search query: @keyword energy effíciency / total: 167
reference: 38 / 167
Author: | Hwang, Zhen-Huan |
Title: | Benchmarking of IP-based Network Storage Systems |
Publication type: | Master's thesis |
Publication year: | 2014 |
Pages: | x + 66 Language: eng |
Department/School: | Perustieteiden korkeakoulu |
Main subject: | Mobile Computing - Services and Security (T-110) |
Supervisor: | Ylä-Jääski, Antti |
Instructor: | Ou, Zhonghong |
Electronic version URL: | http://urn.fi/URN:NBN:fi:aalto-201411123029 |
OEVS: | Electronic archive copy is available via Aalto Thesis Database.
Instructions Reading digital theses in the closed network of the Aalto University Harald Herlin Learning CentreIn the closed network of Learning Centre you can read digital and digitized theses not available in the open network. The Learning Centre contact details and opening hours: https://learningcentre.aalto.fi/en/harald-herlin-learning-centre/ You can read theses on the Learning Centre customer computers, which are available on all floors.
Logging on to the customer computers
Opening a thesis
Reading the thesis
Printing the thesis
|
Location: | P1 Ark Aalto 2014 | Archive |
Keywords: | network storage benchmarking performance throughput cloud computing energy efficiency |
Abstract (eng): | Mobile platforms with access to high speed wireless network have become ubiquitous. Advancements in network technology and consumer electronics have brought traditional storage systems into offices and homes. Services based on cloud technologies, including object based storage, have gained popularity among both private users and enterprises. However, there is still a lack of systematic evaluation of both traditional storage systems and cloud based object storage in a mobile and wireless context. In this thesis, we evaluate the performance of three drastically different storage systems, namely NFS, iSCSI, and OpenStack Swift, which can potentially be used by mobile platforms over wireless network. We build a testbed and an in house, ad hoc microbenchmark to study the impact of various network complexities and different access behaviours of application. In addition, we employ two widely used macrobenchmarks - PostMark and FileBench - to simulate the workloads of typical applications. We find that: (1) iSCSI excels in networks whose condition is as good as LAN; (2) NFS and Swift are more suitable for complex networks such as wireless network and WAN; (3) Swift is a viable replacement for NFS in all scenarios; and (4) System configuration on the client side impacts storage performance significantly and deserve adequate attention. Furthermore, we make several recommendations to practitioners and point out numerous future research directions. |
ED: | 2014-11-16 |
INSSI record number: 50055
+ add basket
INSSI