1,863 Hits in 4.9 sec

Implementation and Evaluation of a Popularity-Based Reconstruction Optimization Algorithm in Availability-Oriented Disk Arrays

Lei Tian, Hong Jiang, Dan Feng, Qin Xin, Xing Shu
2007 24th IEEE Conference on Mass Storage Systems and Technologies (MSST 2007)  
In this paper, we implement the incorporation of a Popularity-based multi-threaded Reconstruction Optimization algorithm, PRO, into the recovery mechanism of the Linux software RAID (MD), which is a well-known  ...  Our results demonstrate PRO's significant performance advantage over the existing reconstruction schemes, especially on a RAID-5 disk array, in terms of the measured reconstruction time and response time  ...  One can see that the response time improvement of PRO for a RAID-5 disk array is more significant than that for a RAID-10 disk array.  ... 
doi:10.1109/msst.2007.4367978 fatcat:aqdzshbeszcvnif2o6ttb3vl7u

REMB: Recoverable External Memory Bitmap of Software RAID

Wang Wei, Yu Li-hua
2016 International Journal of Database Theory and Application  
Experimental results show that REMB improve cloud database performance by 30%~60%. 338 Copyright ⓒ 2016 SERSC by user have high recovery priority, Ceph's I/O response time will still be severely affected  ...  In this paper, we propose REMB(Recoverable External Memory Bitmap) to improve MD(Software RAID) performance, and implement it in distributed block storage of NTES.  ...  We have to create a new LOG bitmap and perform full copy data recovery for the software RAID.  ... 
doi:10.14257/ijdta.2016.9.10.29 fatcat:xvnkjsbnevd6xcao2ly3vquvj4

JOR: A Journal-guided Reconstruction Optimization for RAID-Structured Storage Systems

Suzhen Wu, Dan Feng, Hong Jiang, Bo Mao, Lingfang Zeng, Jianxi Chen
2009 2009 15th International Conference on Parallel and Distributed Systems  
Experimental results obtained from our JOR prototype implementation demonstrate that JOR reduces reconstruction times of two state-of-the-art reconstruction schemes by an amount that is approximately proportional  ...  This paper proposes a simple and practical RAID reconstruction optimization scheme, called JOurnal-guided Reconstruction (JOR).  ...  Studies on RAID reconstruction have been concentrating on how to improve the performance of RAID reconstruction algorithms.  ... 
doi:10.1109/icpads.2009.69 dblp:conf/icpads/WuFJMZC09 fatcat:ma2brzvnxrcnhfe7p4tr2shbqa

An Efficient Penalty-Aware Cache to Improve the Performance of Parity-Based Disk Arrays under Faulty Conditions

Shenggang Wan, Xubin He, Jianzhong Huang, Qiang Cao, Shiyi Li, Changsheng Xie
2013 IEEE Transactions on Parallel and Distributed Systems  
Similarly, VDF-LRU can speed up the online recovery by up to 34.6% in RAID-5 and 38.2% in RAID-6, or improve the system service rate by up to 28.4% in RAID-5.  ...  To address this issue, we propose a novel penalty-aware buffer cache replacement strategy, named Victim Disk(s) First (VDF) cache, to improve the reliability and performance of a storage system consisting  ...  Although we improve the weight of the blocks on faulty disk to n − 1 times for RAID-5 and n − 2 times for RAID-6 in both VDF-LRU and VDF-LFU, the efficiency is not the same.  ... 
doi:10.1109/tpds.2012.266 fatcat:iudtd7ikhngpncsxbas3pn2jkq

The Raid distributed database system

B. Bhargava, J. Riedl
1989 IEEE Transactions on Software Engineering  
A software tool for the evaluation of the implementation of transaction processing algorithms in an operating system kernel is proposed.  ...  Raid is a robust and adaptable distributed database system for transaction processing. Raid is a message-passing system, with server processes on each site.  ...  Lilien of AT&T Bell Laboratories for his help in preparing the first draft of this paper in the Summer of 1987.  ... 
doi:10.1109/32.24726 fatcat:rkch6ow2hbcblcnlh7ic2keobi

S2-RAID: A new RAID architecture for fast data recovery

Jiguang Wan, Jibin Wang, Qing Yang, Changsheng Xie
2010 2010 IEEE 26th Symposium on Mass Storage Systems and Technologies (MSST)  
SPC IO traces and standard benchmarks have been used to measure the performance of S 2 -RAID as compared to existing baseline software RAID, MD.  ...  We have implemented a prototype S 2 -RAID system in Linux operating system for the purpose of evaluating its performance potential.  ...  We would also like to thank the anonymous reviewers for their valuable insights that have improved the quality of the paper greatly.  ... 
doi:10.1109/msst.2010.5496980 dblp:conf/mss/WanWYX10 fatcat:yihawn35j5bp7k7nzl2cahut6a

Using active data repair and migration technologies for RAID-based storage systems

Yin Yang, Wei Liang, Zhihu Tan, Jiguang Wan, Sandeep Subhoyt
2014 Journal of Computers  
The overall results indicate that the RARM can improve performance and reliability of the system with little influence.  ...  This paper proposes a new RAID-based system using Active data Repair and Migration technologies called RARM, which detects the status of sector, then allocates reserved space in every disk and arranges  ...  We measured both front and end performance in terms of average user response times while online recovery operation and the total online recovery times for the three benchmarks.  ... 
doi:10.4304/jcp.9.10.2281-2293 fatcat:hhr5w2h46fbetackndkjvc2lme

