A replacing strategy based least recently used algorithm in storage system
/ 0
- Details
- Category: Information technologies, systems analysis and administration
- Last Updated on 08 February 2016
- Published on 08 February 2016
- Hits: 4570
Authors:
Heng Yang, Nanyang Normal University, Nanyang, China
Abstract:
Purpose. Big data is a very large vocabulary we have heard recently. No matter for business or personal users, there is always a lot of important data to store. When we use the storage system, we often want the system to respond quickly enough to reach a state of no delay. This is a big challenge to the storage system. Scientists have already done many researches on this topic and they found that the use of cache in the storage system can improve the performance of storage system greatly.
Methodology. Cache algorithm is a hot research field in the current storage area. Least Recently Used algorithm (LRU) is a commonly used cache replacement algorithm.
Findings. Since the new hash value that appears atop of the stack needs to adjust the stack even if the visited page is already in memory, this takes much time. We need a better cache replacement algorithm to improve the performance.
Originality. A new replacing strategy based on the LRU algorithm has been developed; it is called Improved LRU algorithm (ILRU). It can increase the hit rate when more users suddenly have a higher access to the unfamiliar page. We determine whether the hash value is added to a page or not through searching the access hash value in the LRU queue.
Practical value. The test results show that the design of ILRU algorithm can improve the performance comparing to traditional LRU algorithm. At the same time, ILRU algorithm has higher hit rate than FIFO algorithm.
References:
1. Gantz, J. and Reinsel, D. (2011), “The digital universe in 2020: Big data, bigger digital shadows, and biggest growth in the far east”, IDC iView: IDC Analyze the Future, pp. 1−16.
2. Xu, Z., Ning, W., Vassilios, G.V. and Michael, P.H. (2015), “A distributed in-network caching scheme for P2P-like content chunk delivery”, Computer Networks, vol. 91, no.14, pp. 577−592.
3. Hamilton, T., Brian, D., Jules, W., Russell, K., Jonathan, P. and Douglas, C.S. (2014), “Aniruddha Gokhale. DRE system performance optimization with the SMACK cache efficiency metric”, Journal of Systems and Software, vol. 98, pp. 25−43.
4. P. Julian B. and F. Sagayaraj F. (2015), “Improving the performance of a proxy cache using very fast decision tree classifier”, Procedia Computer Science, vol. 48, pp. 304−312.
5. Nicaise C.F., Philippe, N., Giovanni, N. and Don, T. (2014), “Performance evaluation of hierarchical TTL-based cache networks”, Computer Networks, vol. 48, pp. 304−312.
6. O'neil, E.J., O'neil, P.E. and Weikum, G. (2012), “The LRU-K page replacement algorithm for database disk buffering”, Proc. of the Conf. on ACM SIGMOD Record, NY, USA, pp. 297−306.
7. Shasha, D. and Johnson, T. (2010), “2Q: A low overhead high performance buffer management replacement algoritm”, Proc. of the 20th International Conference on Very Large Databases. Copenhagen, Danmark, pp. 439−450.
8. Wenjia, N., Gang, L., Endong, T., Xinghua, Y., Liang, C., Zhong, Z.S. and Song, C. (2014), “Interaction relationships of caches in agent-based HD video surveillance: Discovery and utilization”, Journal of Network and Computer Applications, vol. 37, pp. 155−169.
9. Jiang, S., Ding, X. and Chen, F. (2006), “DULO: an effective buffer cache management scheme to exploit both temporal and spatial locality”, Proc. of the 4th USENIX Conference on File and Storage Technologies. California, USA, pp. 8−17.
 2015_06_yang
|
|
 2016-02-08  541.71 KB  988 |
|
Archive