Haryadi Gunawi and Andrew Chien

Size: px
Start display at page:

Download "Haryadi Gunawi and Andrew Chien"

Transcription

1 Haryadi Gunawi and Andrew Chien in collaboration with Gokul Soundararajan and Deepak Kenchammana (NetApp) Rob Ross and Dries Kimpe (Argonne National Labs)

2 2 q Complete fail-stop q Fail partial Rich literature q Corruption q Performance degradation ( limpware )?

3 3 Limping NIC! (1,000,000x) q 1Gb NIC card on a machine that suddenly starts transmitting at 1 kbps, q this one slow machine caused a chain reaction making a 100 node cluster was crawling at a snail's pace Facebook Engineers Cascading impact!

4 4 q q q q q q Disks 4 servers having high wait times on I/O for, up to 103 seconds. This was left uncorrected for 50 Argonne Causes: Weak disk head, bad packaging, missing screws, broken/old fans, too many disks/ box, firmware bugs, bad sector remapping, SSDs Samsung firmware bug (reduce bandwidth by 4x) Network cards and switches On Intrepid, a bad batch of optical transceivers with an extremely high error rate cause an effective throughput of 1-2 Argonne Causes: Broken adapter, error correcting, driver bugs, power fluctuation, Memory Runs only at 25% of normal speed HBase operators Processors 26% variation Aging transistors, overheat, self throttling, Many others: Yes we've seen that in production More anecdotes in our paper [SoCC 13]

5 5

6 6 q Introduction q Impact of limpware to scale-out cloud systems? [HotCloud 13, SoCC 13] q Progress Summary What bugs live in the cloud? [SoCC 14] Detecting performance bugs [HotCloud 15] The Tail at Store [In Submission] Other ongoing work

7 7 q Anecdotes The performance of a 100 node cluster was crawling at a snail's pace Facebook q But, why?

8 8 q Goals: Measure system-level impacts Find design flaws q Run distributed systems/protocols E.g., 3-node write in HDFS q Measure slowdowns under: No failure, crash, a limping NIC Execution slowdown 1000x slower 100x slower 0.1 Mbps NIC 1Mbps NIC workload 10x slower 10 Mbps NIC 1

9 9 HDFS Hadoop ZooKeeper Cassandra HBase

10 Fail-stop tolerant, but not limpware tolerant (no failover recovery) 10

11 q Run Hadoop with 6+ hours of Facebook workload 30-node cluster 30-node cluster (w/ 1 slow 0.1 Mbps) 1 job/hour Also happens in HDFS and ZooKeeper Cluster collapse after ~4 hours 11

12 12 q Single point of performance failure q Coarse-grained timeouts q Bounded thread/queue pool à resource exhaustion q Unbounded thread/queue pool à OOM q No throttling or back-pressure q Limp-oblivious background jobs q Unexploited parallelism of small transactional I/Os q Long lock/resource contention q

13 13 q Introduction q Impact of limpware [SoCC 13] q Progress Summary

14 14 q Study/Analysis Limplock/limpware [HotCloud 13, SoCC 13] What bugs live in the cloud? [SoCC 14]

15 15 q Study/Analysis Limplock/limpware [HotCloud 13, SoCC 13] What bugs live in the cloud? [SoCC 14] - Study of bugs in scale-out distributed systems - New: scalability bugs, single-point-of-failure bugs,

16 16 q Study/Analysis Limplock/limpware [HotCloud 13, SoCC 13] What bugs live in the cloud? [SoCC 14] The Tail at Store [In Submission] - Goal: Anecdotes to real statistics - Collaboration with Gokul Soundararajan and Deepak Kenchammana - Study of over 450,000 disks, 4000 SSDs, and 240 EBS drives - Ask: How many slow drives? How often? Transient? RAID RAID

17 17 q Study/Analysis Limplock/limpware [HotCloud 13, SoCC 13] What bugs live in the cloud? [SoCC 14] The Tail at Store [In Submission] - Limping disks and SSDs are real! - 2-digit slowdowns had occurred in 0.01% of disk and SSD hours - 4- and 3-digit slowdowns in 124 and 2461 disk hours, and 3-digit SSD slowdowns in 10 SSD hours

18 18 q Study/Analysis q Towards Limpware-Tolerant Systems Detecting limpware-intolerant designs in distributed systems [HotCloud 15] Tail-tolerant storage [In Progress] - In flash controller, operating system, and distributed storage - + Coordination with MapReduce Speculative Execution - (A cross-cutting approach) TT Flash Ctrl TT OS/RAID MapReduce Spec. Ex. TT Distr. FS

19 19 XPS à Exploit Scale Limpware à Underexploit Scale ucare.cs.uchicago.edu ceres.uchicago.edu

Haryadi S. Gunawi 1, Riza O. Suminto 1, Russell Sears 2, Casey Golliher 2, Swaminathan Sundararaman 3, Xing Lin 4, Tim Emami 4, Weiguang Sheng 5,

Haryadi S. Gunawi 1, Riza O. Suminto 1, Russell Sears 2, Casey Golliher 2, Swaminathan Sundararaman 3, Xing Lin 4, Tim Emami 4, Weiguang Sheng 5, Haryadi S. Gunawi 1, Riza O. Suminto 1, Russell Sears 2, Casey Golliher 2, Swaminathan Sundararaman 3, Xing Lin 4, Tim Emami 4, Weiguang Sheng 5, Nematollah Bidokhti 5, Caitie McCaffrey 6, Gary Grider

