SlideShare a Scribd company logo
1 of 2
Download to read offline
3.10.2016 DSF ­ Data Sharing Facility
http://www.research.ibm.com/haifa/projects/systems/dsf.html 1/2
Israel 
   Home  |  Products & services  |  Support & downloads  |  My account
 Select a country
 IBM Research Home
 IBM Haifa Labs Home
DSF ­ Data
Sharing
Facility
Storage Research
Project Homepage
 · Publications
 · zFS ­ The next
Generation
 · Contact Information
Feedback
DSF ­ Data Sharing Facility
Data Sharing Facility (DSF) is an experimental project to build a serverless file system that
distributes all aspects of file and storage management over cooperating machines
interconnected by a fast switched network. 
DSF is aimed at building a system that would scale from several to several hundred
machines using commodity components. The first prototype is based on a cluster of NT
workstations connected by a Fast Ethernet of 100Mbps; DSF will be perceived by each
workstation as a native file system (i.e., a file system driver in NT terminology). 
DSF contains four types of entities: Clients, File Managers, Storage Managers, and Logical
Volumes. When configured together over a fast network, these entities provide a global
memory cache, distributed file management, and distributed storage repository. 
DSF Client – runs on some workstation and provides access to DSF files and directories.
The DSF Client maintains a memory cache of data blocks accessed by applications on the
Client workstation. The Client accepts file system requests from user programs, sends data
to Storage Managers on writes, forwards reads to File Managers on cache misses, and
receives replies from Storage Managers or other Clients. It also answers forwarding
requests from File Managers by sending data to other Clients. 
DSF File Manager – manages the metadata and cache consistency for a subset of DSF files.
To provide scalable service, DSF splits the management of its files among several File
Managers. The Manager of a file controls two sets of information about the file: cache
consistency state and file structure metadata blocks. Together, these structures allow the File
Manager to track all copies of the file's data blocks. The File Manager can thus forward
Client read requests to other Clients, thereby implementing cooperative client caching. 
DSF Storage Manager – stores and maintains data and metadata blocks on its local disks.
The Storage Manager reacts to requests from File Managers by supplying data to Clients
that have initiated I/O operations. DSF Storage Managers contain the intelligence to support
DSF Logical Volumes. 
DSF Logical Volume – consists of a number of logical disks that span multiple Storage
 
3.10.2016 DSF ­ Data Sharing Facility
http://www.research.ibm.com/haifa/projects/systems/dsf.html 2/2
Manager machines and provides an abstract interface to disk storage with a rich set of
recovery properties. The logical volume hides the physical distribution of its logical disks so
that new disks and Storage Managers can be dynamically incorporated into the system
without interrupting system operation. This increases storage capacity and throughput. The
storage system may be further reconfigured by moving disks between Storage Managers to
match different working environments and workloads. 
The logical volume supports a rich set of block operations, such as: read, write,
allocateAndWrite, allocate, and delete. Furthermore, special care is taken when handling
metadata blocks, e.g., inodes and indirect blocks. Metadata block operations include soft­
write, soft­allocate, and soft­delete, which can be committed or aborted later on, and thus be
efficiently used for multi­disk atomic transactions. 
The performance and scalability of DSF is affected by several important design features:
Separation of storage from file management
Caching and metadata management can be done on a machine that is different from
the one that stores the data.
Dynamic distribution of file management across multiple machines at file open
time
This offers superior performance and scalability compared to traditional server­
based file systems. Changing the management machine for a particular file can be
done for load balancing, or in response to system configuration changes such as
machine addition or deletion. This also enables dynamic load balancing of the
system at file open time and avoids the problems associated with static load
balancing at discrete time intervals.
Cooperative caching
Client machine memories are treated as one global cooperative cache. Clients are
able to access blocks cached by other clients, thereby reducing the load on Storage
Managers and reducing the cost of local cache misses.
Lack of dedicated machines
This eliminates source bottlenecks. Any machine in the system, including one that
runs user applications, can be made responsible for storing, caching, and managing
any piece of data or metadata. Furthermore, any machine in the system can assume
the responsibilities of a failed component.
Freedom to configure the file system
DSF can be configured to match different system environments depending on
machine memories and CPU speeds. DSF can have multiple configurations ranging
from a "small office system", where the file system is shared between two machines
and only one is responsible for storing the data, to a large "clustered system" of
hundreds of machines such as GPFS, where each machine is made responsible for
storing, caching and managing the file system.
Use of logical volumes
DSF logical volume can be used to dynamically reconfigure the storage subsystem
without interrupting file system operation. The support for transactional operations
over multiple disks improves the performance of file operations that require atomic
multi­block writes, such as file sync and directory operations. The new
allocateAndWrite technique removes the need to allocate a new block prior to its
writing to disk.
  About IBM  |  Privacy  |  Legal  |  Contact

