Loading...

A High-Level SSD Performance Estimation Model based on Workload Features

Azadvar, Soheil | 2018

1112 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 51483 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Asadi, Hossein
  7. Abstract:
  8. With increases in amount of data being processed and generated, the need for highperformance data storage systems is essential. These systems must keep up with request throughputs as high as millions of requests per second. Solid-State Drives (SSD) lack mechanical parts and have lower read and write latency. SSDs have key differences with traditional spinning drives in having a fixed read/write granularity, erase before write and requiring wear-leveling and garbage collection operations. These operations have significant impact on drive performance and also frequency of these operations depends on the workload characteristics, such as randomness. In this thesis, a high-level estimator of SSD performance is used with a study of the effects of workload characteristics on SSDs. In the proposed model, features extracted from time periods of workload are given as input to machine learning methods. Futhermore, the effect of high-level features on hard disk drive performance is reviewed and is further developed for use on SSDs. To achieve this purpose, workloads recorded from real hardware are used for training of machine learning models and finally a model with 86% confidence level is proposed for estimation of SSD performance
  9. Keywords:
  10. Machine Learning ; High-performance Solid State Disk Drive (SSD) ; Solid State Disk Drive ; Work Load ; Workload Characteristics

 Digital Object List

 Bookmark