Multithreaded approach for dynamic load balancing of parallel adaptive PDE computations
نویسنده
چکیده
We present a multithreaded model for the dynamic load-balancing of parallel adaptive PDE computations. Multithreading is used as a means of exploring concurrency in the processor level and for tolerating: (i) long memory latencies, and (ii) synchronization costs inherent to traditional load-balancing methods. We show that under certain values of the parameters (i.e., number of threads, and context switch time) of the model, multithreaded load-balancing systems are expected to perform better than existing software systems based on traditional load-balancing methods.
منابع مشابه
Multithreaded Model for Dynamic Load Bm.ancing Parallel Adafi'ive Pde Computations
We present a multithreaded model for the dynamic load-balancing of numerical, adaptive computations required for the solution of Partial Differential Equations (PDEs) on multiprocessors. Multithreading is used as a means of exploring concurrency at the processor level in order to tolerate synchronization costs inherent to traditional (non-threaded) parallel adaptive PDE solvers. Our preliminary...
متن کاملMultithreaded model for dynamic load balancing parallel adaptive PDE computations
We present a multithreaded model for the dynamic load-balancing of numerical, adaptive computations required for the solution of Partial Di erential Equations (PDEs) on multiprocessors. Multithreading is used as a means of exploring concurrency at the processor level in order to tolerate synchronization costs inherent to traditional (non-threaded) parallel adaptive PDE solvers. Our preliminary ...
متن کاملMultithreaded model for dynamic load balancing parallel adaptive PDE computations
An EEcient thread-to-thread communication for hybrid shared/distributed address space programming paradigms, to be submitted to IEEE Trans. Parallel and Distributed Computing. 36] Juan Miguel del Rosario and N.P. Chrisochoides, An interrupt driven implementation of thread-to-thread communication for distributed address space machines, To be submitted to IEEE Trans. Parallel and Distributed Comp...
متن کاملParleda: 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 کاملA Massive Multilevel-parallel Microscopic Traffic Simulator with Gridlock Detection and Solving
Traffic simulators based on microscopic models are a detailed approach to infrastructure and policy evaluation. They allow a closer to reality representation of the factors that influence traffic flow: individual driver and vehicle attributes, dynamic route decisions, lane changing and restrictions, driver cooperation. Such aspects add to the complexity and volume of computations, leading to sl...
متن کامل