Task4: Mapping and Load-balancing Summary of Work Completed (october '95 -march '96) Dynamic Load Balancing

ثبت نشده
چکیده

The implementation and testing of the mapping tool box has been successfully achieved during the last period. The work on mapping consists now mainly in integrating the mapping tool box with the rest of the Sepp environment. The undergoing integration is described in the last HPCTI report. In the framework of the Sepp project, partners of the task concentrated mainly on dynamic load balancing tool. Several approaches for load balancing schedulers and algorithms are currently being explored: A semi-distributed scheduler for dynamic load balancing based on PVM internal facilities (resource manager) has been designed and partially implemented 1]. The scheduler is responsible for making decisions about the locations of new processes taking into account the load and communication overheads. A rst version is under implementation and should be available in May 96. This approach will be validated on a modiied version of WaTor (an irregular distributed simulation problem which involves complex process creation and terminaison) A rst version of the Diiusion Method has been implemented. This method solves the problem of global unbalance by taking into account the maximal diierence of load between processors of the same local domain , providing an eecient load-balanced solution 3]. The current prototype running on IBM SP-1 has been implemented in a centralised version. Tests are currently in progress to validate this approach. A parallel programming platform, Athapascan 2], providing the func-tionalities and eeciency of a thread kernel and the Pvm library was implemented and used for the development of real-size applications. The concepts underlying this approach, i.e. the remote procedure call (RPC) With contributions from the partners of the task: L.

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

Load Balancing Approaches for Web Servers: A Survey of Recent Trends

Numerous works has been done for load balancing of web servers in grid environment. Reason behinds popularity of grid environment is to allow accessing distributed resources which are located at remote locations. For effective utilization, load must be balanced among all resources. Importance of load balancing is discussed by distinguishing the system between without load balancing and with loa...

متن کامل

Online Distribution and Load Balancing Optimization Using the Robin Hood and Johnson Hybrid Algorithm

Proper planning of assembly lines is one of the production managers’ concerns at the tactical level so that it would be possible to use the machine capacity, reduce operating costs and deliver customer orders on time. The lack of an efficient method in balancing assembly line can create threatening problems for manufacturing organizations. The use of assembly line balancing methods cannot balan...

متن کامل

Observations on Using Genetic Algorithms for Dynamic Load-Balancing

ÐLoad-balancing problems arise in many applications, but, most importantly, they play a special role in the operation of parallel and distributed computing systems. Load-balancing deals with partitioning a program into smaller tasks that can be executed concurrently and mapping each of these tasks to a computational resource such a processor (e.g., in a multiprocessor system) or a computer (e.g...

متن کامل

Mapping and load balancing on distributed memory systems

Two kinds of tools are necessary to optimise the use of available resources by the execution of parallel programs on distributed memory systems: mapping and load balancing tools. A mapping tool is well suited for programs whose behaviour is predictable while for many "real applications", it needs to be complemented by a dynamic load balancing tool. Both tools are currently investigated to be in...

متن کامل

Tools for Mapping, Load Balancing and Monitoring in the LOGFLOW Parallel Prolog Project

LOGFLOW is an all-solution parallel logic programming system able to exploit OR-parallelism and pipeline AND-parallelism of Prolog programs. The LOGFLOW project is intended to implement Prolog in massively parallel distributed memory multicomputers. Porting LOGFLOW to a workstation cluster resulted in a variant of LOGFLOW called WS-LOGFLOW. Implementation of LOGFLOW both on multi-transputers an...

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 1996