More information

Limplock: Understanding the Impact of Limpware on Scale-Out Cloud Systems

Limplock: Understanding the Impact of Limpware on Scale-Out Cloud Systems Limplock: Understanding the Impact of Limpware on Scale-Out Cloud Systems Thanh Do, Mingzhe Hao, Tanakorn Leesatapornwongsa, Tiratat Patana-anake, and Haryadi S Gunawi University of Chicago University

More information

MixApart: Decoupled Analytics for Shared Storage Systems. Madalin Mihailescu, Gokul Soundararajan, Cristiana Amza University of Toronto and NetApp

MixApart: Decoupled Analytics for Shared Storage Systems. Madalin Mihailescu, Gokul Soundararajan, Cristiana Amza University of Toronto and NetApp MixApart: Decoupled Analytics for Shared Storage Systems Madalin Mihailescu, Gokul Soundararajan, Cristiana Amza University of Toronto and NetApp Hadoop Pig, Hive Hadoop + Enterprise storage?! Shared storage

More information

Manylogs Improving CMR/SMR Disk Bandwidth & Latency. Tiratat Patana-anake, Vincentius Martin, Nora Sandler, Cheng Wu, and Haryadi S.

Manylogs Improving CMR/SMR Disk Bandwidth & Latency. Tiratat Patana-anake, Vincentius Martin, Nora Sandler, Cheng Wu, and Haryadi S. Manylogs Improving CMR/SMR Disk Bandwidth & Latency Tiratat Patana-anake, Vincentius Martin, Nora Sandler, Cheng Wu, and Haryadi S. Gunawi Manylogs @ M SST 1 6 2 Got 100% of the read bandwidth User 1 Big

More information

Lecture 11 Hadoop & Spark

Lecture 11 Hadoop & Spark Lecture 11 Hadoop & Spark Dr. Wilson Rivera ICOM 6025: High Performance Computing Electrical and Computer Engineering Department University of Puerto Rico Outline Distributed File Systems Hadoop Ecosystem

More information

Programming Models MapReduce

Programming Models MapReduce Programming Models MapReduce Majd Sakr, Garth Gibson, Greg Ganger, Raja Sambasivan 15-719/18-847b Advanced Cloud Computing Fall 2013 Sep 23, 2013 1 MapReduce In a Nutshell MapReduce incorporates two phases

More information

CSE-E5430 Scalable Cloud Computing Lecture 9

CSE-E5430 Scalable Cloud Computing Lecture 9 CSE-E5430 Scalable Cloud Computing Lecture 9 Keijo Heljanko Department of Computer Science School of Science Aalto University keijo.heljanko@aalto.fi 15.11-2015 1/24 BigTable Described in the paper: Fay

More information

YCSB++ Benchmarking Tool Performance Debugging Advanced Features of Scalable Table Stores

YCSB++ Benchmarking Tool Performance Debugging Advanced Features of Scalable Table Stores YCSB++ Benchmarking Tool Performance Debugging Advanced Features of Scalable Table Stores Swapnil Patil Milo Polte, Wittawat Tantisiriroj, Kai Ren, Lin Xiao, Julio Lopez, Garth Gibson, Adam Fuchs *, Billie

More information

PACM: A Prediction-based Auto-adaptive Compression Model for HDFS. Ruijian Wang, Chao Wang, Li Zha

PACM: A Prediction-based Auto-adaptive Compression Model for HDFS. Ruijian Wang, Chao Wang, Li Zha PACM: A Prediction-based Auto-adaptive Compression Model for HDFS Ruijian Wang, Chao Wang, Li Zha Hadoop Distributed File System Store a variety of data http://popista.com/distributed-filesystem/distributed-file-system:/125620

More information

HIGH PERFORMANCE STORAGE SOLUTION PRESENTATION All rights reserved RAIDIX

HIGH PERFORMANCE STORAGE SOLUTION PRESENTATION All rights reserved RAIDIX HIGH PERFORMANCE STORAGE SOLUTION PRESENTATION 2017 All rights reserved RAIDIX ABOUT COMPANY RAIDIX is a innovative solution provider and developer of high-performance storage systems. Patented erasure

More information

EsgynDB Enterprise 2.0 Platform Reference Architecture

EsgynDB Enterprise 2.0 Platform Reference Architecture EsgynDB Enterprise 2.0 Platform Reference Architecture This document outlines a Platform Reference Architecture for EsgynDB Enterprise, built on Apache Trafodion (Incubating) implementation with licensed

More information

YCSB++ benchmarking tool Performance debugging advanced features of scalable table stores

YCSB++ benchmarking tool Performance debugging advanced features of scalable table stores YCSB++ benchmarking tool Performance debugging advanced features of scalable table stores Swapnil Patil M. Polte, W. Tantisiriroj, K. Ren, L.Xiao, J. Lopez, G.Gibson, A. Fuchs *, B. Rinaldi * Carnegie

More information

SparkStreaming. Large scale near- realtime stream processing. Tathagata Das (TD) UC Berkeley UC BERKELEY

SparkStreaming. Large scale near- realtime stream processing. Tathagata Das (TD) UC Berkeley UC BERKELEY SparkStreaming Large scale near- realtime stream processing Tathagata Das (TD) UC Berkeley UC BERKELEY Motivation Many important applications must process large data streams at second- scale latencies

More information

Analytics in the cloud

Analytics in the cloud Analytics in the cloud Dow we really need to reinvent the storage stack? R. Ananthanarayanan, Karan Gupta, Prashant Pandey, Himabindu Pucha, Prasenjit Sarkar, Mansi Shah, Renu Tewari Image courtesy NASA

More information

Improving Hadoop MapReduce Performance on Supercomputers with JVM Reuse

Improving Hadoop MapReduce Performance on Supercomputers with JVM Reuse Thanh-Chung Dao 1 Improving Hadoop MapReduce Performance on Supercomputers with JVM Reuse Thanh-Chung Dao and Shigeru Chiba The University of Tokyo Thanh-Chung Dao 2 Supercomputers Expensive clusters Multi-core

