HDFS and Availability Data Retragement

Similar documents
Hadoop Architecture. Part 1

Welcome to the unit of Hadoop Fundamentals on Hadoop architecture. I will begin with a terminology review and then cover the major components

Distributed File Systems

Distributed Filesystems

Apache Hadoop FileSystem and its Usage in Facebook

BookKeeper. Flavio Junqueira Yahoo! Research, Barcelona. Hadoop in China 2011

HADOOP MOCK TEST HADOOP MOCK TEST I

Hadoop IST 734 SS CHUNG

Hadoop: A Framework for Data- Intensive Distributed Computing. CS561-Spring 2012 WPI, Mohamed Y. Eltabakh

Distributed File Systems

Server Virtualization with Windows Server Hyper-V and System Center

Hadoop Distributed File System (HDFS) Overview

Design and Evolution of the Apache Hadoop File System(HDFS)

Server Virtualization with Windows Server Hyper-V and System Center

THE HADOOP DISTRIBUTED FILE SYSTEM

Take An Internal Look at Hadoop. Hairong Kuang Grid Team, Yahoo! Inc

Server Virtualization with Windows Server Hyper-V and System Center

INTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY

Hadoop & its Usage at Facebook

Hadoop Distributed File System. Dhruba Borthakur June, 2007

Hadoop Distributed File System. Jordan Prosch, Matt Kipps

Distributed File System. MCSN N. Tonellotto Complements of Distributed Enabling Platforms

Server Virtualization with Windows Server Hyper-V and System Center

Journal of science STUDY ON REPLICA MANAGEMENT AND HIGH AVAILABILITY IN HADOOP DISTRIBUTED FILE SYSTEM (HDFS)

The Google File System

Big Data Technology Core Hadoop: HDFS-YARN Internals

HDFS Users Guide. Table of contents

Hadoop & its Usage at Facebook

Server Virtualization with Windows Server Hyper-V and System Center

Outline. MCSA: Server Virtualization

EMC IRODS RESOURCE DRIVERS

Hadoop Distributed File System. Dhruba Borthakur Apache Hadoop Project Management Committee June 3 rd, 2008

Apache Hadoop. Alexandru Costan

Intro to Map/Reduce a.k.a. Hadoop

Apache Hadoop FileSystem Internals

20409B: Server Virtualization with Windows Server Hyper-V and System Center

Chapter 7. Using Hadoop Cluster and MapReduce

Hadoop Scalability at Facebook. Dmytro Molkov YaC, Moscow, September 19, 2011

NoSQL Data Base Basics

Open source software framework designed for storage and processing of large scale data on clusters of commodity hardware

Pervasive PSQL Meets Critical Business Requirements

Brian LaGoe, Systems Administrator Benjamin Jellema, Systems Administrator Eastern Michigan University

HDFS Reliability. Tom White, Cloudera, 12 January 2008

Snapshots in Hadoop Distributed File System

"Charting the Course... MOC B Server Virtualization with Windows Hyper-V and System Center. Course Summary

Course Outline. Create and configure virtual hard disks. Create and configure virtual machines. Install and import virtual machines.

Server-Virtualisierung mit Windows Server Hyper-V und System Center MOC 20409

Distributed System Principles

Overview. Big Data in Apache Hadoop. - HDFS - MapReduce in Hadoop - YARN. Big Data Management and Analytics

High Availability and Disaster Recovery Solutions for Perforce

The Hadoop Distributed File System

Prepared By : Manoj Kumar Joshi & Vikas Sawhney

Introduction to Cloud : Cloud and Cloud Storage. Lecture 2. Dr. Dalit Naor IBM Haifa Research Storage Systems. Dalit Naor, IBM Haifa Research

Lecture 2 (08/31, 09/02, 09/09): Hadoop. Decisions, Operations & Information Technologies Robert H. Smith School of Business Fall, 2015

CS2510 Computer Operating Systems

CS2510 Computer Operating Systems

Research on Reliability of Hadoop Distributed File System

RAID. Tiffany Yu-Han Chen. # The performance of different RAID levels # read/write/reliability (fault-tolerant)/overhead

Westek Technology Snapshot and HA iscsi Replication Suite

International Journal of Advance Research in Computer Science and Management Studies

High Availability with Windows Server 2012 Release Candidate

Hadoop Distributed File System. T Seminar On Multimedia Eero Kurkela

Introduction to Hadoop. New York Oracle User Group Vikas Sawhney

Web DNS Peer-to-peer systems (file sharing, CDNs, cycle sharing)

Server Virtualization with Windows Server Hyper-V and System Center

Diagram 1: Islands of storage across a digital broadcast workflow

Storage Architectures for Big Data in the Cloud

Department of Computer Science University of Cyprus EPL646 Advanced Topics in Databases. Lecture 14

!"#$%&' ( )%#*'+,'-#.//"0( !"#$"%&'()*$+()',!-+.'/', 4(5,67,!-+!"89,:*$;'0+$.<.,&0$'09,&)"/=+,!()<>'0, 3, Processing LARGE data sets

HDFS Under the Hood. Sanjay Radia. Grid Computing, Hadoop Yahoo Inc.

