MapReduce Job Scheduling in Hybrid Storage Modes
CSTR:
Author:
  • Article
  • | |
  • Metrics
  • |
  • Reference [35]
  • |
  • Related [20]
  • | | |
  • Comments
    Abstract:

    In a heterogeneous Hadoop cluster scenario, the hybrid use of erasure codes and replica storage modes, as well as the real-time computing capability difference of server nodes lead to the low efficiency of MapReduce job processing. To deal with this problem, this study implements a scheduling strategy that dynamically adjusts MapReduce job assignment in multi-concurrent scenarios according to data storage situations and the real-time load of nodes. This strategy dynamically controls the concurrent amount of tasks of each node by modifying data storage location strategies in the current Hadoop framework, so as to achieve more balanced resource allocation among jobs when multiple jobs are concurrent. The experimental results show that the scheduling mode proposed in this study can shorten the job completion time by about 17% and effectively avoid the starvation phenomenon faced by some jobs compared with the two default job scheduling strategies of Hadoop.

    Reference
    [1] Dean J, Ghemawat S. MapReduce: Simplified data processing on large clusters. Communications of the ACM, 2008, 51(1): 107–113. [doi: 10.1145/1327452.1327492
    [2] 徐鹏. Hadoop 2. X HDFS源码剖析. 北京: 电子工业出版社, 2016. 1–25.
    [3] Apache. HDFS architecture. https://hadoop.apache.org/docs/r3.3.1/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html. (2021-06-15)[2022-08-03].
    [4] Chen YP, Ganapathi A, Griffith R, et al. The case for evaluating mapreduce performance using workload suites. Proceedings of the 19th IEEE Annual International Symposium on Modelling, Analysis, and Simulation of Computer and Telecommunication Systems. Singapore: IEEE, 2011. 390–399.
    [5] Ahmad F, Chakradhar ST, Raghunathan A, et al. Tarazu: Optimizing mapReduce on heterogeneous clusters. ACM SIGARCH Computer Architecture News, 2012, 40(1): 61–74. [doi: 10.1145/2189750.2150984
    [6] IDC. The digitization of the world from edge to core. https://www.seagate.com/files/www-content/our-story/trends/files/idc-seagate-dataage-whitepaper.pdf. (2018-12-04)[2022-08-03].
    [7] CAICT中国信通院. 数据中心产业图谱研究报告. http://www.caict.ac.cn/kxyj/qwfb/ztbg/202201/P020220125529907466991.pdf. (2022-01-26)[2022-08-03].
    [8] Wikipedia. Era sure code. https://en.wikipedia.org/wiki/Era sure_code. (2022-06-26).
    [9] Ceph. Ceph documentation. https://docs.ceph.com/en/latest/rados/operations/erasure-code/. (2019-04-23)[2022-08-03].
    [10] Huang C, Simitci H, Xu YK, et al. Erasure coding in Windows azure storage. Proceedings of the 2012 USENIX Conference on Annual Technical Conference. Boston: USENIX Association, 2012. 15–26.
    [11] Muralidhar S, Lloyd W, Roy S, et al. F4: Facebook’s warm BLOB storage system. Proceedings of the 11th USENIX Symposium on Operating Systems Design and Implementation. Broomfield: USENIX Association, 2014. 383–398.
    [12] Wang J, Shang PJ, Yin JL. DRAW: A new data-grouping-aware data placement scheme for data intensive applications with interest locality. In: Li XL, Qiu J, eds. Cloud Computing for Data-intensive Applications. New York: Springer, 2014. 149–174.
    [13] Bawankule KL, Dewang RK, Singh AK. Historical data based approach to mitigate stragglers from the Reduce phase of MapReduce in a heterogeneous Hadoop cluster. Cluster Computing, 2022, 25(5): 3193–3211. [doi: 10.1007/s10586-021-03530-x
    [14] Jeyaraj R, Ananthanarayana VS, Paul A. MapReduce scheduler to minimize the size of intermediate data in shuffle phase. Proceedings of the 18th IEEE/ACIS International Conference on Computer and Information Science. Beijing: IEEE, 2019. 30–34.
    [15] Dai XM, Bensaou B. Scheduling for response time in Hadoop MapReduce. Proceedings of the 2016 IEEE International Conference on Communications. Kuala Lumpur: IEEE, 2016. 1–6.
    [16] Kavulya S, Tan JQ, Gandhi R, et al. An analysis of traces from a production MapReduce cluster. Proceedings of the 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing. Melbourne: IEEE, 2010. 94–103.
    [17] Maleki N, Rahmani AM, Conti M. SPO: A secure and performance-aware optimization for MapReduce scheduling. Journal of Network and Computer Applications, 2021, 176: 102944. [doi: 10.1016/j.jnca.2020.102944