More information

Big Data Hadoop Stack

Big Data Hadoop Stack Big Data Hadoop Stack Lecture #1 Hadoop Beginnings What is Hadoop? Apache Hadoop is an open source software framework for storage and large scale processing of data-sets on clusters of commodity hardware

More information

Disks and RAID. CS 4410 Operating Systems. [R. Agarwal, L. Alvisi, A. Bracy, E. Sirer, R. Van Renesse]

Disks and RAID. CS 4410 Operating Systems. [R. Agarwal, L. Alvisi, A. Bracy, E. Sirer, R. Van Renesse] Disks and RAID CS 4410 Operating Systems [R. Agarwal, L. Alvisi, A. Bracy, E. Sirer, R. Van Renesse] Storage Devices Magnetic disks Storage that rarely becomes corrupted Large capacity at low cost Block

More information

Accelerating Big Data: Using SanDisk SSDs for Apache HBase Workloads

Accelerating Big Data: Using SanDisk SSDs for Apache HBase Workloads WHITE PAPER Accelerating Big Data: Using SanDisk SSDs for Apache HBase Workloads December 2014 Western Digital Technologies, Inc. 951 SanDisk Drive, Milpitas, CA 95035 www.sandisk.com Table of Contents

More information

Hadoop. Introduction / Overview

Hadoop. Introduction / Overview Hadoop Introduction / Overview Preface We will use these PowerPoint slides to guide us through our topic. Expect 15 minute segments of lecture Expect 1-4 hour lab segments Expect minimal pretty pictures

More information

Large- Scale Sor,ng: Breaking World Records. Mike Conley CSE 124 Guest Lecture 12 March 2015

Large- Scale Sor,ng: Breaking World Records. Mike Conley CSE 124 Guest Lecture 12 March 2015 Large- Scale Sor,ng: Breaking World Records Mike Conley CSE 124 Guest Lecture 12 March 2015 Sor,ng Given an array of items, put them in order 5 2 8 0 2 5 4 9 0 1 0 0 0 0 0 0 1 2 2 4 5 5 8 9 Many algorithms

More information

How Apache Hadoop Complements Existing BI Systems. Dr. Amr Awadallah Founder, CTO Cloudera,

How Apache Hadoop Complements Existing BI Systems. Dr. Amr Awadallah Founder, CTO Cloudera, How Apache Hadoop Complements Existing BI Systems Dr. Amr Awadallah Founder, CTO Cloudera, Inc. Twitter: @awadallah, @cloudera 2 The Problems with Current Data Systems BI Reports + Interactive Apps RDBMS

More information

DiskReduce: Making Room for More Data on DISCs. Wittawat Tantisiriroj

DiskReduce: Making Room for More Data on DISCs. Wittawat Tantisiriroj DiskReduce: Making Room for More Data on DISCs Wittawat Tantisiriroj Lin Xiao, Bin Fan, and Garth Gibson PARALLEL DATA LABORATORY Carnegie Mellon University GFS/HDFS Triplication GFS & HDFS triplicate

More information

ONOS: TOWARDS AN OPEN, DISTRIBUTED SDN OS. Chun Yuan Cheng

ONOS: TOWARDS AN OPEN, DISTRIBUTED SDN OS. Chun Yuan Cheng ONOS: TOWARDS AN OPEN, DISTRIBUTED SDN OS Chun Yuan Cheng OUTLINE - Introduction - Two prototypes - Conclusion INTRODUCTION - An open, vendor neutral, control-data plane interface such as OpenFlow allows

More information

Next-Generation Cloud Platform

Next-Generation Cloud Platform Next-Generation Cloud Platform Jangwoo Kim Jun 24, 2013 E-mail: jangwoo@postech.ac.kr High Performance Computing Lab Department of Computer Science & Engineering Pohang University of Science and Technology

More information

Cloud Computing and Hadoop Distributed File System. UCSB CS170, Spring 2018

Cloud Computing and Hadoop Distributed File System. UCSB CS170, Spring 2018 Cloud Computing and Hadoop Distributed File System UCSB CS70, Spring 08 Cluster Computing Motivations Large-scale data processing on clusters Scan 000 TB on node @ 00 MB/s = days Scan on 000-node cluster

More information

Algorithms and Data Structures for Efficient Free Space Reclamation in WAFL

Algorithms and Data Structures for Efficient Free Space Reclamation in WAFL Algorithms and Data Structures for Efficient Free Space Reclamation in WAFL Ram Kesavan Technical Director, WAFL NetApp, Inc. SDC 2017 1 Outline Garbage collection in WAFL Usenix FAST 2017 ACM Transactions

More information

ibench: Quantifying Interference in Datacenter Applications

ibench: Quantifying Interference in Datacenter Applications ibench: Quantifying Interference in Datacenter Applications Christina Delimitrou and Christos Kozyrakis Stanford University IISWC September 23 th 2013 Executive Summary Problem: Increasing utilization

More information

Nutanix Tech Note. Virtualizing Microsoft Applications on Web-Scale Infrastructure

Nutanix Tech Note. Virtualizing Microsoft Applications on Web-Scale Infrastructure Nutanix Tech Note Virtualizing Microsoft Applications on Web-Scale Infrastructure The increase in virtualization of critical applications has brought significant attention to compute and storage infrastructure.

