Abstract: A Simultaneous Multithreading (SMT) Processor is
capable of executing instructions from multiple threads in the same
cycle. SMT in fact was introduced as a powerful architecture to
superscalar to increase the throughput of the processor.
Simultaneous Multithreading is a technique that permits multiple
instructions from multiple independent applications or threads to
compete limited resources each cycle. While the fetch unit has been
identified as one of the major bottlenecks of SMT architecture, several
fetch schemes were proposed by prior works to enhance the fetching
efficiency and overall performance.
In this paper, we propose a novel fetch policy called queue situation
identifier (QSI) which counts some kind of long latency instructions of
each thread each cycle then properly selects which threads to fetch
next cycle. Simulation results show that in best case our fetch policy
can achieve 30% on speedup and also can reduce the data cache level 1
miss rate.
Abstract: The Partitioned Global Address Space (PGAS) programming
paradigm offers ease-of-use in expressing parallelism
through a global shared address space while emphasizing performance
by providing locality awareness through the partitioning of
this address space. Therefore, the interest in PGAS programming
languages is growing and many new languages have emerged and
are becoming ubiquitously available on nearly all modern parallel
architectures. Recently, new parallel machines with multiple cores
are designed for targeting high performance applications. Most of the
efforts have gone into benchmarking but there are a few examples of
real high performance applications running on multicore machines.
In this paper, we present and evaluate a parallelization technique
for implementing a local DNA sequence alignment algorithm using
a PGAS based language, UPC (Unified Parallel C) on a chip
multithreading architecture, the UltraSPARC T1.
Abstract: The policies governing the business of any
organization are well reflected in her business rules. The business
rules are implemented by data validation techniques, coded during
the software development process. Any change in business
policies results in change in the code written for data validation
used to enforce the business policies. Implementing the change in
business rules without changing the code is the objective of this
paper. The proposed approach enables users to create rule sets at
run time once the software has been developed. The newly defined
rule sets by end users are associated with the data variables for
which the validation is required. The proposed approach facilitates
the users to define business rules using all the comparison
operators and Boolean operators. Multithreading is used to
validate the data entered by end user against the business rules
applied. The evaluation of the data is performed by a newly
created thread using an enhanced form of the RPN (Reverse Polish
Notation) algorithm.
Abstract: This research simulates one of the natural phenomena,
the ocean wave. Our goal is to be able to simulate the ocean wave at
real-time rate with the water surface interacting with objects. The
wave in this research is calm and smooth caused by the force of the
wind above the ocean surface. In order to make the simulation of the
wave real-time, the implementation of the GPU and the
multithreading techniques are used here. Based on the fact that the
new generation CPUs, for personal computers, have multi cores, they
are useful for the multithread. This technique utilizes more than one
core at a time. This simulation is programmed by C language with
OpenGL. To make the simulation of the wave look more realistic, we
applied an OpenGL technique called cube mapping (environmental
mapping) to make water surface reflective and more realistic.
Abstract: The shortest path question is in a graph theory model
question, and it is applied in many fields. The most short-path
question may divide into two kinds: Single sources most short-path,
all apexes to most short-path. This article mainly introduces the
problem of all apexes to most short-path, and gives a new parallel
algorithm of all apexes to most short-path according to the Dijkstra
algorithm. At last this paper realizes the parallel algorithms in the
technology of C # multithreading.
Abstract: In this paper, we consider components of discrete event imitating model, implementing a simulation model by using JAVA and performing an input analysis of the data and an output analysis of the simulation results. Was lead development of imitating model of mass service system with n (n≥1) devices of service. On the basis of the developed process of a multithreading simulated the distributed processes with presence of synchronization. Was developed the algorithm of event-oriented simulation, was received results of system functioning with n devices of service.