Zone-Based Shortest Positioning Time First Scheduling for MEMS-Based Storage Devices
نویسندگان
چکیده
Access latency to secondary storage devices is frequently a limiting factor in computer system performance. New storage technologies promise to provide greater storage densities at lower latencies that is currently obtainable with hard disk drives. MEMS-based storage devices [1, 11, 20] use orthogonal magnetic or physical recording techniques and thousands of simultaneously active MEMS-based read-write tips to provide high-density low-latency non-volatile storage. These devices promise seek times 10–20 times faster than hard drives, storage densities 10 times greater, and power consumption an order of magnitude lower. Previous research has examined data layout and request ordering algorithms that are analogs of those developed for hard drives. We present an analytical model of MEMS device performance that motivates a computationally simple MEMS-based request scheduling algorithm called ZSPTF, which has average response times comparable to Shortest Positioning Time First (SPTF) but with response time variability comparable to Circular Scan (C-SCAN).
منابع مشابه
Towards Optimal I/O Scheduling for MEMS-Based Storag
Magnetic disks have been the dominant on-line storage technology for more than three decades. Disks have maintained this dominance in spite of other competing storage technologies such as bubble memory, holographic stores, and improved DRAMs. Although magnetic disks have enjoyed this position in the persistent storage arena, more recently the memory hierarchy has suffered from problems of laten...
متن کاملComparison of I/O Scheduling Algorithms for High Parallelism MEMS-Based Storage Devices
MEMS-based storage is one of the leading candidates as tomorrow’s storage medium due to its salient characteristics such as high-parallelism, high density, and low-power consumption. Because physical structures of MEMS-based storage is different from those of hard disks, new software management techniques for MEMS-based storage are needed. Specifically, MEMS-based storage has thousands of paral...
متن کاملAn Analytical Solution to a MEMS Seek Time Model
The huge disparity between memory access times and disk access times has been the subject of extensive research. CPU speed has been increasing rapidly but disk access latency has lagged behind—disk transfer rates have been increasing at 40% per year, while seek times and rotational latency have been increasing at less than 10% per year [4]. This disparity has created a performance bottleneck in...
متن کاملTowards Optimal I/O Scheduling for MEMS-based Storage
MEMS-based storage devices are currently being developed to narrow the gap between processor and disk speeds. MEMS-based storage devices have a different architecture from disk devices, thus algorithms, such as I/O scheduling and data placement, designed for disks need to be revisited. In this paper, we focus on developing an I/O scheduling algorithm for MEMS-based storage devices. Our theoreti...
متن کامل