More information

MixApart: Decoupled Analytics for Shared Storage Systems

MixApart: Decoupled Analytics for Shared Storage Systems MixApart: Decoupled Analytics for Shared Storage Systems Madalin Mihailescu, Gokul Soundararajan, Cristiana Amza University of Toronto, NetApp Abstract Data analytics and enterprise applications have very

More information

Topics. Big Data Analytics What is and Why Hadoop? Comparison to other technologies Hadoop architecture Hadoop ecosystem Hadoop usage examples

Topics. Big Data Analytics What is and Why Hadoop? Comparison to other technologies Hadoop architecture Hadoop ecosystem Hadoop usage examples Hadoop Introduction 1 Topics Big Data Analytics What is and Why Hadoop? Comparison to other technologies Hadoop architecture Hadoop ecosystem Hadoop usage examples 2 Big Data Analytics What is Big Data?

More information

Distributed Systems 16. Distributed File Systems II

Distributed Systems 16. Distributed File Systems II Distributed Systems 16. Distributed File Systems II Paul Krzyzanowski pxk@cs.rutgers.edu 1 Review NFS RPC-based access AFS Long-term caching CODA Read/write replication & disconnected operation DFS AFS

More information

Konstantin Shvachko, Hairong Kuang, Sanjay Radia, Robert Chansler Yahoo! Sunnyvale, California USA {Shv, Hairong, SRadia,

Konstantin Shvachko, Hairong Kuang, Sanjay Radia, Robert Chansler Yahoo! Sunnyvale, California USA {Shv, Hairong, SRadia, Konstantin Shvachko, Hairong Kuang, Sanjay Radia, Robert Chansler Yahoo! Sunnyvale, California USA {Shv, Hairong, SRadia, Chansler}@Yahoo-Inc.com Presenter: Alex Hu } Introduction } Architecture } File

More information

Hadoop File System S L I D E S M O D I F I E D F R O M P R E S E N T A T I O N B Y B. R A M A M U R T H Y 11/15/2017

Hadoop File System S L I D E S M O D I F I E D F R O M P R E S E N T A T I O N B Y B. R A M A M U R T H Y 11/15/2017 Hadoop File System 1 S L I D E S M O D I F I E D F R O M P R E S E N T A T I O N B Y B. R A M A M U R T H Y Moving Computation is Cheaper than Moving Data Motivation: Big Data! What is BigData? - Google

More information

Hadoop An Overview. - Socrates CCDH

Hadoop An Overview. - Socrates CCDH Hadoop An Overview - Socrates CCDH What is Big Data? Volume Not Gigabyte. Terabyte, Petabyte, Exabyte, Zettabyte - Due to handheld gadgets,and HD format images and videos - In total data, 90% of them collected

More information

Parallel Programming Principle and Practice. Lecture 10 Big Data Processing with MapReduce

Parallel Programming Principle and Practice. Lecture 10 Big Data Processing with MapReduce Parallel Programming Principle and Practice Lecture 10 Big Data Processing with MapReduce Outline MapReduce Programming Model MapReduce Examples Hadoop 2 Incredible Things That Happen Every Minute On The

More information

Storage Technologies - 3

Storage Technologies - 3 Storage Technologies - 3 COMP 25212 - Lecture 10 Antoniu Pop antoniu.pop@manchester.ac.uk 1 March 2019 Antoniu Pop Storage Technologies - 3 1 / 20 Learning Objectives - Storage 3 Understand characteristics

More information

Data Management. Parallel Filesystems. Dr David Henty HPC Training and Support

Data Management. Parallel Filesystems. Dr David Henty HPC Training and Support Data Management Dr David Henty HPC Training and Support d.henty@epcc.ed.ac.uk +44 131 650 5960 Overview Lecture will cover Why is IO difficult Why is parallel IO even worse Lustre GPFS Performance on ARCHER

More information

April Final Quiz COSC MapReduce Programming a) Explain briefly the main ideas and components of the MapReduce programming model.

April Final Quiz COSC MapReduce Programming a) Explain briefly the main ideas and components of the MapReduce programming model. 1. MapReduce Programming a) Explain briefly the main ideas and components of the MapReduce programming model. MapReduce is a framework for processing big data which processes data in two phases, a Map

More information

Cloud Computing CS

Cloud Computing CS Cloud Computing CS 15-319 Programming Models- Part III Lecture 6, Feb 1, 2012 Majd F. Sakr and Mohammad Hammoud 1 Today Last session Programming Models- Part II Today s session Programming Models Part

More information

Bandura High-speed disk duplicator. User s Manual v1.4

Bandura High-speed disk duplicator. User s Manual v1.4 Bandura High-speed disk duplicator User s Manual v1.4 Thank you for purchasing an Atola Technology product The Atola Bandura is a stand-alone high-speed 1-to-1 disk drive duplicator built for professional

More information

HDFS: Hadoop Distributed File System. CIS 612 Sunnie Chung

HDFS: Hadoop Distributed File System. CIS 612 Sunnie Chung HDFS: Hadoop Distributed File System CIS 612 Sunnie Chung What is Big Data?? Bulk Amount Unstructured Introduction Lots of Applications which need to handle huge amount of data (in terms of 500+ TB per

More information

Massive Online Analysis - Storm,Spark

Massive Online Analysis - Storm,Spark Massive Online Analysis - Storm,Spark presentation by R. Kishore Kumar Research Scholar Department of Computer Science & Engineering Indian Institute of Technology, Kharagpur Kharagpur-721302, India (R

More information

2013 AWS Worldwide Public Sector Summit Washington, D.C.

2013 AWS Worldwide Public Sector Summit Washington, D.C. 2013 AWS Worldwide Public Sector Summit Washington, D.C. EMR for Fun and for Profit Ben Butler Sr. Manager, Big Data butlerb@amazon.com @bensbutler Overview 1. What is big data? 2. What is AWS Elastic

More information

Shiqin Yan, Huaicheng Li, Mingzhe Hao, Michael Hao Tong, Swaminathan Sundararaman *, Andrew Chien, and Haryadi S. Gunawi

Shiqin Yan, Huaicheng Li, Mingzhe Hao, Michael Hao Tong, Swaminathan Sundararaman *, Andrew Chien, and Haryadi S. Gunawi Shiqin Yan, Huaicheng Li, Mingzhe Hao, Michael Hao Tong, Swaminathan Sundararaman *, Andrew Chien, and Haryadi S. Gunawi * 2 if your read is stuck behind an erase you may have wait 10s of milliseconds.

