Data placement schemes in replicated mirrored disk systems

نویسندگان

  • Athena Vakali
  • Yannis Manolopoulos
چکیده

In the present paper, we study the most frequent data placement schemes, i.e., the organ-pipe and camel arrangements, in a mirrored disk system which supports cylinder replication at appropriate locations across the disk surface. Five schemes are proposed to identify the appropriate replica positions for the most frequently accessed cylinders. Three schemes, which are called Left, Right and Symmetric Replication Techniques, are based on an analytical approach. More speci®cally, given a single replica in the above strategies, estimates are derived for the expected seek distance decrease with respect to a non-replicated mirrored disk system. On the other hand, the remaining two schemes, which are called Positional and Frequency Replication Techniques, are based on heuristic approaches. We globally compare the performance of these ®ve schemes as a function of the number of replicas. These results are also compared with the corresponding metrics of a conventional (with no replication) mirrored disk system. Ó 2000 Elsevier Science Inc. All rights reserved.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Novel Replica Placement Strategy for Video Servers

Mirroring-based reliability as compared to parity-based reliability signi cantly simpli es the design and the implementation of video servers, since in case of failure mirroring does not require any synchronization of reads or decoding to reconstruct the lost video data. While mirroring doubles the amount of storage volume required, the steep decrease of the cost of magnetic disk storage makes ...

متن کامل

Mirrored and Hybrid Disk Arrays: Organization, Scheduling, Reliability, and Performance

Basic mirroring (BM) classified as RAID level 1 replicates data on two disks, thus doubling disk access bandwidth for read requests. RAID1/0 is an array of BM pairs with balanced loads due to striping. When a disk fails the read load on its pair is doubled, which results in halving the maximum attainable bandwidth. We review RAID1 organizations which attain a balanced load upon disk failure, bu...

متن کامل

Placement of Mirrored Web Servers in Ring Networks with Budget Constraint

This paper investigates the placement problem of mirrored web servers in ring networks, where each client accesses the closest mirrored server. Without budget constraint, we propose an efficient algorithm that computes the optimal placement with time complexity . With budget constraint, if all candidate servers charge the same fee, we propose an algorithm to compute the optimal placement in tim...

متن کامل

Coding-Based Replication Schemes for Distributed Systems

Data is often replicated in distributed systems to improve availability and performance. This replication is expensive in terms of disk storage since the existing schemes generally require full files to be stored at each site. In this paper, we present schemes which significantly reduce the storage requirements in replication based systems. These schemes use the coding method suggested by Rabin...

متن کامل

Using Coding to Support Data Resiliency in Distributed Systems

Data is often replicated in distributed systems to improve availability and performance. This replication is expensive in terms of disk storage since the existing schemes generally require full les to be stored at each site. In this paper, we present how the coding scheme suggested by Rabin can be used to store replicated data and how the voting algorithm and the quorum requirements change to m...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • Journal of Systems and Software

دوره 55  شماره 

صفحات  -

تاریخ انتشار 2000