Jonathan E. Cook, Artur W. Klauser, Alexander L. Wolf, and Benjamin G. Zorn. Semi-automatic, self-adaptive control of garbage collection rates in object databases. In Proc. 1996 SIGMOD. ACM Press, 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....Therefore, some systems partition the heap into independently collectible areas [Bis77, This research was supported in part by the Advanced Research Projects Agency of the Department of Defense, monitored by the Office of Naval Research under contract N00014 91 J 4136. HM92, YNY94, AGF95, MMH96, CKWZ96] This is also the approach taken in many distributed system [LQP92, LL92, ML94, FS96] Generational collectors are a variant of partitioned collection that use the ages of objects to optimize the collection of younger, smaller partitions [LH83] however, the age based heuristics are not ....

Jonathan E. Cook, Artur W. Klauser, Alexander L. Wolf, and Benjamin G. Zorn. Semi-automatic, self-adaptive control of garbage collection rates in object databases. In Proc. 1996 SIGMOD. ACM Press, 1996.

....benchmark distribute several implementations, including an E version and a C version. In one of our studies of garbage collection policies, we hand instrumented a subset of the E version, which is based on the Exodus storage manager, and used it to generate the PTF traces that drove our analyses [18]. After developing the AMPS toolkit, we created a second, AMPSbased C version of the benchmark. Using these two implementations of the same benchmark, we were able to assess the correctness of the automation provided by the toolkit simply by comparing the generated PTF traces from each ....

....of primary memory led us to apply it to the related study of performance of storage management in persistent object system implementations. Specifically, in prior work we used trace driven simulation to investigate the performance of storage management algorithms in persistent object systems [2,3,18]. We developed a simulator, ODBsim, that uses traces as input [1] Through trace driven simulations, we investigated methods to improve the performance of algorithms for automatic storage reclamation, focusing on policies to effectively select partitions to collect and the rate at which to perform ....

Cook JE, Klauser AW, Wolf AL, Zorn BG. Semi-automatic, self-adaptive control of garbage collection rates in object databases. Proceedings of the ACM SIGMOD International Conference on the Management of Data; 1996; 377--388.

....are crucial resources for good performance of applications. Thus, timeliness and efficiency require that the disk space be divided into smaller partitions that are traced independently. This is the approach taken in many single site systems with large, disk based heaps [Bis77, YNY94, AGF95, MMH96, CKWZ96] Ideally, partitions should be large enough to be an efficient unit of tracing, and small enough to fit in a fraction of main memory. However, to trace a site independently of other sites, and a partition independently of other partitions, objects reachable from other sites or other partitions ....

....respect to other transactions and crashes [Gra78] Partitioned garbage collection involves a wide range of other issues such as the choice of the tracing algorithm (marking, copying [Bak78] replicating [ONG93] etc. the selection of partitions to trace [CWZ94] the rate of starting traces [CKWZ96] handling transactional rollback [AGF95] etc. This thesis either does not cover the above issues or addresses them only marginally. These issues have been discussed by other researchers and most of their proposals can be applied in conjunction with the techniques proposed in this thesis. 1.3.2 ....

J. E. Cook, A. W. Klauser, A. L. Wolf, and B. G. Zorn. Semi-automatic, self-adaptive control of garbage collection rates in object databases. In Proc. SIGMOD. ACM Press, 1996.

....log updates for crash recovery. Schemes that trace the entire heap [Bak78, KW93, ONG93] do not scale to very large heaps because the nonlocal nature of tracing causes random disk accesses. Therefore, large systems partition the heap into independently collectible areas [Bis77, YNY94, AGF95, MMH96, CKWZ96] This is also the approach taken in many distributed systems [LQP92, LL92, ML94, FS96] Generational collectors are a variant of partitioned collection that use the ages of ob This paper appeared in ACM SIGMOD International Conference on Management of Data, 1997. This research was supported in ....

J. E. Cook, A. W. Klauser, A. L. Wolf, and B. G. Zorn. Semi-automatic, self-adaptive control of garbage collection rates in object databases. In Proc. 1996 SIGMOD. ACM Press, 1996.

....this partition structure. Operation Partition Regime Object Kind Operation Partition Regime Operations on Partitions on Objects Figure 2: Operation Matrices. scheme is considered to be the most efficient way to incrementally garbage collect large spaces, such as persistent object stores [11, 12, 23]. This view is also supported by recent experiments conducted by Printezis on garbage collecting small stores [24] These experiments showed that the time needed to garbage collect stores of sizes between 27MB and 30MB varied from 3 secs to 43 secs, depending on the object kinds included and the ....

E. J. Cook, A. W. Klauser, A. L. Wolf, and B. G. Zorn. Semi-Automatic, Self-Adaptive Control of Garbage Collection Rates in Object Databases. In Proceedings of SIGMOD'96, pages 377--388, Montreal, Canada, October 1996.

....distribute several implementations, including an E version and a C version. In one of our studies of garbage collection policies, we hand instrumented a subset of the E version, which is based on the Exodus storage manager, and used it to generate the PTF trace files that drove our analyses [8]. After developing the AMPS toolkit, we created a second, AMPS based C version of the benchmark. Using these two implementations of the same benchmark we were able to assess the correctness of the automation provided by the toolkit simply by comparing the generated PTF trace files from each ....

....in this domain led us to apply it to the closely related study of performance of storage management in persistent object system implementations. Specifically, in prior work we used trace driven simulation to investigate the performance of storage management algorithms in persistent object systems [8, 10, 11]. We developed a simulator, ODBsim, that uses traces as input [9] Through trace driven simulations, we investigated methods to improve the performance of algorithms for automatic storage reclamation, focusing on policies to e#ectively select partitions to collect and the rate at which to perform ....

J.E. Cook, A.W. Klauser, A.L. Wolf, and B.G. Zorn. Semi-automatic, Self-adaptive Control of Garbage Collection Rates in Object Databases. In Proceedings of the ACM SIGMOD International Conference on the Management of Data, pages 377--388. SIGMOD, June 1996.

....The POSSE (Persistent Object SyStem Evaluation) project seeks to design, implement, and evaluate alternative implementation methods for persistent object systems 1 . For the past three years, our work is specifically focused on the problem of storage management in such systems [CWZ 94, CKWZ 96, CWZ 98] In our efforts to develop new policies for storage management, we have developed a trace file format, PTFF [KWZ 96] and hand instrumented an implementation of the OO7 benchmark [CDN 93] with methods to record its structure and behavior using that format. To evaluate or improve ....

....developing and evaluating high performance algorithms for storage management in object database systems. Our research has concentrated on partition based algorithms, and our previous results include new methods for partition selection [CWZ 94] and methods for controlling garbage collection rates [CKWZ 96] Our approach to performance evaluation has evolved in the last several years, and the interfaces and formats described in this paper are the result of our experiences. Our initial approach to performance evaluation involved simultaneously developing a simulation system and synthetic database ....

Jonathan E. Cook, Artur W. Klauser, Alexander L. Wolf, and Benjamin G. Zorn. Semiautomatic, self-adaptive control of garbage collection rates in object databases. In Proceedings of the ACM SIGMOD International Conference on the Management of Data, pages 377-388, Montreal, Canada, June 1996. 10

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