More information

2/26/2017. For instance, consider running Word Count across 20 splits

2/26/2017. For instance, consider running Word Count across 20 splits Based on the slides of prof. Pietro Michiardi Hadoop Internals https://github.com/michiard/disc-cloud-course/raw/master/hadoop/hadoop.pdf Job: execution of a MapReduce application across a data set Task:

More information

IBM InfoSphere Streams v4.0 Performance Best Practices

IBM InfoSphere Streams v4.0 Performance Best Practices Henry May IBM InfoSphere Streams v4.0 Performance Best Practices Abstract Streams v4.0 introduces powerful high availability features. Leveraging these requires careful consideration of performance related

More information

Hadoop MapReduce Framework

Hadoop MapReduce Framework Hadoop MapReduce Framework Contents Hadoop MapReduce Framework Architecture Interaction Diagram of MapReduce Framework (Hadoop 1.0) Interaction Diagram of MapReduce Framework (Hadoop 2.0) Hadoop MapReduce

More information

PLATFORM AND SOFTWARE AS A SERVICE THE MAPREDUCE PROGRAMMING MODEL AND IMPLEMENTATIONS

PLATFORM AND SOFTWARE AS A SERVICE THE MAPREDUCE PROGRAMMING MODEL AND IMPLEMENTATIONS PLATFORM AND SOFTWARE AS A SERVICE THE MAPREDUCE PROGRAMMING MODEL AND IMPLEMENTATIONS By HAI JIN, SHADI IBRAHIM, LI QI, HAIJUN CAO, SONG WU and XUANHUA SHI Prepared by: Dr. Faramarz Safi Islamic Azad

More information

CS370 Operating Systems

CS370 Operating Systems CS370 Operating Systems Colorado State University Yashwant K Malaiya Spring 2018 Lecture 24 Mass Storage, HDFS/Hadoop Slides based on Text by Silberschatz, Galvin, Gagne Various sources 1 1 FAQ What 2

More information

Processing of big data with Apache Spark

Processing of big data with Apache Spark Processing of big data with Apache Spark JavaSkop 18 Aleksandar Donevski AGENDA What is Apache Spark? Spark vs Hadoop MapReduce Application Requirements Example Architecture Application Challenges 2 WHAT

More information

Page 1. Goals for Today" Background of Cloud Computing" Sources Driving Big Data" CS162 Operating Systems and Systems Programming Lecture 24

Page 1. Goals for Today Background of Cloud Computing Sources Driving Big Data CS162 Operating Systems and Systems Programming Lecture 24 Goals for Today" CS162 Operating Systems and Systems Programming Lecture 24 Capstone: Cloud Computing" Distributed systems Cloud Computing programming paradigms Cloud Computing OS December 2, 2013 Anthony

More information

Programming Distributed Systems

Programming Distributed Systems Annette Bieniusa Programming Distributed Systems Summer Term 2018 1/ 26 Programming Distributed Systems 09 Testing Distributed Systems Annette Bieniusa AG Softech FB Informatik TU Kaiserslautern Summer

More information

Outline. Spanner Mo/va/on. Tom Anderson

Outline. Spanner Mo/va/on. Tom Anderson Spanner Mo/va/on Tom Anderson Outline Last week: Chubby: coordina/on service BigTable: scalable storage of structured data GFS: large- scale storage for bulk data Today/Friday: Lessons from GFS/BigTable

More information

MapR Enterprise Hadoop

MapR Enterprise Hadoop 2014 MapR Technologies 2014 MapR Technologies 1 MapR Enterprise Hadoop Top Ranked Cloud Leaders 500+ Customers 2014 MapR Technologies 2 Key MapR Advantage Partners Business Services APPLICATIONS & OS ANALYTICS

More information

朱义普. Resolving High Performance Computing and Big Data Application Bottlenecks with Application-Defined Flash Acceleration. Director, North Asia, HPC

朱义普. Resolving High Performance Computing and Big Data Application Bottlenecks with Application-Defined Flash Acceleration. Director, North Asia, HPC October 28, 2013 Resolving High Performance Computing and Big Data Application Bottlenecks with Application-Defined Flash Acceleration 朱义普 Director, North Asia, HPC DDN Storage Vendor for HPC & Big Data

More information

Request-Oriented Durable Write Caching for Application Performance appeared in USENIX ATC '15. Jinkyu Jeong Sungkyunkwan University

Request-Oriented Durable Write Caching for Application Performance appeared in USENIX ATC '15. Jinkyu Jeong Sungkyunkwan University Request-Oriented Durable Write Caching for Application Performance appeared in USENIX ATC '15 Jinkyu Jeong Sungkyunkwan University Introduction Volatile DRAM cache is ineffective for write Writes are dominant

More information

CS3600 SYSTEMS AND NETWORKS

CS3600 SYSTEMS AND NETWORKS CS3600 SYSTEMS AND NETWORKS NORTHEASTERN UNIVERSITY Lecture 11: File System Implementation Prof. Alan Mislove (amislove@ccs.neu.edu) File-System Structure File structure Logical storage unit Collection

