The management of parallelism, the simultaneous managementofmultiple environments, and the synchronization of cooperating threads are some of the fundamental issues that need to be solved inamultithreaded pipelined architecture. In this paper, we present several di erent solutions to these problems which are currently implemented in a cycle-accurate software simulator of a multithreaded architecture called the SAM architecture. The simulator is a software prototype of achip we expect to design and build within the coming year. The architecture contains two special instructions that allow parallelism to vary with time, and a two-level thread queue that allows the degree of parallelism to expand, theoretically, without limit, while still allo...
Existing multiprocessor synchronization mechanisms are relatively heavyweight, due in part to the le...
Computer architects heavily rely on software simulation to evaluate new and existing processor desig...
To achieve high performance, contemporary computer systems rely on two forms of parallelism: instruc...
Modem processors are designed to achieve greater amounts of instruction level parallelism (ILP) and ...
Performance improvements for microprocessors have traditionally been achieved by increasing their cl...
A multithreaded architecture exploits instruction level parallelism by interleaving instructions fr...
The architecture of future high performance computer systems will respond to the possibilities offer...
This paper analyzes the basic design issues of multithreaded processors and discusses how they may r...
To achieve high performance, contemporary computer systems rely on two forms of parallelism: instruc...
Abstract 1 This paper discusses the preliminary performance study of hybrid multithreaded execution ...
This thesis proposes, develops, and evaluates hardware and software mechanisms that enhance the effi...
The performance of a concurrent multithreaded architectural model, called superthreading [15), is st...
This paper describes initial results for an architecture called the Shared-Thread Multiprocessor (ST...
Present-day parallel computers often face the problems of large software overheads for process switc...
This dissertation focuses on design and implementation issues of a multithreaded parallel programmin...
Existing multiprocessor synchronization mechanisms are relatively heavyweight, due in part to the le...
Computer architects heavily rely on software simulation to evaluate new and existing processor desig...
To achieve high performance, contemporary computer systems rely on two forms of parallelism: instruc...
Modem processors are designed to achieve greater amounts of instruction level parallelism (ILP) and ...
Performance improvements for microprocessors have traditionally been achieved by increasing their cl...
A multithreaded architecture exploits instruction level parallelism by interleaving instructions fr...
The architecture of future high performance computer systems will respond to the possibilities offer...
This paper analyzes the basic design issues of multithreaded processors and discusses how they may r...
To achieve high performance, contemporary computer systems rely on two forms of parallelism: instruc...
Abstract 1 This paper discusses the preliminary performance study of hybrid multithreaded execution ...
This thesis proposes, develops, and evaluates hardware and software mechanisms that enhance the effi...
The performance of a concurrent multithreaded architectural model, called superthreading [15), is st...
This paper describes initial results for an architecture called the Shared-Thread Multiprocessor (ST...
Present-day parallel computers often face the problems of large software overheads for process switc...
This dissertation focuses on design and implementation issues of a multithreaded parallel programmin...
Existing multiprocessor synchronization mechanisms are relatively heavyweight, due in part to the le...
Computer architects heavily rely on software simulation to evaluate new and existing processor desig...
To achieve high performance, contemporary computer systems rely on two forms of parallelism: instruc...