Towards fast and lightweight checkpointing for mobile virtualization using NVRAM
Checkpointing is a key enabler of hibernation, live migration and fault-tolerance for virtual machines (VMs) in mobile devices. However, checkpointing a VM is usually heavyweight: the VM's entire memory needs to be dumped to storage, which induces a significant amount of (slow) I/O operations,...
Saved in:
Main Authors: | , , , , , , , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2020
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/144764 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-144764 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1447642020-11-24T01:13:08Z Towards fast and lightweight checkpointing for mobile virtualization using NVRAM Zhong, Kan Liu, Duo Wu, Yunsong Long, Linbo Liu, Weichen Ren, Jinting Liu, Renping Liang, Liang Shao, Zili Li, Tao School of Computer Science and Engineering Engineering::Computer science and engineering Mobile Virutalization Checkpointing Checkpointing is a key enabler of hibernation, live migration and fault-tolerance for virtual machines (VMs) in mobile devices. However, checkpointing a VM is usually heavyweight: the VM's entire memory needs to be dumped to storage, which induces a significant amount of (slow) I/O operations, degrading system performance and user experience. In this paper, we propose FLIC, a fast and lightweight checkpointing machinery for virtualized mobile devices by taking advantages of recent byte-addressable, non-volatile memory (NVRAM). Instead of saving the VM's entire memory to storage, we store its working set pages in NVRAM, avoiding accessing slow flash memory (compared to server-grade SSDs). To further reduce the write activities to flash memory, we propose an energy-efficient data deduplication to eliminate redundant data in VM snapshot and save storage space. Experimental results based on an Exynos 5250 SoC show that our approach can effectively improve the performance of checkpointing in mobile virutalization and save energy. This work is partially supported by grants from the National Natural Science Foundation of China (61672116 and 61601067), Chongqing High-Tech Research Program cstc2016jcyjA0332, the Fundamental Research Funds for the Central Universities under Grant 0214005207005, the Science and Technology Research Program of Chongqing Municipal Education Commission (KJ1704085), the Chongqing Research Program of Basic Research and Frontier Technology (cstc2017jcyjAX0164), the Research Grants Council of the Hong Kong Special Administrative Region, China (GRF 15213814, GRF 15222315), and Direct Grant for Research, The Chinese University of Hong Kong (Project No. 4055096). A preliminary version of this paper was presented at the ACM/IEEE 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE) [1]. 2020-11-24T01:13:08Z 2020-11-24T01:13:08Z 2018 Journal Article Zhong, K., Liu, D., Wu, Y., Long, L., Liu, W., Ren, J., . . . Li, T. (2019). Towards fast and lightweight checkpointing for mobile virtualization using NVRAM. IEEE Transactions on Parallel and Distributed Systems, 30(6), 1421-1433. doi:10.1109/TPDS.2018.2886906 1045-9219 https://hdl.handle.net/10356/144764 10.1109/TPDS.2018.2886906 6 30 1421 1433 en IEEE Transactions on Parallel and Distributed Systems © 2018 IEEE. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering Mobile Virutalization Checkpointing |
spellingShingle |
Engineering::Computer science and engineering Mobile Virutalization Checkpointing Zhong, Kan Liu, Duo Wu, Yunsong Long, Linbo Liu, Weichen Ren, Jinting Liu, Renping Liang, Liang Shao, Zili Li, Tao Towards fast and lightweight checkpointing for mobile virtualization using NVRAM |
description |
Checkpointing is a key enabler of hibernation, live migration and fault-tolerance for virtual machines (VMs) in mobile devices. However, checkpointing a VM is usually heavyweight: the VM's entire memory needs to be dumped to storage, which induces a significant amount of (slow) I/O operations, degrading system performance and user experience. In this paper, we propose FLIC, a fast and lightweight checkpointing machinery for virtualized mobile devices by taking advantages of recent byte-addressable, non-volatile memory (NVRAM). Instead of saving the VM's entire memory to storage, we store its working set pages in NVRAM, avoiding accessing slow flash memory (compared to server-grade SSDs). To further reduce the write activities to flash memory, we propose an energy-efficient data deduplication to eliminate redundant data in VM snapshot and save storage space. Experimental results based on an Exynos 5250 SoC show that our approach can effectively improve the performance of checkpointing in mobile virutalization and save energy. |
author2 |
School of Computer Science and Engineering |
author_facet |
School of Computer Science and Engineering Zhong, Kan Liu, Duo Wu, Yunsong Long, Linbo Liu, Weichen Ren, Jinting Liu, Renping Liang, Liang Shao, Zili Li, Tao |
format |
Article |
author |
Zhong, Kan Liu, Duo Wu, Yunsong Long, Linbo Liu, Weichen Ren, Jinting Liu, Renping Liang, Liang Shao, Zili Li, Tao |
author_sort |
Zhong, Kan |
title |
Towards fast and lightweight checkpointing for mobile virtualization using NVRAM |
title_short |
Towards fast and lightweight checkpointing for mobile virtualization using NVRAM |
title_full |
Towards fast and lightweight checkpointing for mobile virtualization using NVRAM |
title_fullStr |
Towards fast and lightweight checkpointing for mobile virtualization using NVRAM |
title_full_unstemmed |
Towards fast and lightweight checkpointing for mobile virtualization using NVRAM |
title_sort |
towards fast and lightweight checkpointing for mobile virtualization using nvram |
publishDate |
2020 |
url |
https://hdl.handle.net/10356/144764 |
_version_ |
1688665341375283200 |