Study of Different Types of Storage and Memory Workload Estimation and Management For Dbms, Rdbms and Ddbms |
Modern storage systems aresophisticated. Simple direct attached storage devices are giving way to storagesystems that are shared, flexible, virtualized and network-attached. Today,storage systems have their own administrators, who use specialized tools andexpertise to configure and manage storage resources. Although the separation ofstorage management and database management has many advantages, it alsointroduces problems. Database physical design and storage configuration areclosely related tasks, and the separation makes it more difficult to achieve agood end-toend design. In this paper, we attempt to close this gap byaddressing the problem of predicting the storage workload that will begenerated by a database management system. Specifically, we show how totranslate a database workload description, together with a database physicaldesign, into a characterization of the storage workload that will result. Sucha characterization can be used by a storage administrator to guide storageconfiguration. The ultimate goal of this work is to enable effective end-to-enddesign and configuration spanning both the database and storage system tiers.We present an empirical assessment of the cost of workload prediction as wellas the accuracy of the result.