Lamport Leslie. On-the-fly garbage collection: Once more with rigor. Technical Report CA-7508-1611, Computer Associates, Wakefield, MA, August 1975. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
List Processing in Real Time on a Serial Computer - Baker, Jr. (1978) (175 citations) (Correct)
....first to consider the problem, and sketches a multiprogramming solution in which the garbage collector shares time with the main list processing program. Steele s paper [30] was the first in a flurry of papers about multiprocessing garbage collection which included contributions by [16,17] and [23,24]. 28] independently detailed the Minsky Knuth Steele method, and both [28] and [33] analyzed the time and storage required to make it work. The Minsky Knuth Steele Muller Wadler (MKSMW) method for real time garbage collection has two processes running in parallel. The list processor process is ....
....bounds for nonequilibrium situations so long as the ratios of the rate of CONSING to the rates of marking, sweeping, and relocating are constant; i.e. we relativize the rates of marking, sweeping, and relocating with respect to a CONS counter rather than a clock. The Dijkstra Lamport (DL) method [16,17,23,24] also has the mutator and collector running in parallel, but the collector uses no stack. It marks by scanning all of storage for a mark bit it can propagate to the marked cell s offspring. This simple method of garbage collection was considered because their concern was proving that the collector ....
[Article contains additional citation context not shown here]
Lamport, L. On-the-fly garbage collection: Once more with rigor. CA-7508-1611, Mass. Computer Associates, Wakefield, Mass., Aug. 1975.
The Incremental Garbage Collection of Processes - Baker, Hewitt (1977) (29 citations) (Correct)
....Hence it may not be collected even if it becomes irrelevant. This is one reason why busy waiting is not a good synchronization method. Garbage collection is made incremental by using some of the ideas from an earlier paper [1] which in turn is based on the work of Dijkstra [5,6] and Lamport [14,15]. The mark phase of our incremental garbage collector process employs three colors for every object white, grey, and black. Intuitively, white nodes are not yet known to be accessible, grey nodes are known to be accessible, but whose offspring have not yet been checked, and black nodes are ....
....by the garbage collector, it must stop and allow itself to be colored black before continuing. The notion that processes must be marked as well as storage may explain some of the trouble that Dijkstra and Lamport had when trying to prove their parallel garbage collection algorithm correct [5,6,14,15]. Since their algorithm does not mark a user process by coloring it black (thereby prohibiting it from directly touching white nodes) and allows these white processes to run, the proof that the algorithm collects only and all garbage is long and very subtle (see [15] 3. Coroutines and ....
[Article contains additional citation context not shown here]
Lamport, L. "On-the-fly Garbage Collection: Once More with Rigor". Mass. Comp. Associates, CA-7508-1611, Aug. 1975.
the Garbage Collection Bibliography - Richard Jones (2003) (Correct)
No context found.
Lamport Leslie. On-the-fly garbage collection: Once more with rigor. Technical Report CA-7508-1611, Computer Associates, Wakefield, MA, August 1975.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC