Optimal scheduling of arithmetic operations in parallel with memory access (preliminary version)

Authors:
David Bernstein;Ron Y. Pinter;Michael Rodeh
Affiliations:
Dept. of Electrical Engineering, Technion - Israel Institute of Technology, Haifa 32000, ISRAEL;IBM Israel Scientific Center, Technion City, Haifa 32000, ISRAEL;IBM Israel Scientific Center, Technion City, Haifa 32000, ISRAEL
Venue:
POPL '85 Proceedings of the 12th ACM SIGACT-SIGPLAN symposium on Principles of programming languages
Year:
1985

Citing 6
Cited 6

The Generation of Optimal Code for Arithmetic Expressions

Journal of the ACM (JACM)
Optimal Code Generation for Expression Trees

Journal of the ACM (JACM)
Graph Algorithms

Graph Algorithms
Data Structures and Algorithms

Data Structures and Algorithms
Decoupled access/execute computer architectures

ISCA '82 Proceedings of the 9th annual symposium on Computer Architecture
SCHEDULING TASK SYSTEMS WITH RESOURCES

SCHEDULING TASK SYSTEMS WITH RESOURCES

Optimal chaining in expression trees

SIGPLAN '86 Proceedings of the 1986 SIGPLAN symposium on Compiler construction
Scheduling arithmetic and load operations in parallel with no spilling

POPL '87 Proceedings of the 14th ACM SIGACT-SIGPLAN symposium on Principles of programming languages
Optimal Chaining in Expression Trees

IEEE Transactions on Computers
On the Complexity of Scheduling Problems for Parallel/Pipelined Machines

IEEE Transactions on Computers
Linear-time, optimal code scheduling for delayed-load architectures

PLDI '91 Proceedings of the ACM SIGPLAN 1991 conference on Programming language design and implementation
Efficient instruction scheduling for delayed-load architectures

ACM Transactions on Programming Languages and Systems (TOPLAS)

Quantified Score

Hi-index	0.01

Visualization

Abstract

We propose a new machine model in which load operations can be performed in parallel with arithmetic operations by two separate functional units. For this model, the evaluation of expression trees is considered. An efficient algorithm to produce an optimal order of evaluation is described and analyzed. For a tree with n vertices the algorithm runs in time &Ogr;(n log2n). If the arithmetic operations have at most two arguments, the complexity goes down to &Ogr;(n logn).