| 
|
QuoteRef: blacSM6_2004
Topic:
memory management by garbage collection
Topic:
memory cache
Topic:
critical regions
Topic:
memory management by free list
Topic:
memory management by reference counting
Topic:
heap memory management
Reference
Blackburn, S.M., Cheng, P., McKinley, K.S., "Myths and realities: the performance impact of garbage collection", Proceedings of the Joint International Conference on Measurement and Modeling of Computer Systems, New York, New York, USA, June 2004, ACM SIGMETRICS, pp. 25-36. Google
25 ;;Quote: comparative study of semi-space, mark-sweek, reference counting, and generational garbage collection; direct and indirect costs of mutator, collector, L1 and L2 cache, TLB misses
25 ;;Quote: contiguous allocation important for cache performance of garbage collection; generational, copying collection is best
26 ;;Quote: generational collectors are better than whole heap collectors in almost all circumstances; lower collection time; write barrier is reasonable
26 ;;Quote: generational collectors perform similarly; use mark-sweep mature space in small heaps
26+;;Quote: generational collectors provide good spatial locality for young objects; mature objects have good temporal locality
26 ;;Quote: nursery size for generational collectors should be well above the L2 cache size
32 ;;Quote: free-list allocation has poor spatial locality; contiguous allocation for young objects is better; use free-list for mature objects
32+;;Quote: generational collectors improving relative to explicit memory management; good use of spatial locality for young objects
33 ;;Quote: reference counting is expensive, especially with cache memory
Topic: memory management by garbage collection (116 items)
Topic: memory cache (29 items)
Topic: critical regions (58 items)
Topic: memory management by free list (25 items)
Topic: memory management by reference counting (23 items)
Topic: heap memory management (33 items)