More information

Data Intensive Scalable Computing

Data Intensive Scalable Computing Data Intensive Scalable Computing Randal E. Bryant Carnegie Mellon University http://www.cs.cmu.edu/~bryant Examples of Big Data Sources Wal-Mart 267 million items/day, sold at 6,000 stores HP built them

More information

Best Practices for Deploying Hadoop Workloads on HCI Powered by vsan

Best Practices for Deploying Hadoop Workloads on HCI Powered by vsan Best Practices for Deploying Hadoop Workloads on HCI Powered by vsan Chen Wei, ware, Inc. Paudie ORiordan, ware, Inc. #vmworld HCI2038BU #HCI2038BU Disclaimer This presentation may contain product features

More information

Practical MySQL Performance Optimization. Peter Zaitsev, CEO, Percona July 02, 2015 Percona Technical Webinars

Practical MySQL Performance Optimization. Peter Zaitsev, CEO, Percona July 02, 2015 Percona Technical Webinars Practical MySQL Performance Optimization Peter Zaitsev, CEO, Percona July 02, 2015 Percona Technical Webinars In This Presentation We ll Look at how to approach Performance Optimization Discuss Practical

More information

Big Data and Object Storage

Big Data and Object Storage Big Data and Object Storage or where to store the cold and small data? Sven Bauernfeind Computacenter AG & Co. ohg, Consultancy Germany 28.02.2018 Munich Volume, Variety & Velocity + Analytics Velocity

More information

<Insert Picture Here> Filesystem Features and Performance

<Insert Picture Here> Filesystem Features and Performance Filesystem Features and Performance Chris Mason Filesystems XFS Well established and stable Highly scalable under many workloads Can be slower in metadata intensive workloads Often

More information

SSD Architecture for Consistent Enterprise Performance

SSD Architecture for Consistent Enterprise Performance SSD Architecture for Consistent Enterprise Performance Gary Tressler and Tom Griffin IBM Corporation August 21, 212 1 SSD Architecture for Consistent Enterprise Performance - Overview Background: Client

More information

Building Self-Healing Mass Storage Arrays. for Large Cluster Systems

Building Self-Healing Mass Storage Arrays. for Large Cluster Systems Building Self-Healing Mass Storage Arrays for Large Cluster Systems NSC08, Linköping, 14. October 2008 Toine Beckers tbeckers@datadirectnet.com Agenda Company Overview Balanced I/O Systems MTBF and Availability

More information

Fail-Slow at Scale: Evidence of Hardware Performance Faults in Large Production Systems

Fail-Slow at Scale: Evidence of Hardware Performance Faults in Large Production Systems Fail-Slow at Scale: Evidence of Hardware Performance Faults in Large Production Systems Haryadi S. Gunawi 1, Riza O. Suminto 1, Russell Sears 2, Casey Golliher 2, Swaminathan Sundararaman 3, Xing Lin 4,

More information

Service Oriented Performance Analysis

Service Oriented Performance Analysis Service Oriented Performance Analysis Da Qi Ren and Masood Mortazavi US R&D Center Santa Clara, CA, USA www.huawei.com Performance Model for Service in Data Center and Cloud 1. Service Oriented (end to

More information

Storage for HPC, HPDA and Machine Learning (ML)

Storage for HPC, HPDA and Machine Learning (ML) for HPC, HPDA and Machine Learning (ML) Frank Kraemer, IBM Systems Architect mailto:kraemerf@de.ibm.com IBM Data Management for Autonomous Driving (AD) significantly increase development efficiency by

More information

Benefits of Multi-Node Scale-out Clusters running NetApp Clustered Data ONTAP. Silverton Consulting, Inc. StorInt Briefing

Benefits of Multi-Node Scale-out Clusters running NetApp Clustered Data ONTAP. Silverton Consulting, Inc. StorInt Briefing Benefits of Multi-Node Scale-out Clusters running NetApp Clustered Data ONTAP Silverton Consulting, Inc. StorInt Briefing BENEFITS OF MULTI- NODE SCALE- OUT CLUSTERS RUNNING NETAPP CDOT PAGE 2 OF 7 Introduction

More information

NetApp: Solving I/O Challenges. Jeff Baxter February 2013

NetApp: Solving I/O Challenges. Jeff Baxter February 2013 NetApp: Solving I/O Challenges Jeff Baxter February 2013 1 High Performance Computing Challenges Computing Centers Challenge of New Science Performance Efficiency directly impacts achievable science Power

More information

Accelerate Big Data Insights

Accelerate Big Data Insights Accelerate Big Data Insights Executive Summary An abundance of information isn t always helpful when time is of the essence. In the world of big data, the ability to accelerate time-to-insight can not

More information

Outline: ONTAP 9 Cluster Administration and Data Protection Bundle (CDOTDP9)

Outline: ONTAP 9 Cluster Administration and Data Protection Bundle (CDOTDP9) Outline: ONTAP 9 Cluster Administration and Data Protection Bundle (CDOTDP9) Cluster Module 1: ONTAP Overview Data Fabric ONTAP software Fabric layers The cluster Nodes High-availability pairs Networks

More information

Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li

Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li Raghav Sethi Michael Kaminsky David G. Andersen Michael J. Freedman Goal: fast and cost-effective key-value store Target: cluster-level storage for

More information

Nowcasting. D B M G Data Base and Data Mining Group of Politecnico di Torino. Big Data: Hype or Hallelujah? Big data hype?