More Related Content

What's hot

Demystifying Cloud: What is Cloud?
Demystifying Cloud: What is Cloud?Demystifying Cloud: What is Cloud?
Demystifying Cloud: What is Cloud?sriramr
 
Distributed processing
Distributed processingDistributed processing
Distributed processingNeil Stein
 
Network attached storage different from traditional file servers & implemen
Network attached storage different from traditional file servers & implemenNetwork attached storage different from traditional file servers & implemen
Network attached storage different from traditional file servers & implemenIAEME Publication
 
Network Attached Storage (NAS) Initiative
Network Attached Storage (NAS) Initiative Network Attached Storage (NAS) Initiative
Network Attached Storage (NAS) Initiative Gary Wilhelm
 
CDW: SAN vs. NAS
CDW: SAN vs. NASCDW: SAN vs. NAS
CDW: SAN vs. NASSpiceworks
 
Personal storage to enterprise storage system journey
Personal storage to enterprise storage system journeyPersonal storage to enterprise storage system journey
Personal storage to enterprise storage system journeySoumen Sarkar
 
Net App Unified Storage Architecture
Net App Unified Storage ArchitectureNet App Unified Storage Architecture
Net App Unified Storage Architecturenburgett
 
Netapp online training-30Hours Classes, 30Hours Lab,Assignments, Project
Netapp online training-30Hours Classes, 30Hours Lab,Assignments, ProjectNetapp online training-30Hours Classes, 30Hours Lab,Assignments, Project
Netapp online training-30Hours Classes, 30Hours Lab,Assignments, ProjectVidhyalive
 
Storage Efficiency Customer Success Stories Sept 2010 power point
Storage Efficiency Customer Success Stories Sept 2010 power pointStorage Efficiency Customer Success Stories Sept 2010 power point
Storage Efficiency Customer Success Stories Sept 2010 power pointMichael Hudak
 
FAQ on Dedupe NetApp
FAQ on Dedupe NetAppFAQ on Dedupe NetApp
FAQ on Dedupe NetAppAshwin Pawar
 
Network Attached Storage (NAS)
Network Attached Storage (NAS) Network Attached Storage (NAS)
Network Attached Storage (NAS) Kshitij Shete
 
Storage Area Network (San)
Storage Area Network (San)Storage Area Network (San)
Storage Area Network (San)sankcomp
 
Understanding das-nas-san
Understanding das-nas-sanUnderstanding das-nas-san
Understanding das-nas-sanAshwin Pawar
 

What's hot (20)

Demystifying Cloud: What is Cloud?
Demystifying Cloud: What is Cloud?Demystifying Cloud: What is Cloud?
Demystifying Cloud: What is Cloud?
 
