Benchmarking Tape System Performance

摘要

In spite of the rapid decrease in magnetic disk prices, tertiary storage (i.e., removable media in a robotic storage library) is becoming increasingly popular. The fact that so much data can be stored encourages applications that use ever more massive data sets. Application drivers include multimedia databases, data warehouses, scientific databases, dataintensive scientific research, and digital libraries and archives. The research community, has responded with investigations into systems integration, performance modeling, and performance optimization. Tertiary storage systems present special challenges because of their unusual performance characteristics. Access latencies can range into minutes even on unloaded systems, but transfer rates can be very high. Tertiary storage is implemented with a wide array of technologies, each with its own performance quirks. However, little detailed performance information about tertiary storage devices has been published. As a result, mass storage system (MSS) implementers must rely on vendor-reported numbers or their own tests to select appropriate tertiary storage devices. Additionally, MSS designers must have detailed knowledge of the performance characteristics of their devices to optimally place files on media and perform other optimizations. In this paper we present detailed measurements of several tape drives and describe the tests used to gather this data. The tape drives we measured include the DLT 4000, Ampex 310, IBM 3590, 4mm DAT, and the Sony DTF drive. This mixture of equipment includes high and low performance drives, serpentine and helical scan drives, and cartridge and cassette tapes. This data is suitable for system performance modeling or system performance optimization studies. By measuring and modeling a variety ...