Processing of Hadoop using Highly Available NameNode

HDFS Federation. Sanjay Radia Founder and Hortonworks. Page 1

CSE-E5430 Scalable Cloud Computing Lecture 2

Course Server Virtualization with Windows Server Hyper-V and System Center

Processing of massive data: MapReduce. 2. Hadoop. New Trends In Distributed Systems MSc Software and Systems

Big data management with IBM General Parallel File System

Big + Fast + Safe + Simple = Lowest Technical Risk

ATLAS Tier 3

Hadoop Distributed File System. Dhruba Borthakur Apache Hadoop Project Management Committee

Hadoop Architecture and its Usage at Facebook

SciDAC Petascale Data Storage Institute

Lecture 32 Big Data. 1. Big Data problem 2. Why the excitement about big data 3. What is MapReduce 4. What is Hadoop 5. Get started with Hadoop

Long term retention and archiving the challenges and the solution

Comparing the Hadoop Distributed File System (HDFS) with the Cassandra File System (CFS)

Accelerating and Simplifying Apache

Weekly Report. Hadoop Introduction. submitted By Anurag Sharma. Department of Computer Science and Engineering. Indian Institute of Technology Bombay

Enhancing UNICORE Storage Management using Hadoop

WOS Cloud. ddn.com. Personal Storage for the Enterprise. DDN Solution Brief

A Multilevel Secure MapReduce Framework for Cross-Domain Information Sharing in the Cloud

Comparative analysis of mapreduce job by keeping data constant and varying cluster size technique

Hadoop Distributed Filesystem. Spring 2015, X. Zhang Fordham Univ.

A Brief Analysis on Architecture and Reliability of Cloud Based Data Storage

marlabs driving digital agility WHITEPAPER Big Data and Hadoop

Designing a Cloud Storage System

MOC SERVER VIRTUALIZATION WITH WINDOWS SERVER HYPER-V AND SYSTEM CENTER

Deploying Hadoop with Manager

Perforce Disaster Recovery at Google. ! Google's mission is to organize the world's information and make it universally accessible and useful.

NET ACCESS VOICE PRIVATE CLOUD

HADOOP MOCK TEST HADOOP MOCK TEST

Apache HBase. Crazy dances on the elephant back

Transcription:

마스터 제목 스타일 편집 마스터 부제목 Availability 스타일 편집 and Data durability in HDFS 3 Jun 2011 nfracatals, 고등기술 연구소 / 이문수 moon@nfractals.com Company Profile and Business 1

Who we are? Since 2009 Consulting Solution for cloud computing Media delivery service 2

Topics We discuss about HDFS's Availability Data durability Not Performance Scalability 3

Overview of HDFS HDFS is a distributed storage system for reliably storing petabytes of data on clusters of commodity hardware Name node Managing metadata. Issues replication Secondary name node (Aka. Checkpoint node), Backup node Checkpoint, Backup metadata of Name node Data node Storing data Reliability backed by multiple replicas 4

Challenge Is reliable? Our case. Cloud storage for CDN Service 1PB data 99.999% Availability (less than 5 min / year). Files are not reproducible. The only copy. 5

Data durability Data durability = (1-P_blockloss) * (1-P_metadataloss) * (1-P_humanerror) * (1-P_sortwareerror) 6

P_metadataloss P_metadataloss is reduced by Multiple local disk RAID Remote NFS Separate secondary node or backup node Backup multiple copies of different ages 7

P_humanerror P_humanerror is reduced by Enable Trash facility Permissions 8

P_softwareerror P_softwareerror is reduced by Common HDFS configuration Move instead of programmatic delete 9

P_blockloss r = number of replicas f = number of datanode that fail concurrently n = number of datanode b = number of blocks p = probability of failure of a single machine P_single_failure = E[downtime] / (E[downtime] + E[uptime]) = MTTR / (MTTR+MTBF) MTTR = ( b / n ) / (3*n) P_single_failure = b/3n 2 / (MTBF + b/3n 2 ) P_failure = C(n,f) * p_single_failure f * (1-p_single_failure) (n-f) P_single_blockloss = ( C(n,f)*C(f,r) ) / ( C(n,f)*C(n,r) ) = C(f,r) / C(n,r) P_no_blockloss = (1 - P_single_blockloss) b n P_blockloss = 1 - P_failure * P_no_blockloss f=0 10

P_blockloss To reduce P_blockloss Minimize number of blocks Enable dfs.datanode.failed.volumes.tolerated Local redundancy for datanode OS volume 11

Availability Availability = (1-P_failure_NN) * (1-P_blockloss) * (1-P_networkerror) * (1-P_maintenance) where P_failure_NN = MTTR / (MTBF+MTTR) 12

Improve Availability To improve Availability Active - Standby HA Cluster Fault tolerant Name node Standby HDFS Cluster Virtualization Layer for Multiple HDFS Cluster 13

Active - Standby HA Cluster DRBD+LinuxHA (http://www.cloudera.com/blog/2009/07/hadoop-ha-configuration/) Reduce MTTR Reduced MTTR is still very large from few minutes to several hours 14

Fault tolerant Name node Remove MTTR, Reduce P_maintenance NN availability (https://issues.apache.org/jira/browse/hdfs-1064) Avatar node (https://issues.apache.org/jira/browse/hdfs-976) - Not yet available EMC Greenplum Google GFS 15

Standby HDFS Cluster Standby cluster keep synchronized with Master cluster Reduce MTTR, P_maintenance, P_blockloss, P_networkerror Google app engine (GFS) 5 hours down, 2, July, 2009 http://groups.google.com/group/google-appengine/msg/ba95ded980c8c179?pli=1 16

Virtualization layer for Multiple HDFS Virtualization layer provides Different version HDFS support Virtual namespace on the top of multiple HDFS cluster Redundancy over HDFS clusters Remove MTTR, P_maintenance Reduce P_blockloss, P_networkerror 17

Summary To get greater availability, data durability Keep number of blocks as small as possible Multiple age, location of metadata backup Fast recovery is very important for both availability and durability More replica for important data 18

For Your Market Leading Questions nfractals www.nfractals.com Moonsoo Lee(moon@nfractals.com) 19