Distributed processing
Distributed processingDistributed processing
Distributed processing
 
Storage Basics
Storage BasicsStorage Basics
Storage Basics
 
Network attached storage different from traditional file servers & implemen
Network attached storage different from traditional file servers & implemenNetwork attached storage different from traditional file servers & implemen
Network attached storage different from traditional file servers & implemen
 
Network Attached Storage (NAS) Initiative
Network Attached Storage (NAS) Initiative Network Attached Storage (NAS) Initiative
Network Attached Storage (NAS) Initiative
 
CDW: SAN vs. NAS
CDW: SAN vs. NASCDW: SAN vs. NAS
CDW: SAN vs. NAS
 
Personal storage to enterprise storage system journey
Personal storage to enterprise storage system journeyPersonal storage to enterprise storage system journey
Personal storage to enterprise storage system journey
 
Net App Unified Storage Architecture
Net App Unified Storage ArchitectureNet App Unified Storage Architecture
Net App Unified Storage Architecture
 
Netapp online training-30Hours Classes, 30Hours Lab,Assignments, Project
Netapp online training-30Hours Classes, 30Hours Lab,Assignments, ProjectNetapp online training-30Hours Classes, 30Hours Lab,Assignments, Project
Netapp online training-30Hours Classes, 30Hours Lab,Assignments, Project
 
Computer cluster
Computer clusterComputer cluster
Computer cluster
 
San nas-
San nas-San nas-
San nas-
 
Storage Efficiency Customer Success Stories Sept 2010 power point
Storage Efficiency Customer Success Stories Sept 2010 power pointStorage Efficiency Customer Success Stories Sept 2010 power point
Storage Efficiency Customer Success Stories Sept 2010 power point
 
Nas Ashok1
Nas Ashok1Nas Ashok1
Nas Ashok1
 
Cluster computer
Cluster  computerCluster  computer
Cluster computer
 
Storage basics
Storage basicsStorage basics
Storage basics
 
FAQ on Dedupe NetApp
FAQ on Dedupe NetAppFAQ on Dedupe NetApp
FAQ on Dedupe NetApp
 
Network Attached Storage (NAS)
Network Attached Storage (NAS) Network Attached Storage (NAS)
Network Attached Storage (NAS)
 
Storage Networks
Storage NetworksStorage Networks
Storage Networks
 
Storage Area Network (San)
Storage Area Network (San)Storage Area Network (San)
Storage Area Network (San)
 
Understanding das-nas-san
Understanding das-nas-sanUnderstanding das-nas-san
Understanding das-nas-san
 

Viewers also liked

HDFS for Geographically Distributed File System
HDFS for Geographically Distributed File SystemHDFS for Geographically Distributed File System
HDFS for Geographically Distributed File SystemKonstantin V. Shvachko
 
Use Distributed Filesystem as a Storage Tier
Use Distributed Filesystem as a Storage TierUse Distributed Filesystem as a Storage Tier
Use Distributed Filesystem as a Storage TierManfred Furuholmen
 
File models and file accessing models
File models and file accessing modelsFile models and file accessing models
File models and file accessing modelsishmecse13
 
Dfs (Distributed computing)
Dfs (Distributed computing)Dfs (Distributed computing)
Dfs (Distributed computing)Sri Prasanna
 
Hadoop distributed file system
Hadoop distributed file systemHadoop distributed file system
Hadoop distributed file systemAnshul Bhatnagar
 
Distributed File Systems
Distributed File Systems Distributed File Systems
Distributed File Systems Maurvi04
 
Distributed File Systems: An Overview
Distributed File Systems: An OverviewDistributed File Systems: An Overview
Distributed File Systems: An OverviewAnant Narayanan
 
Dfs presentation
Dfs presentationDfs presentation
Dfs presentationAlizay Khan
 
Distributed File Systems
Distributed File SystemsDistributed File Systems
Distributed File Systemsawesomesos
 
3. distributed file system requirements
3. distributed file system requirements3. distributed file system requirements
3. distributed file system requirementsAbDul ThaYyal
 
Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...
Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...
Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...Cloudera, Inc.
 