    [18] Chen L, Liu ZH. Energy- and locality-efficient multi-job scheduling based on MapReduce for heterogeneous datacenter. Service Oriented Computing and Applications, 2019, 13(4): 297–308. [doi: 10.1007/s11761-019-00273-x
    [19] Rashmi KV, Shah NB, Ramchandran K, et al. Regenerating codes for errors and erasures in distributed storage. Proceedings of the 2012 IEEE International Symposium on Information Theory. Cambridge: IEEE, 2012. 1202–1206.
    [20] Chen HCH, Hu YC, Lee PPC, et al. NCCloud: A network-coding-based storage system in a cloud-of-clouds. IEEE Transactions on Computers, 2014, 63(1): 31–44. [doi: 10.1109/TC.2013.167
    [21] Shi HY, Lu XY. TriEC: Tripartite graph based erasure coding NIC offload. Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. Denver: ACM, 2019. 44.
    [22] Wang F, Tang YJ, Xie YW, et al. XORInc: Optimizing data repair and update for erasure-coded systems with XOR-based in-network computation. Proceedings of the 35th Symposium on Mass Storage Systems and Technologies. Santa Clara: IEEE, 2019. 244–256.
    [23] Mitra S, Panta R, Ra MR, et al. Partial-parallel-repair (PPR): A distributed technique for repairing erasure coded storage. Proceedings of the 11th European Conference on Computer Systems. London: ACM, 2016. 30.
    [24] Li XL, Yang ZR, Li JH, et al. Repair pipelining for erasure-coded storage: Algorithms and evaluation. ACM Transactions on Storage, 2021, 17(2): 13
    [25] Xu LL, Lyu M, Li QL, et al. SelectiveEC: Towards balanced recovery load on erasure-coded storage systems. IEEE Transactions on Parallel and Distributed Systems, 2022, 33(10): 2386–2400. [doi: 10.1109/TPDS.2021.3129973
    [26] Chan JCW, Ding Q, Lee PPC, et al. Parity logging with reserved space: Towards efficient updates and recovery in erasure-coded clustered storage. Proceedings of the 12th USENIX Conference on File and Storage Technologies. Santa Clara: USENIX Association, 2014. 163–176.
    [27] Rashmi KV, Shah NB, Gu DK, et al. A “hitchhiker’s” guide to fast and efficient data reconstruction in erasure-coded data centers. Proceedings of the 2014 ACM Conference on SIGCOMM. Chicago: ACM, 2014. 331–342.
    [28] Rawat AS, Vishwanath S, Bhowmick A, et al. Update efficient codes for distributed storage. Proceedings of the 2011 IEEE International Symposium on Information Theory. St. Petersburg: IEEE, 2011. 1457–1461.
    [29] Hashem IAT, Anuar NB, Marjani M, et al. Multi-objective scheduling of MapReduce jobs in big data processing. Multimedia Tools and Applications, 2018, 77(8): 9979–9994. [doi: 10.1007/s11042-017-4685-y
    [30] Jiang YW, Zhou P, Cheng TCE, et al. Optimal online algorithms for MapReduce scheduling on two uniform machines. Optimization Letters, 2019, 13(7): 1663–1676. [doi: 10.1007/s11590-018-01384-8
    [31] Naik NS, Negi A, Br TB, et al. A data locality based scheduler to enhance MapReduce performance in heterogeneous environments. Future Generation Computer Systems, 2019, 90: 423–434. [doi: 10.1016/j.future.2018.07.043
    [32] Darrous J. Scalable and efficient data management in distributed clouds: Service provisioning and data processing [Ph.D. thesis]. Lyon: Université de Lyon, 2019.
    [33] Ford D, Labelle F, Popovici FI, et al. Availability in globally distributed storage systems. Proceedings of the 9th USENIX Symposium on Operating Systems Design and Implementation. Vancouver: USENIX Association, 2010. 61–74.
    [34] Ahmad F, Chakradhar ST, Raghunathan A, et al. ShuffleWatcher: Shuffle-aware scheduling in multi-tenant MapReduce clusters. Proceedings of the 2014 USENIX Conference on Annual Technical Conference. Philadelphia: USENIX Association, 2014. 1–12.
    [35] Intel. Intel-bigdata/HiBench. https://github.com/Intel-bigdata/HiBench. (2020-06-20)[2022-08-03].
    Cited by
    Comments
    Comments
    分享到微博
    Submit
Get Citation

杨振宇,牛天洋,吕敏.混合存储模式下MapReduce作业调度.计算机系统应用,2023,32(3):70-85

Copy
Share
Article Metrics
  • Abstract:720
  • PDF: 2004
  • HTML: 1534
  • Cited by: 0
History
  • Received:August 09,2022
  • Revised:September 15,2022
  • Online: December 09,2022
Article QR Code
You are the first990387Visitors
Copyright: Institute of Software, Chinese Academy of Sciences Beijing ICP No. 05046678-3
Address:4# South Fourth Street, Zhongguancun,Haidian, Beijing,Postal Code:100190
Phone:010-62661041 Fax: Email:csa (a) iscas.ac.cn
Technical Support:Beijing Qinyun Technology Development Co., Ltd.

Beijing Public Network Security No. 11040202500063