DOI QR코드

DOI QR Code

An Efficient Hybrid Replication Protocol for High Available Distributed System

고 가용성 분산 시스템을 위한 효율적인 하이브리드 복제 프로토콜

  • 윤희용 (성균관대학교 정보통신공학부) ;
  • 최성춘 (성균관대학교 정보통신공학부)
  • Published : 2005.04.01

Abstract

In distributed systems data are replicated and stored at several nodes to increase the availability and overall performance. Here Quorum protocol doffing a certain set of replicas required for read/write operation exists for global concurrency control. One of the representative replication Protocols - the Tree Quorum protocol - has a drawback of rapidly growing number of replicas as the level increases, while the Grid protocol requires the same operation cost even without any failure. In this paper, thus, we propose a new replication protocol called hybrid protocol which capitalizes the merits of the existing protocols and solves the problems of them at the same time. The proposed hybrid protocol has very low operation cost in the absence of failure like the tree quorum protocol, and has relatively lower operation cost and higher availability than existing protocols when failure occurs by employing tree architecture as the overall organization while each level of the tree is organized as a row of a grid architecture. It is thus effective to be applied to survival storage system. We conduct cost and availability analysis of the proposed protocol through mathematical modeling, and response time and throughput are compared with those of the Tree Quorum protocol through computer simulation.

분산 시스템에서 가용성을 높이고 전체 시스템의 성능을 향상시키기 위해 데이터는 여러 노드에 중복하여 저장된다. 여기서는 전역적 접근제어를 위해서 읽기/쓰기 동작을 수행하는데 필요한 노드의 집합을 정의하는 Quorum 프로토콜이 존재한다. Quorum 프로토콜을 사용하는 대표적인 복제 프로토콜인 Tree Quorum 프로토콜은 트리의 높이가 증가할수록 노드의 수가 기하급수적으로 증가하고, Grid 프로토콜은 노드에 장애가 발생하지 않아도 언제나 같은 읽기/쓰기 비용을 갖는다는 단점을 갖고 있다. 따라서, 본 논문에서는 기존 프로토콜의 장점을 가지면서 단점을 해결할 수 있는 새로운 하이브리드 프로토콜을 제안한다. 제안된 하이브리드 프로토콜은 전체적으로는 트리 구조를 가지면서 각 레벨에서는 그리드의 열과 같은 구조를 가짐으로써 노드에 장애가 없을 때에는 Tree Quorum 프로토콜과 같이 적은 동작 비용을 요구하며, 노드에 장애가 존재할 경우에도 기존 프로토콜에 비해 상대적으로 적은 동자 비용과 높은 가용성을 보인다. 그러므로 높은 데이터 가용성이 요구되는 서바이벌 스토리지 시스템에 효율적으로 적용 가능하다. 본 논문에서는 수학적 모델링을 통하여 제안된 프로토콜의 비용과 가용성을 평가하고, 시뮬레이션을 통해 응답시간과 처리율을 기존의 Tree quorum프로토콜과 비교한다.

Keywords

References

  1. C. Amza., A.L. Cox, W. Zwaenepoel, Data replication strategies for fault tolerance and availability on commodity clusters, Proc. Int'l Conf on Dependable Systems and Networks (DSN), 2000, 459-467 https://doi.org/10.1109/ICDSN.2000.857576
  2. K. Arai, K. Tanaka, M. Takizawa, Group protocol for quorum-based replication, Proc. Seventh Int'l Conf on Parallel and Distributed Systems, 2000, 57-64 https://doi.org/10.1109/ICPADS.2000.857683
  3. G. Alonso, Partial Database Replication and Group Communication Primitives, Proc. of the 2nd European Research Seminar on Advances in Distributed Systems (ERSADS'97), March 1997, 171-176
  4. R.H. Thomas. A Majority Consensus Approach to Concurrency Control for Multiple Copy Data-based, ACM Trans on Database Systems, 4(2),1979,180-207 https://doi.org/10.1145/320071.320076
  5. D. Davcev, A Dynamic Voting Scheme in Distributed Systems. IEEE Trans on Software Engineering, 15(1), 1989, 93-97 https://doi.org/10.1109/32.21731
  6. D. Saha, S. Rangarajan, S.K. Tripathi, An Analysis of the Average Message Overhead in Replica Control Protocols, IEEE Trans an Parallel and Distributed Systems, 7(10), Oct. 1996, 1026-1034 https://doi.org/10.1109/71.539734
  7. P.A. Bernstein and N. Goodman, An Algorithm for Concurrency Control and Recovery in Replicated Distributed Databases, ACM Trans on Distributed Systems, 9(4), 1984, 596-615 https://doi.org/10.1145/1994.2207
  8. D. Freisleben, H.H. Koch, and O. Theel, Designing Multi-Level Quorum Schemes for Highly Replicated Data. Proc. of the 1991 Pacific Rim Int'l Symp on Fault Tolerant Systems, IEEE, 1991, 154-159 https://doi.org/10.1109/{RFTS.1991.212950
  9. D.K. Gifford, Weighted Voting for Replicated Data, Proc. of the 7th ACM Symp on Operating Systems Principles, 1979, 150-162 https://doi.org/10.1145/800215.806583
  10. A. Kumar, Hierarchical quotum consensus : A new algorithm for managing replicated data', IEEE Transaction on Computer, 40(9), September 1991, 996-1004 https://doi.org/10.1109/12.83661
  11. S. Cheung, M. Ammar, and M. Ahamad, The Grid Protocol: A High Performance Scheme for Maintaining Replicated Data, Proc of the 6th Int'l Conf on Data Engineering, 1990, 438-445 https://doi.org/10.1109/ICDE.1990.113497
  12. T. Anderson, Y. Breitbart, H. Korth, A Wool, Replication, Consistency, and Practicality: Are These Mutually Exclusive', ACM SIGMOD Int'l Conf on Management of Data, June 1998, 484-495 https://doi.org/10.1145/276304.276347
  13. D. Agrawal and A. El Abbadi, The tree Quorum protocol: An Efficient Approach for Managing Replicated Data, Proc of the 16th Very Large Databases (VLDB) Conf, 1990, 243-254