Viewers also liked (11)

HDFS for Geographically Distributed File System
HDFS for Geographically Distributed File SystemHDFS for Geographically Distributed File System
HDFS for Geographically Distributed File System
 
Use Distributed Filesystem as a Storage Tier
Use Distributed Filesystem as a Storage TierUse Distributed Filesystem as a Storage Tier
Use Distributed Filesystem as a Storage Tier
 
File models and file accessing models
File models and file accessing modelsFile models and file accessing models
File models and file accessing models
 
Dfs (Distributed computing)
Dfs (Distributed computing)Dfs (Distributed computing)
Dfs (Distributed computing)
 
Hadoop distributed file system
Hadoop distributed file systemHadoop distributed file system
Hadoop distributed file system
 
Distributed File Systems
Distributed File Systems Distributed File Systems
Distributed File Systems
 
Distributed File Systems: An Overview
Distributed File Systems: An OverviewDistributed File Systems: An Overview
Distributed File Systems: An Overview
 
Dfs presentation
Dfs presentationDfs presentation
Dfs presentation
 
Distributed File Systems
Distributed File SystemsDistributed File Systems
Distributed File Systems
 
3. distributed file system requirements
3. distributed file system requirements3. distributed file system requirements
3. distributed file system requirements
 
Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...
Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...
Hadoop Distributed File System (HDFS) Encryption with Cloudera Navigator Key ...
 

Similar to DSF - Data Sharing Facility

PARALLEL FILE SYSTEM FOR LINUX CLUSTERS
PARALLEL FILE SYSTEM FOR LINUX CLUSTERSPARALLEL FILE SYSTEM FOR LINUX CLUSTERS
PARALLEL FILE SYSTEM FOR LINUX CLUSTERSRaheemUnnisa1
 
OSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio Manfred
OSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio ManfredOSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio Manfred
OSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio ManfredNETWAYS
 
DFS PPT.pptx
DFS PPT.pptxDFS PPT.pptx
DFS PPT.pptxVMahesh5
 
Unit-1 Introduction to Big Data.pptx
Unit-1 Introduction to Big Data.pptxUnit-1 Introduction to Big Data.pptx
Unit-1 Introduction to Big Data.pptxAnkitChauhan817826
 
final-unit-ii-cc-cloud computing-2022.pdf
final-unit-ii-cc-cloud computing-2022.pdffinal-unit-ii-cc-cloud computing-2022.pdf
final-unit-ii-cc-cloud computing-2022.pdfSamiksha880257
 
S014066 scale-ess-orlando-v1705a
S014066 scale-ess-orlando-v1705aS014066 scale-ess-orlando-v1705a
S014066 scale-ess-orlando-v1705aTony Pearson
 
Big data interview questions and answers
Big data interview questions and answersBig data interview questions and answers
Big data interview questions and answersKalyan Hadoop
 
Authenticated Key Exchange Protocols for Parallel Network File Systems
Authenticated Key Exchange Protocols for Parallel Network File SystemsAuthenticated Key Exchange Protocols for Parallel Network File Systems
Authenticated Key Exchange Protocols for Parallel Network File Systems1crore projects
 
Consolidating File Servers into the Cloud
Consolidating File Servers into the CloudConsolidating File Servers into the Cloud
Consolidating File Servers into the CloudBuurst
 
EMC Isilon Multitenancy for Hadoop Big Data Analytics
EMC Isilon Multitenancy for Hadoop Big Data AnalyticsEMC Isilon Multitenancy for Hadoop Big Data Analytics
EMC Isilon Multitenancy for Hadoop Big Data AnalyticsEMC
 
Authenticated key exchange protocols for parallel network file systems
Authenticated key exchange protocols for parallel network file systemsAuthenticated key exchange protocols for parallel network file systems
Authenticated key exchange protocols for parallel network file systemsPvrtechnologies Nellore
 
7. emc isilon hdfs enterprise storage for hadoop
7. emc isilon hdfs   enterprise storage for hadoop7. emc isilon hdfs   enterprise storage for hadoop
7. emc isilon hdfs enterprise storage for hadoopTaldor Group
 
Liberate Your Files with a Private Cloud Storage Solution powered by Open Source
Liberate Your Files with a Private Cloud Storage Solution powered by Open SourceLiberate Your Files with a Private Cloud Storage Solution powered by Open Source
Liberate Your Files with a Private Cloud Storage Solution powered by Open SourceIsaac Christoffersen
 
Azure fb-google Web Services
Azure fb-google Web ServicesAzure fb-google Web Services
Azure fb-google Web ServicesShreya Srivastava
 