Efficient Data Protection for Enterprise Data Centers

Khaled Aldossari
2015 International Journal of Computer Science & Information Technology (IJCSIT)  
Some of these challenges include recovery objectives, data explosion, cost and the nature of data. The paper explores multiple methods of data protection at different storage levels.  ...  RAID disk arrays, snapshot technology, storage mirroring, and backup and archive strategies all are methods used by many large organizations to protect their data.  ...  ACKNOWLEDGEMENTS The author would like to thank his colleagues in Saudi Aramco, Hussain Al-Raqa for his encouragement to write this paper and Edward Liu for his valuable comments.  ... 
doi:10.5121/ijcsit.2015.7112 fatcat:hflpwuq2yncn7awzksi6xc3hty

RAID10L: A high performance RAID10 storage architecture based on logging technique

Bo Mao, Dan Feng, Suzhen Wu, Jianxi Chen, Lingfang Zeng, Lei Tian
2008 2008 13th Asia-Pacific Computer Systems Architecture Conference  
The. goal of RAIDJOL is to signi.ficant~v improve the write performance of RAIDJ0 at some little expense of reliability. In RAIDJOL both read and write requests are processed in a balance scheme.  ...  Driven by some real-life traces, RAIDJOL gains improvement up to 30.7% with an average of27.7% than RAIDJO in terms ofaverage response time.  ...  Algorithm 1: Recovery from data disk failure in data loss.  ... 
doi:10.1109/apcsac.2008.4625430 dblp:conf/aPcsac/MaoFWCZT08 fatcat:4fbia46kcjcmbn7vcc5o2emio4

BMCloud: Minimizing Repair Bandwidth and Maintenance Cost in Cloud Storage

Chao Yin, Changsheng Xie, Jiguang Wan, Chih-Cheng Hung, Jinjiang Liu, Yihua Lan
2013 Mathematical Problems in Engineering  
The other one is repairing bandwidth, which is a concern for most of us. We addressed these two problems from both theoretical and practical perspectives.  ...  To protect data in cloud storage, fault tolerance and efficient recovery become very important.  ...  It can improve recovery bandwidth performance and ensure data integrity.  ... 
doi:10.1155/2013/756185 fatcat:6nsyk23645hrpnnydueb7q7tg4

Online availability upgrades for parity-based RAIDs through supplementary parity augmentations

Lei Tian, Qiang Cao, Hong Jiang, Dan Feng, Changsheng Xie, Qin Xin
2011 ACM Transactions on Storage  
by one disk for each stripe to balance the workload and thus maximize the performance of reconstruction during recovery.  ...  is proportional to the supplementary parity coverage, thus significantly reducing the overhead of data regeneration and decreasing recovery time in parity-based RAID systems.  ...  The average user response time during recovery and the reconstruction time are the two most important metrics in evaluating the recovery performance for RAID systems.  ... 
doi:10.1145/1970338.1970341 fatcat:4kemju36d5hurlgqa4jyi7ps24

Improving Availability of RAID-Structured Storage Systems by Workload Outsourcing

Suzhen Wu, Hong Jiang, Dan Feng, Lei Tian, Bo Mao
2011 IEEE transactions on computers  
improves the performance of the low-priority background tasks, such as RAID reconstruction and RAID resynchronization.  ...  Importantly, WorkOut is portable and can be easily incorporated into any existing optimizing algorithms for RAID-structured storage systems.  ...  While all the above algorithms focus on improving the performance by optimizing the organization of work within a disk or a single RAID set, our work increases the performance of the low-priority background  ... 
doi:10.1109/tc.2010.206 fatcat:t2rg7yvaq5fxvlqg2wmwhli57a

Storage architecture with integrity, redundancy and encryption

Henning Klein, Jorg Keller
2009 2009 IEEE International Symposium on Parallel & Distributed Processing  
We present performance figures of our experiments.  ...  We provide an implementation, i.e. a driver, which encapsulates all these features and uses parallel algorithms exploiting multicore processor performance to match the bandwidth available from multiple  ...  Therefore the positions of any computed possible solution for three simultaneous errors can be checked 15 times when Figure 10 shows the improvement of error correction capabilities before and after  ... 
doi:10.1109/ipdps.2009.5160901 dblp:conf/ipps/KleinK09 fatcat:ru563cl4yze5tcbpl4xstzlqkq

HRAID6ML: A hybrid RAID6 storage architecture with mirrored logging

Lingfang Zeng, Dan Feng, Jianxi Chen, Qingsong Wei, Bharadwaj Veeravalli, Wenguo Liu
2012 012 IEEE 28th Symposium on Mass Storage Systems and Technologies (MSST)  
In this paper, we propose HRAID6ML, a new logging architecture for RAID6 systems for enhanced energy efficiency, performance and reliability.  ...  Furthermore, HRAID6ML prolongs the lifecycle of the SSDs and improves the systems energy efficiency by reducing the SSDs write frequency. We have implemented proposed HRAID6ML.  ...  At the same time, the parity part can be recovered through the RAID6 recovery algorithm.  ... 
doi:10.1109/msst.2012.6232374 dblp:conf/mss/ZengFCWVL12 fatcat:nauwmreezrcx7a7iyulzl5o3pe

Parity Declustering for Continuous Operation in Redundant Disk Arrays [chapter]

2009 High Performance Mass Storage and Parallel I/O  
This result arises from the simple model of disk access performance used in previous work, which did not consider throughput variations due to positioning delays. Accesion For NTIS CRA&, d  ...  This declustered parity organization balances cost against data reliability and performance during failure recovery in highly-available parity-based arrays for use in continuous-operation systems.  ...  overhead for reduced user performance degradation during on-line failure recovery, can be effectively implemented in array controlling software.  ... 
doi:10.1109/9780470544839.ch13 fatcat:yj2eylluibdy5kgwzhmfzavdte
« Previous Showing results 1 — 15 out of 1,863 results