Nowcasting. D B M G Data Base and Data Mining Group of Politecnico di Torino. Big Data: Hype or Hallelujah? Big data hype? Big data hype? Big Data: Hype or Hallelujah? Data Base and Data Mining Group of 2 Google Flu trends On the Internet February 2010 detected flu outbreak two weeks ahead of CDC data Nowcasting http://www.internetlivestats.com/

More information

Data Processing at the Speed of 100 Gbps using Apache Crail. Patrick Stuedi IBM Research

Data Processing at the Speed of 100 Gbps using Apache Crail. Patrick Stuedi IBM Research Data Processing at the Speed of 100 Gbps using Apache Crail Patrick Stuedi IBM Research The CRAIL Project: Overview Data Processing Framework (e.g., Spark, TensorFlow, λ Compute) Spark-IO Albis Pocket

More information

MapReduce: Simplified Data Processing on Large Clusters 유연일민철기

MapReduce: Simplified Data Processing on Large Clusters 유연일민철기 MapReduce: Simplified Data Processing on Large Clusters 유연일민철기 Introduction MapReduce is a programming model and an associated implementation for processing and generating large data set with parallel,

More information

Decentralized Distributed Storage System for Big Data

Decentralized Distributed Storage System for Big Data Decentralized Distributed Storage System for Big Presenter: Wei Xie -Intensive Scalable Computing Laboratory(DISCL) Computer Science Department Texas Tech University Outline Trends in Big and Cloud Storage

More information

Big Data com Hadoop. VIII Sessão - SQL Bahia. Impala, Hive e Spark. Diógenes Pires 03/03/2018

Big Data com Hadoop. VIII Sessão - SQL Bahia. Impala, Hive e Spark. Diógenes Pires 03/03/2018 Big Data com Hadoop Impala, Hive e Spark VIII Sessão - SQL Bahia 03/03/2018 Diógenes Pires Connect with PASS Sign up for a free membership today at: pass.org #sqlpass Internet Live http://www.internetlivestats.com/

More information

Hadoop 2.x Core: YARN, Tez, and Spark. Hortonworks Inc All Rights Reserved

Hadoop 2.x Core: YARN, Tez, and Spark. Hortonworks Inc All Rights Reserved Hadoop 2.x Core: YARN, Tez, and Spark YARN Hadoop Machine Types top-of-rack switches core switch client machines have client-side software used to access a cluster to process data master nodes run Hadoop

More information

Cassandra, MongoDB, and HBase. Cassandra, MongoDB, and HBase. I have chosen these three due to their recent

Cassandra, MongoDB, and HBase. Cassandra, MongoDB, and HBase. I have chosen these three due to their recent Tanton Jeppson CS 401R Lab 3 Cassandra, MongoDB, and HBase Introduction For my report I have chosen to take a deeper look at 3 NoSQL database systems: Cassandra, MongoDB, and HBase. I have chosen these

More information

Crossing the Chasm: Sneaking a parallel file system into Hadoop

Crossing the Chasm: Sneaking a parallel file system into Hadoop Crossing the Chasm: Sneaking a parallel file system into Hadoop Wittawat Tantisiriroj Swapnil Patil, Garth Gibson PARALLEL DATA LABORATORY Carnegie Mellon University In this work Compare and contrast large

More information

15-440/15-640: Homework 3 Due: November 8, :59pm

15-440/15-640: Homework 3 Due: November 8, :59pm Name: 15-440/15-640: Homework 3 Due: November 8, 2018 11:59pm Andrew ID: 1 GFS FTW (25 points) Part A (10 points) The Google File System (GFS) is an extremely popular filesystem used by Google for a lot

More information

Concepts Introduced in Chapter 6. Warehouse-Scale Computers. Programming Models for WSCs. Important Design Factors for WSCs

Concepts Introduced in Chapter 6. Warehouse-Scale Computers. Programming Models for WSCs. Important Design Factors for WSCs Concepts Introduced in Chapter 6 Warehouse-Scale Computers A cluster is a collection of desktop computers or servers connected together by a local area network to act as a single larger computer. introduction

More information

Distributed Computation Models

Distributed Computation Models Distributed Computation Models SWE 622, Spring 2017 Distributed Software Engineering Some slides ack: Jeff Dean HW4 Recap https://b.socrative.com/ Class: SWE622 2 Review Replicating state machines Case

More information

Frequently asked questions from the previous class survey

Frequently asked questions from the previous class survey CS 370: OPERATING SYSTEMS [MASS STORAGE] Shrideep Pallickara Computer Science Colorado State University L29.1 Frequently asked questions from the previous class survey How does NTFS compare with UFS? L29.2

More information

Data Clustering on the Parallel Hadoop MapReduce Model. Dimitrios Verraros

Data Clustering on the Parallel Hadoop MapReduce Model. Dimitrios Verraros Data Clustering on the Parallel Hadoop MapReduce Model Dimitrios Verraros Overview The purpose of this thesis is to implement and benchmark the performance of a parallel K- means clustering algorithm on

More information

When Hadoop-like Distributed Storage Meets NAND Flash: Challenge and Opportunity

When Hadoop-like Distributed Storage Meets NAND Flash: Challenge and Opportunity When Hadoop-like Distributed Storage Meets NAND Flash: Challenge and Opportunity Jupyung Lee Intelligent Computing Lab Future IT Research Center Samsung Advanced Institute of Technology November 9, 2011

More information

ΕΠΛ 602:Foundations of Internet Technologies. Cloud Computing

ΕΠΛ 602:Foundations of Internet Technologies. Cloud Computing ΕΠΛ 602:Foundations of Internet Technologies Cloud Computing 1 Outline Bigtable(data component of cloud) Web search basedonch13of thewebdatabook 2 What is Cloud Computing? ACloudis an infrastructure, transparent