Large Scale On-Demand Image Processing For Disaster Relief
Large Scale On-Demand Image Processing For Disaster ReliefLarge Scale On-Demand Image Processing For Disaster Relief
Large Scale On-Demand Image Processing For Disaster ReliefRobert Grossman
 

Similar to DSF - Data Sharing Facility (20)

PARALLEL FILE SYSTEM FOR LINUX CLUSTERS
PARALLEL FILE SYSTEM FOR LINUX CLUSTERSPARALLEL FILE SYSTEM FOR LINUX CLUSTERS
PARALLEL FILE SYSTEM FOR LINUX CLUSTERS
 
Nas fundamentals
Nas fundamentalsNas fundamentals
Nas fundamentals
 
Hadoop data management
Hadoop data managementHadoop data management
Hadoop data management
 
Hadoop
HadoopHadoop
Hadoop
 
OSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio Manfred
OSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio ManfredOSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio Manfred
OSDC 2010 | Use Distributed Filesystem as a Storage Tier by Fabrizio Manfred
 
DFS PPT.pptx
DFS PPT.pptxDFS PPT.pptx
DFS PPT.pptx
 
Unit-1 Introduction to Big Data.pptx
Unit-1 Introduction to Big Data.pptxUnit-1 Introduction to Big Data.pptx
Unit-1 Introduction to Big Data.pptx
 
final-unit-ii-cc-cloud computing-2022.pdf
final-unit-ii-cc-cloud computing-2022.pdffinal-unit-ii-cc-cloud computing-2022.pdf
final-unit-ii-cc-cloud computing-2022.pdf
 
S014066 scale-ess-orlando-v1705a
S014066 scale-ess-orlando-v1705aS014066 scale-ess-orlando-v1705a
S014066 scale-ess-orlando-v1705a
 
Big data interview questions and answers
Big data interview questions and answersBig data interview questions and answers
Big data interview questions and answers
 
Authenticated Key Exchange Protocols for Parallel Network File Systems
Authenticated Key Exchange Protocols for Parallel Network File SystemsAuthenticated Key Exchange Protocols for Parallel Network File Systems
Authenticated Key Exchange Protocols for Parallel Network File Systems
 
Consolidating File Servers into the Cloud
Consolidating File Servers into the CloudConsolidating File Servers into the Cloud
Consolidating File Servers into the Cloud
 
Bsd ppt
Bsd pptBsd ppt
Bsd ppt
 
EMC Isilon Multitenancy for Hadoop Big Data Analytics
EMC Isilon Multitenancy for Hadoop Big Data AnalyticsEMC Isilon Multitenancy for Hadoop Big Data Analytics
EMC Isilon Multitenancy for Hadoop Big Data Analytics
 
Authenticated key exchange protocols for parallel network file systems
Authenticated key exchange protocols for parallel network file systemsAuthenticated key exchange protocols for parallel network file systems
Authenticated key exchange protocols for parallel network file systems
 
7. emc isilon hdfs enterprise storage for hadoop
7. emc isilon hdfs   enterprise storage for hadoop7. emc isilon hdfs   enterprise storage for hadoop
7. emc isilon hdfs enterprise storage for hadoop
 
module 2.pptx
module 2.pptxmodule 2.pptx
module 2.pptx
 
Liberate Your Files with a Private Cloud Storage Solution powered by Open Source
Liberate Your Files with a Private Cloud Storage Solution powered by Open SourceLiberate Your Files with a Private Cloud Storage Solution powered by Open Source
Liberate Your Files with a Private Cloud Storage Solution powered by Open Source
 
Azure fb-google Web Services
Azure fb-google Web ServicesAzure fb-google Web Services
Azure fb-google Web Services
 
Large Scale On-Demand Image Processing For Disaster Relief
Large Scale On-Demand Image Processing For Disaster ReliefLarge Scale On-Demand Image Processing For Disaster Relief
Large Scale On-Demand Image Processing For Disaster Relief
 

DSF - Data Sharing Facility