Garbage collection is fast, but a stack is faster

Garbage collection is fast, but a stack is faster (1994)

by James S. Miller , Guillermo J. Rozas

Venue:	Tech. Rep. Memo 1462, MIT Artificial Intelligence Laboratory
Citations:	13 - 0 self

Datum	Value	Source
TITLE	Garbage collection is fast, but a stack is faster	INFERENCE
AUTHOR NAME	James S. Miller	SVM HeaderParse 0.1
AUTHOR AFFIL	and	SVM HeaderParse 0.2
AUTHOR NAME	Guillermo J. Rozas	SVM HeaderParse 0.1
ABSTRACT	This publication can be retrieved by anonymous ftp to publications.ai.mit.edu. The pathname for this publication is: ai-publications/1994/AIM-1462.ps.Z Prompted by claims that garbage collection can outperform stack allocation when su cient physical memory is available, we present a careful analysis and set of cross-architecture measurements comparing these two approaches for the implementation of continuation (procedure call) frames. When the frames are allocated on a heap they require additional space, increase the amount of data transferred between memory and registers, and, on current architectures, require more instructions. We nd that stack allocation of continuation frames outperforms heap allocation in some cases by almost a factor of three. Thus, stacks remain an important implementation technique for procedure calls, even in the presence of an e cient, compacting garbage collector and large amounts of memory.	SVM HeaderParse 0.1
YEAR	1994	INFERENCE
VENUE	Tech. Rep. Memo 1462, MIT Artificial Intelligence Laboratory	INFERENCE
VENUE TYPE	TECHREPORT	INFERENCE
TECH	Technical Report Memo 1462	INFERENCE
CITATIONS	0 found