More information

Shark: SQL and Rich Analytics at Scale. Michael Xueyuan Han Ronny Hajoon Ko

Shark: SQL and Rich Analytics at Scale. Michael Xueyuan Han Ronny Hajoon Ko Shark: SQL and Rich Analytics at Scale Michael Xueyuan Han Ronny Hajoon Ko What Are The Problems? Data volumes are expanding dramatically Why Is It Hard? Needs to scale out Managing hundreds of machines

More information

Software Based Fault Injection Framework For Storage Systems Vinod Eswaraprasad Smitha Jayaram Wipro Technologies

Software Based Fault Injection Framework For Storage Systems Vinod Eswaraprasad Smitha Jayaram Wipro Technologies Software Based Fault Injection Framework For Storage Systems Vinod Eswaraprasad Smitha Jayaram Wipro Technologies The agenda Reliability in Storage systems Types of errors/faults in distributed storage

More information

PracticeDump. Free Practice Dumps - Unlimited Free Access of practice exam

PracticeDump.  Free Practice Dumps - Unlimited Free Access of practice exam PracticeDump http://www.practicedump.com Free Practice Dumps - Unlimited Free Access of practice exam Exam : 74-409 Title : Server Virtualization with Windows Server Hyper-V and System Center Vendor :

More information

ONTAP 9 Cluster Administration. Course outline. Authorised Vendor e-learning. Guaranteed To Run. DR Digital Learning. Module 1: ONTAP Overview

ONTAP 9 Cluster Administration. Course outline. Authorised Vendor e-learning. Guaranteed To Run. DR Digital Learning. Module 1: ONTAP Overview ONTAP 9 Cluster Administration Course Code: Duration: 3 Days Product Page: https://digitalrevolver.com/product/ontap-9-cluster-administration-2/ This 3-day, instructor led course uses lecture and hands-on

More information

Facilitating Magnetic Recording Technology Scaling for Data Center Hard Disk Drives through Filesystem-level Transparent Local Erasure Coding

Facilitating Magnetic Recording Technology Scaling for Data Center Hard Disk Drives through Filesystem-level Transparent Local Erasure Coding Facilitating Magnetic Recording Technology Scaling for Data Center Hard Disk Drives through Filesystem-level Transparent Local Erasure Coding Yin Li, Hao Wang, Xuebin Zhang, Ning Zheng, Shafa Dahandeh,

More information

Predictable Time-Sharing for DryadLINQ Cluster. Sang-Min Park and Marty Humphrey Dept. of Computer Science University of Virginia

Predictable Time-Sharing for DryadLINQ Cluster. Sang-Min Park and Marty Humphrey Dept. of Computer Science University of Virginia Predictable Time-Sharing for DryadLINQ Cluster Sang-Min Park and Marty Humphrey Dept. of Computer Science University of Virginia 1 DryadLINQ What is DryadLINQ? LINQ: Data processing language and run-time

More information

The Hadoop Ecosystem. EECS 4415 Big Data Systems. Tilemachos Pechlivanoglou

The Hadoop Ecosystem. EECS 4415 Big Data Systems. Tilemachos Pechlivanoglou The Hadoop Ecosystem EECS 4415 Big Data Systems Tilemachos Pechlivanoglou tipech@eecs.yorku.ca A lot of tools designed to work with Hadoop 2 HDFS, MapReduce Hadoop Distributed File System Core Hadoop component

More information

A+ Guide to Hardware: Managing, Maintaining, and Troubleshooting, 5e. Chapter 6 Supporting Hard Drives

A+ Guide to Hardware: Managing, Maintaining, and Troubleshooting, 5e. Chapter 6 Supporting Hard Drives A+ Guide to Hardware: Managing, Maintaining, and Troubleshooting, 5e Chapter 6 Supporting Hard Drives Objectives Learn about the technologies used inside a hard drive and how data is organized on the drive

More information

18-hdfs-gfs.txt Thu Oct 27 10:05: Notes on Parallel File Systems: HDFS & GFS , Fall 2011 Carnegie Mellon University Randal E.

18-hdfs-gfs.txt Thu Oct 27 10:05: Notes on Parallel File Systems: HDFS & GFS , Fall 2011 Carnegie Mellon University Randal E. 18-hdfs-gfs.txt Thu Oct 27 10:05:07 2011 1 Notes on Parallel File Systems: HDFS & GFS 15-440, Fall 2011 Carnegie Mellon University Randal E. Bryant References: Ghemawat, Gobioff, Leung, "The Google File

More information

Data Storage Infrastructure at Facebook

Data Storage Infrastructure at Facebook Data Storage Infrastructure at Facebook Spring 2018 Cleveland State University CIS 601 Presentation Yi Dong Instructor: Dr. Chung Outline Strategy of data storage, processing, and log collection Data flow

More information

An update on the scalability limits of the Condor batch system

An update on the scalability limits of the Condor batch system An update on the scalability limits of the Condor batch system D Bradley 1, T St Clair 1, M Farrellee 1, Z Guo 1, M Livny 1, I Sfiligoi 2, T Tannenbaum 1 1 University of Wisconsin, Madison, WI, USA 2 University

More information

*Prof.Komal Shringare

*Prof.Komal Shringare results by the use of table boundaries detection techniques and the use of text post-processing techniques to detect the noise and to correct bad-recognized words. Appendix OCR:- Optical character recognition,

More information

A BigData Tour HDFS, Ceph and MapReduce

A BigData Tour HDFS, Ceph and MapReduce A BigData Tour HDFS, Ceph and MapReduce These slides are possible thanks to these sources Jonathan Drusi - SCInet Toronto Hadoop Tutorial, Amir Payberah - Course in Data Intensive Computing SICS; Yahoo!

More information