Data migration on parallel disks: Algorithms and evaluation

Our work is motivated by the problem of managing data on storage devices, typically a set of disks. Such storage servers are used as web servers or multimedia servers, for handling high demand for data. As the system is running, to exhibit good performance, it needs to respond dynamically to changes in demand for different data items. There are known algorithms for mapping demand to a layout. When the demand changes, a new layout can be computed. In this work we study thedata migration problem, which arises when we need to change one layout to another quickly. This problem has been studied earlier where for each disk a new layout has been prescribed. However, to apply these algorithms effectively, we identify another problem that we refer to as the correspondence problem, whose solution has a significant impact on the overall solution for the data migration problem. We study algorithms for the data migration problem in more detail and identify variations of the basic algorithm that seem to improve performance in practice, even though some of the variations have poor worst-case behavior.

[1]  Shahram Ghandeharizadeh,et al.  Design and Implementation of Scalable Continuous Media Servers , 1998, Parallel Comput..

[2]  Claude Berge,et al.  A short proof for a generalization of Vizing's theorem , 1991, J. Graph Theory.

[3]  Hadas Shachnai,et al.  Polynominal time approximation schemes for class-constrained packing problem , 2000, APPROX.

[4]  Samir Khuller,et al.  Algorithms for Data Migration with Cloning , 2004, SIAM J. Comput..

[5]  Anna R. Karlin,et al.  On Algorithms for EÆcient Data Migration , 2022 .

[6]  J. A. Bondy,et al.  Graph Theory with Applications , 1978 .

[7]  Tami Tamir,et al.  Polynominal time approximation schemes for class-constrained packing problem , 2000, APPROX.

[8]  Samir Khuller,et al.  Approximation algorithms for data placement on parallel disks , 2000, SODA '00.

[9]  Philip S. Yu,et al.  DASD dancing: a disk load balancing optimization scheme for video-on-demand computer systems , 1995, SIGMETRICS '95/PERFORMANCE '95.

[10]  Samir Khuller,et al.  Algorithms for non-uniform size data placement on parallel disks , 2003, J. Algorithms.

[11]  Hadas Shachnai,et al.  Approximation Schemes for Generalized 2-Dimensional Vector Packing with Application to Data Placement , 2003, RANDOM-APPROX.

[12]  Ann L. Chervenak,et al.  Tertiary Storage: An Evaluation of New Applications , 1994 .

[13]  Éva Tardos,et al.  An approximation algorithm for the generalized assignment problem , 1993, Math. Program..

[14]  John C. S. Lui,et al.  Design of Scalable Continuous Media Servers , 2004, Multimedia Tools and Applications.

[15]  Joseph Hall,et al.  An Experimental Study of Data Migration Algorithms , 2001, WAE.

[16]  Joseph Hall,et al.  On algorithms for efficient data migration , 2001, SODA '01.

[17]  Hadas Shachnai,et al.  On Two Class-Constrained Versions of the Multiple Knapsack Problem , 2001, Algorithmica.