. Despite suffering from delayed execution, this leads to better overall performance by preventing the harmful effects of paging and thras...

متن کامل

Performance Evaluation of Gang Scheduling for Parallel and Distributed Multiprogramming

In this paper we explore the performance of various aspects of gang scheduling designs. We developed an event-driven simulator of a vanilla gang scheduler that relies on the Distributed Hierarchical Control (DHC) structure. We also developed three variations of the vanilla gang scheduler that rely on a push-down heuristic and on two job-migration schemes to decrease response times by reducing p...

متن کامل

An Efficient Video Request Scheduling Framework for Video-on-Demand Systems

Recent years have witnessed dramatic increase in computational capability and network bandwidth available to end-users in commercial, home, and mobile environments. Similarly, the field of entertainment has also seen a rapid increase in pay-per-view offerings and specialinterest satellite channels. Even the educational environment has seen offerings which provide video and audio captures of cla...

متن کامل

Comparative evaluation of the recent Linux and Solaris kernel architectures

The paper compares core kernel architecture and functionality of two modern open source systems. The subsystems examined are scheduling, memory management, and file system architecture. These subsystems are common to any operating system (not just Unix and Unix-like systems), and they tend to be the most wellunderstood components of the operating system. One of the more interesting aspects conc...

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

برای دانلود 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

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2013