A Cost-Effective, High-Bandwidth Storage Architecture Garth A. Gibson, David F. Nagle, Khalil Amiri, Jeff Butler, Fay W. Chang, Howard Gobioff, Charles.

Slides:



Advertisements
Similar presentations
System Area Network Abhiram Shandilya 12/06/01. Overview Introduction to System Area Networks SAN Design and Examples SAN Applications.
Advertisements

NAS vs. SAN 10/2010 Palestinian Land Authority IT Department By Nahreen Ameen 1.
Network Storage and Cluster File Systems Jeff Chase CPS 212, Fall 2000.
Ceph: A Scalable, High-Performance Distributed File System Sage Weil Scott Brandt Ethan Miller Darrell Long Carlos Maltzahn University of California, Santa.
Network-Attached Storage
S. Barua – CPSC 440 CHAPTER 8 INTERFACING PROCESSORS AND PERIPHERALS Topics to be covered  How to.
1 Recap (RAID and Storage Architectures). 2 RAID To increase the availability and the performance (bandwidth) of a storage system, instead of a single.
Asper School of Business University of Manitoba Systems Analysis & Design Instructor: Bob Travica System architectures Updated: November 2014.
EEC-681/781 Distributed Computing Systems Lecture 3 Wenbing Zhao Department of Electrical and Computer Engineering Cleveland State University
1 Netprog 2002 Network Terminology Motivation, Terminology, Layered systems (and other random stuff)
Database System Architectures  Client-server Database System  Parallel Database System  Distributed Database System Wei Jiang.
Storage Area Network (SAN)
Storage Networking Technologies and Virtualization Section 2 DAS and Introduction to SCSI1.
Module – 7 network-attached storage (NAS)
RAID-x: A New Distributed Disk Array for I/O-Centric Cluster Computing Kai Hwang, Hai Jin, and Roy Ho.
Frangipani: A Scalable Distributed File System C. A. Thekkath, T. Mann, and E. K. Lee Systems Research Center Digital Equipment Corporation.
File Systems and N/W attached storage (NAS) | VTU NOTES | QUESTION PAPERS | NEWS | VTU RESULTS | FORUM | BOOKSPAR ANDROID APP.
Sanjay Ghemawat, Howard Gobioff, and Shun-Tak Leung Google∗
Storage Systems in HPC John A. Chandy Department of Electrical and Computer Engineering University of Connecticut.
Chapter 2 Architectural Models. Keywords Middleware Interface vs. implementation Client-server models OOP.
Object-based Storage Long Liu Outline Why do we need object based storage? What is object based storage? How to take advantage of it? What's.
Mobile Databases: a Selection of Open Issues and Research Directions Authors: Rachid Guerraoui et al. Sources: SIGMOD Record, 33(2), pp.78-83, 2004 Adviser:
Distributed File Systems
Chapter 5 Section 2 : Storage Networking Technologies and Virtualization.
Garth A. Gibson*, David F. Nagle**, William Courtright II*, Nat Lanza*, Paul Mazaitis*, Marc Unangst*, Jim Zelenka* "NASD Scalable Storage Systems",USENIX99,
Latest Relevant Techniques and Applications for Distributed File Systems Ela Sharda
Khalil Amiri*, David Petrou, Gregory R. Ganger* and Garth A. Gibson "Dynamic Function Placement for Data-intensive Cluster Computing," Proceedings of the.
Scalable Web Server on Heterogeneous Cluster CHEN Ge.
DATABASE MANAGEMENT SYSTEMS IN DATA INTENSIVE ENVIRONMENNTS Leon Guzenda Chief Technology Officer.
What is a Distributed File System?? Allows transparent access to remote files over a network. Examples: Network File System (NFS) by Sun Microsystems.
The Client/Server Database Environment Ployphan Sornsuwit KPRU Ref.
Advanced Computer Networks Topic 2: Characterization of Distributed Systems.
Introduction to DFS. Distributed File Systems A file system whose clients, servers and storage devices are dispersed among the machines of a distributed.
Presenters: Rezan Amiri Sahar Delroshan
Large Scale Parallel File System and Cluster Management ICT, CAS.
Using NAS as a Gateway to SAN Dave Rosenberg Hewlett-Packard Company th Street SW Loveland, CO 80537
Server to Server Communication Redis as an enabler Orion Free
 Distributed file systems having transaction facility need to support distributed transaction service.  A distributed transaction service is an extension.
1 Database Management Systems (DBMS). 2 Database Management Systems (DBMS) n Overview of: ä Database Management Components ä Database Systems Architecture.
Eduardo Gutarra Velez. Outline Distributed Filesystems Motivation Google Filesystem Architecture The Metadata Consistency Model File Mutation.
INTRODUCTION TO DBS Database: a collection of data describing the activities of one or more related organizations DBMS: software designed to assist in.
Jinyong Yoon,  Andrew File System  The Prototype  Changes for Performance  Effect of Changes for Performance  Comparison with A Remote-Open.
© 2009 IBM Corporation Statements of IBM future plans and directions are provided for information purposes only. Plans and direction are subject to change.
HADOOP DISTRIBUTED FILE SYSTEM HDFS Reliability Based on “The Hadoop Distributed File System” K. Shvachko et al., MSST 2010 Michael Tsitrin 26/05/13.
The Mach System Silberschatz et al Presented By Anjana Venkat.
Mr. P. K. GuptaSandeep Gupta Roopak Agarwal
Accelerating High Performance Cluster Computing Through the Reduction of File System Latency David Fellinger Chief Scientist, DDN Storage ©2015 Dartadirect.
Internet Protocol Storage Area Networks (IP SAN)
AFS/OSD Project R.Belloni, L.Giammarino, A.Maslennikov, G.Palumbo, H.Reuter, R.Toebbicke.
Multimedia Retrieval Architecture Electrical Communication Engineering, Indian Institute of Science, Bangalore – , India Multimedia Retrieval Architecture.
HNC COMPUTING - Network Concepts 1 Network Concepts Network Concepts Network Operating Systems Network Operating Systems.
Self-Tuned Distributed Multiprocessor System Xiaoyan Bi CSC Operating Systems Dr. Mirela Damian.
1 CEG 2400 Fall 2012 Network Servers. 2 Network Servers Critical Network servers – Contain redundant components Power supplies Fans Memory CPU Hard Drives.
ARUN S CS-7 NO:6. HIGH SPEED OPTICAL CABLE TECHNOLOGY HIGH BANDWIDTH UNIVERSAL CONNECTOR SUPPORTS MULTIPLE PROTOCOLS  10Gb/s to 100Gb/s  single universal.
© 2007 EMC Corporation. All rights reserved. Internet Protocol Storage Area Networks (IP SAN) Module 3.4.
IT 5433 LM1. Learning Objectives Understand key terms in database Explain file processing systems List parts of a database environment Explain types of.
Chapter 16 Client/Server Computing Dave Bremer Otago Polytechnic, N.Z. ©2008, Prentice Hall Operating Systems: Internals and Design Principles, 6/E William.
Amazon Web Services. Amazon Web Services (AWS) - robust, scalable and affordable infrastructure for cloud computing. This session is about:
DISTRIBUTED FILE SYSTEM- ENHANCEMENT AND FURTHER DEVELOPMENT BY:- PALLAWI(10BIT0033)
Databases and DBMSs Todd S. Bacastow January 2005.
Direct Attached Storage and Introduction to SCSI
Chapter 1: Introduction
Parallel Data Laboratory, Carnegie Mellon University
Direct Attached Storage and Introduction to SCSI
Object-Based Storage Advanced Operating System HoSeok Seo
Building a Database on S3
Tiers vs. Layers.
Specialized Cloud Architectures
Outline Review of Quiz #1 Distributed File Systems 4/20/2019 COP5611.
The Gamma Database Machine Project
Presentation transcript:

A Cost-Effective, High-Bandwidth Storage Architecture Garth A. Gibson, David F. Nagle, Khalil Amiri, Jeff Butler, Fay W. Chang, Howard Gobioff, Charles Hardin, Erik Riedel, David Rochberg, Jim Zelenka

Topics What is NASD Goals of design and design principle Architecture of NASD File system NASD-PFS Performance Future work Conclusions Questions

NAPD Network-Attached Secure Disk Network-Attached Secure Disks (NASD) enable cost-effective bandwidth scaling. NASD eliminates the server bandwidth bottleneck by modifying storage devices to transfer data directly to clients and also repartitions traditional file server or database functionality between the drive, client and server.

Goals of design Cost-effective bandwidth scaling No traditional storage file server Secure

Design principles direct transfer to clients Data is transferred between drive and client without indirection or store-and-forward through a file server machine. secure interfaces via cryptographic support By attaching storage to the network, we open drives to direct attack from adversaries. Thus, it is necessary to apply cryptographic techniques to defend against potential attacks.

Design principles cont. asynchronous oversight The ability of the client to perform most operations without synchronous appeal to the file manager. variably-sized data objects To allow drives direct knowledge of the relationships between disk blocks and to minimize security overhead. This also improves opportunities for storage self-management by extending into a disk an understanding of the relationships between blocks on the disk.

Enabling technology I/O-bound applications Video,audio, application such as data mining of retail transactions, telecommunication call records New drive attachment technology Fibrechannel, a serial, switched, packet-based peripheral network Excess of on-drive transistors 0.68 micron CMOS->0.35 micron CMOS, put StrongArm, DRAM or cryptographic support on disk drive

Enabling technology cont. Convergence of peripheral and interprocessor networks Low latency network protocol, such as VIA which narrow the gap between the channel properties of peripheral interconnects and the network properties of client interconnects. Cost-ineffective storage servers

Architecture

File system Port NFS and AFS to NASD NFS: data moving operations( read, write ) and attribute reads ( getattr ) are directed to the NASD drive while all other requests are handled by the file manager. Andrew benchmark, NASD-NFS and NFS had benchmark times within 5% of each other. NASD-optimized parallel filesystem

File system cont. The filesystem manages objects which are not directly backed by data. Instead, they are backed by a storage manager, Cheops, which redirects clients to the underlying component NASD objects. Cheops maintains the mapping of higher-level objects to the objects on the individual devices. NASD PFS employs Cheops as its storage management.

NASD-PFS Performance NASD PFS delivers nearly all of the bandwidth of the NASD drives, while the same application using a powerful NFS server fails to deliver half the performance of the underlying Cheetah drives. One disk: 6.2MB/s

Future work Active disk This next generation of storage devices provides an execution environment directly at individual drives and allows code to execute near the data and before it is placed on the interconnect network. NASD enables this type of extension functionality for the first time because the object-based interface provides sufficient knowledge of the data at the individual devices without having to resort to external metadata.

Conclusions Traditional storage servers are not effective. Storage tranfering data directly on the client’s network will reduce the cost greatly NASD is cost-effective NASD can provide scalable bandwidth Conventional distributed filesystems(NFS and AFS) can be ported to use NASD with performance comparable to current server-based systems.

Questions What is NASD? Compared to storage servers, what is the major benefit of NASD? What the major properties of NASD? (four)