Abstract
| Original language | Undefined |
|---|---|
| Pages (from-to) | 117-128 |
| Number of pages | 12 |
| Journal | Structured programming |
| Volume | 11 |
| Issue number | 117 |
| Publication status | Published - 1990 |
Keywords
- EWI-1224
- IR-56006
Cite this
}
A comparison of three garbage collection algorithms. / Hartel, Pieter H.
In: Structured programming, Vol. 11, No. 117, 1990, p. 117-128.Research output: Contribution to journal › Article › Academic › peer-review
TY - JOUR
T1 - A comparison of three garbage collection algorithms
AU - Hartel, Pieter H.
N1 - Imported from DIES
PY - 1990
Y1 - 1990
N2 - Abstract. The running cost of garbage collection is studied as a function of the amount of available store. A performance model originally proposed by Hoare is modified to support experiments with three garbage collection methods: reference count, mark/scan and two-space copy. By also taking the boundary effects into account, we show that adding more store to a list processing system with a non-reference counting garbagecollector does not always make it run faster. We present an explanation for this anomaly in the behaviour of garbage collection algorithms and discuss some of its consequences.
AB - Abstract. The running cost of garbage collection is studied as a function of the amount of available store. A performance model originally proposed by Hoare is modified to support experiments with three garbage collection methods: reference count, mark/scan and two-space copy. By also taking the boundary effects into account, we show that adding more store to a list processing system with a non-reference counting garbagecollector does not always make it run faster. We present an explanation for this anomaly in the behaviour of garbage collection algorithms and discuss some of its consequences.
KW - EWI-1224
KW - IR-56006
M3 - Article
VL - 11
SP - 117
EP - 128
JO - Structured programming
JF - Structured programming
SN - 0935-1183
IS - 117
ER -