| Laurent Amsaleg, P. Ferreira, Michael Franklin, and Marc Shapiro. Evaluating garbage collection for large persistent stores. In Addendum to Proc. 1995 OOPSLA Workshop on Object Database Behavior. ACM Press, 1995. |
....needed on truncating the stable log. While these require care in implementing them correctly, we do not expect them to introduce a significant performance cost for the reasons given in Section 3.5. 5. 3 Workload Amsaleg et al. pointed out the lack of a standard benchmark for garbage collectors [AFFS95] They identified certain metrics to evaluate collectors; these include isolated costs and benefits such as bookkeeping overheads and the rate of collection, as well as the effect on the overall application performance. They also identified certain parameters to control the measurement of these ....
Laurent Amsaleg, P. Ferreira, Michael Franklin, and Marc Shapiro. Evaluating garbage collection for large persistent stores. In Addendum to Proc. 1995 OOPSLA Workshop on Object Database Behavior. ACM Press, 1995.
....needed on truncating the stable log described in Section 3.3.2. In our experiments, however, we accounted for the expected log overhead from these actions; they were found to be negligible. 3.5. 2 Workload Amsaleg et al. pointed out the lack of a standard benchmark for database garbage collectors [AFFS95] such a benchmark remains absent today. Therefore, we designed a micro benchmark specifically for evaluating the overhead of maintaining insets. The benchmark database consists of a homogenous collection of small objects, each of which has a single reference and some data fields. This is similar ....
L. Amsaleg, P. Ferreira, M. Franklin, and M. Shapiro. Evaluating garbage collection for large persistent stores. In Addendum to Proc. OOPSLA Workshop on Object Database Behavior. ACM Press, 1995.
....We have not yet implemented crash recovery and the actions needed on truncating the stable log. In our experiments, however, we accounted for the expected log overhead from these actions. 5. 3 Workload Amsaleg et al. pointed out the lack of a standard benchmark for database garbage collectors [AFFS95] such a benchmark remains absent today. Therefore, we designed a microbenchmark specifically for evaluating the overhead of maintaining inter partition reference information. The benchmark database consists of a homogenous collection of small objects, each of which has a single reference and ....
L. Amsaleg, P. Ferreira, M. Franklin, and M. Shapiro. Evaluating garbage collection for large persistent stores. In Addendum to Proc. 1995 OOPSLA Workshop on Object Database Behavior. ACM Press, 1995.
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