One Address Computers are Faster and Use Less Memory Space to Execute Arithmetic Assignment Statements
نویسندگان
چکیده
A notation is developed which permits space and time efficiemcy comparisons of four basic computer architectures in use today for executing Fortran-style assignment statements. From the com~arisona, we discover that a suitably designed l-address archit.cture (one accumulator machine) outperforms toe other architectures in speed of execution and in encoded size ~ compiled Fortran statements. The comparisons are valid for CPU's ranging from very inexpensive designs with few registers to the most expensive designs having many registers and employing "pipelining" teohniques or lookahead fetches of operands or instructions into fast cache memories.
منابع مشابه
Evolving the UNIX System Interface to Support Multithreaded Programs
Allowing multiple threads to execute within the same address space makes it easier to write programs that deal with related asynchronous activities and that execute faster on shared-memory multiprocessors. Supporting multiple threads places new constraints on the design of operating system interfaces. We present several guidelines for designing or redesigning interfaces for multithreaded client...
متن کاملGenerating Global Name-space Communication Sets for Array Assignment Statements 1
This paper is concerned with the design of eecient algorithms for generating global name-space communication sets based on execution of array assignment statements on distributed-memory parallel computers. For general cases, although the communication sets can be represented by the union of a functional number of closed forms, these sets cannot be represented by a xed number of closed forms. Cl...
متن کاملCase Studies and Exercises Case Study 1: Optimizing Cache Performance via Advanced Techniques Non-blocking Caches Compiler Optimizations for Caches Calculating Impact of Cache Performance on More Complex Processors Historical Perspective and References Case Studies and Exercises
parallelism took this a step further by providing more parallelism and hence more latency-hiding opportunities. It is likely that the use of instructionand threadlevel parallelism will be the primary tool to combat whatever memory delays are encountered in modern multilevel cache systems. One idea that periodically arises is the use of programmer-controlled scratchpad or other high-speed memori...
متن کاملAnalysis of Partitioned Global Address Space Programs
The introduction of multi-core processors by the major microprocessor vendors has brought parallel programming into the mainstream. Analysis of parallel languages is critical both for safety and optimization purposes. In this report, we consider the specific case of languages with barrier synchronization and global address space abstractions. Two of the fundamental problems in the analysis of p...
متن کاملAn Effective Heuristic for Simple Offset Assignment with Variable Coalescing
In many Digital Signal Processors (DSPs) with limited memory, programs are loaded in the ROM and thus it is very important to optimize the size of the code to reduce the memory requirement. Many DSP processors include address generation units (AGUs) that can perform address arithmetic (auto-increment and auto-decrement) in parallel to instruction execution, and without the need for extra instru...
متن کامل