Managing distributed memory to meet multiclass workload response time goals

In this paper we present an online method for managing a goal-oriented buffer partitioning in the distributed memory of a network of workstations. Our algorithm implements a feedback mechanism which dynamically changes the sizes of dedicated buffer areas and thereby the buffer hit rate for the different classes in such a way that user-specified response time goals are satisfied. The aggregated size of the buffer memory across all network nodes remains constant and only the partitioning is changed. The algorithm is based on efficiently approximating the trajectory of the per-class response time curves as a function of the available buffer. Changes in the workload that would lead to violation of response time goals are counteracted by accordingly adjusting the buffer allocation. For local replacement decisions, we integrate a cost-based buffer replacement algorithm to fit into our goal-oriented approach. We have implemented our algorithm in a detailed simulation prototype and we present first results obtained from this prototype.

[1]  Miron Livny,et al.  Goal-oriented buffer management revisited , 1996, SIGMOD '96.

[2]  John H. Hartman,et al.  Efficient cooperative caching using hints , 1996, OSDI '96.

[3]  Philip A. Bernstein,et al.  Recovery Algorithms for Database Systems , 1983, IFIP Congress.

[4]  Kurt P. Brown.  Goal-oriented memory allocation in database management systems , 1996 .

[5]  Philip S. Yu,et al.  Efficient LRU-Based Buffering in a LAN Remote Caching Architecture , 1996, IEEE Trans. Parallel Distributed Syst..

[6]  Gerhard Weikum,et al.  The LRU-K page replacement algorithm for database disk buffering , 1993, SIGMOD Conference.

[7]  Michael J. Franklin,et al.  Client Data Caching , 1996 .

[8]  David K. Smith.  Theory of Linear and Integer Programming , 1987 .

[9]  Anna R. Karlin,et al.  Implementing global memory management in a workstation cluster , 1995, SOSP.

[10]  Philip E. Gill,et al.  Numerical Linear Algebra and Optimization , 1991 .

[11]  Laszlo A. Belady,et al.  An anomaly in space-time characteristics of certain programs running in a paging machine , 1969, CACM.

[12]  Philip S. Yu,et al.  Policies for efficient memory utilization in a remote caching architecture , 1991, [1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems.

[13]  Donald F. Ferguson,et al.  Goal-oriented dynamic buffer pool management for data base systems , 1995, Proceedings of First IEEE International Conference on Engineering of Complex Computer Systems. ICECCS'95.

[14]  Michael Dahlin,et al.  Cooperative caching: using remote client memory to improve file system performance , 1994, OSDI '94.

[15]  Miron Livny,et al.  Managing Memory to Meet Multiclass Workload Response Time Goals , 1993, VLDB.

[16]  Irving L. Traiger,et al.  The notions of consistency and predicate locks in a database system , 1976, CACM.

[17]  Miron Livny,et al.  Memory management for scalable Web data servers , 1997, Proceedings 13th International Conference on Data Engineering.

[18]  Erhard Rahm,et al.  Goal-oriented Performance Control for Transaction Processing , 1997, MMB.

[19]  Brian Randell,et al.  Operating Systems, An Advanced Course , 1978 .

[20]  Gerhard Weikum,et al.  A cost-model-based online method for distributed caching , 1997, Proceedings 13th International Conference on Data Engineering.

[21]  SarkarPrasenjit,et al.  Efficient cooperative caching using hints , 1996 .

[22]  David E. Culler,et al.  A case for NOW (networks of workstation) , 1995, PODC '95.