Fragmentation in Large Object Repositories

نویسندگان

  • Russell Sears
  • Catharine van Ingen
چکیده

Fragmentation leads to unpredictable and degraded application performance. While these problems have been studied in detail for desktop filesystem workloads, this study examines newer systems such as scalable object stores and multimedia repositories. Such systems use a get/put interface to store objects. In principle, databases and filesystems can support such applications efficiently, allowing system designers to focus on complexity, deployment cost and manageability. Although theoretical work proves that certain storage policies behave optimally for some workloads, these policies often behave poorly in practice. Most storage benchmarks focus on short-term behavior or do not measure fragmentation. We compare SQL Server to NTFS and find that fragmentation dominates performance when object sizes exceed 256KB-1MB. NTFS handles fragmentation better than SQL Server. Although the performance curves will vary with other systems and workloads, we expect the same interactions between fragmentation and free space to apply. It is well-known that fragmentation is related to the percentage free space. We found that the ratio of free space to object size also impacts performance. Surprisingly, in both systems, storing objects of a single size causes fragmentation, and changing the size of write requests affects fragmentation. These problems could be addressed with simple changes to the filesystem and database interfaces. It is our hope that an improved understanding of fragmentation will lead to predictable storage systems that require less maintenance after deployment.

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

An Integration Model of Distributed Objects in a Virtual Repository Implementation basing on The ODRA OODBMS1

The paper deals with integration of distributed fragmented collections of data, being the basis for virtual repositories in the data grid or P2P architecture. The core of the described architecture is based on the StackBased Query Language (SBQL) and virtual updateable SBQL views. Our virtual repository transparently integrates distributed, heterogeneous and fragmented data producing conceptual...

متن کامل

Numerical Study of Spherical Vapor Layer Growth Due to Contact of a Hot Object and Water

Vapor film formation and growth due to contact of a hot body and other liquids arise in some industrial applications including nuclear fuel rods, foundry and production of paper. The possibility of a steam explosion remains in most of these cases which could result in injuries and financial damage. Due to the importance of such phenomenon, this study deals with vapor layer forming, growth, and ...

متن کامل

Transparency in Object-Oriented Grid Database Systems

The paper presents various transparency issues that have to be considered during development of object-oriented Grid applications based on virtual repositories. Higher-level transparencies, such as location, heterogeneity, fragmentation, replication, redundancy, indexing and service provider transparency assure new information processing culture greatly supporting the development, operation and...

متن کامل

Distributed XML Repositories: Top-down Design and Transparent Query Processing

XML is increasingly used not only for data exchange but also to represent arbitrary data sources as virtual XML repositories. In many application scenarios, fragments of such repositories are distributed over the Web. However, design and query processing models for distributed XML data have not yet been studied in detail. The goal of this paper is to study the design and management of distribut...

متن کامل

Complete metadata records in learning object repositories: some evidence and requirements

A learning object can be considered as a unit of instructional content for which a metadata record describing its characteristics and intended educational usage is provided. Metadata records can be used to develop effective search and location of learning objects, and also to develop automated or semi-automated selection and composition tools. In consequence, the quality of metadata records is ...

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:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007