CS 91: Cloud Systems & Datacenter Networks Failures & Replica=on

Size: px
Start display at page:

Download "CS 91: Cloud Systems & Datacenter Networks Failures & Replica=on"

Transcription

1 CS 91: Cloud Systems & Datacenter Networks Failures & Replica=on

2 Types of Failures fail stop : process/machine dies and doesn t come back. Rela=vely easy to detect. (oien planned) performance degrada=on: something has failed that makes it slow, but it s s=ll correct (straggler). Harder to detect. Byzan=ne : process has failed but is s=ll running. Might be incorrect (spewing garbage) or even malicious. Can t be trusted. VERY hard to detect.

3 Failure Impact: Availability The degree to which your system is opera=ng Op=ons aren t necessarily discrete: Fully opera=onal: all is good, full capacity Down: all is broken, no service for anybody In- between: service is degraded, but accessible Example: 100 servers needed to handle load, but only 75 are currently online

4 Failure Impact: Availability The degree to which your system is opera=ng OIen measured by nines of up=me

5 Failure Impact: Availability The degree to which your system is opera=ng OIen measured by nines of up=me

6 Failure Sources Hardware (oien disks, why?) h]ps:// v=tdacjrsceq4

7 Failure Sources Hardware (oien disks, why?) SoIware bugs Configura=on / human mistakes Network (Internet) connec=vity Planned maintenance Most common: soiware bugs

8 Cloud / Datacenter Scale So, how reliable must our hardware and soiware be to become reliable enough? If it s not 100%, then it doesn t really ma]er Even if the failure rate of any one thing is really low, there are SO MANY things in a datacenter, something will fail soon.

9 Fault- tolerant SoIware With so many failure sources, it s cri=cal that soiware be made reliable. Pros: can handle unexpected failures can handle planned maintenance, (de)commissions Con: increased soiware complexity

10 Fault- tolerant SoIware Common solu=on: (What s the most important principle in systems design?)

11 Fault- tolerant SoIware Common solu=on: Abstrac=on! Hide complex details whenever possible Typically build a layer of soiware infrastructure that can handle common failures Build applica=on logic on top of that Get to assume that (many) failures won t happen

12 Seen Before: ISIS (+ others) Important system that doesn t want to worry about failures: Air Traffic Control ISIS SoIware Reliability Layer The reali=es of networks and distributed systems

13 Will See Again: Harp (+ others) Important system that doesn t want to worry about failures: Storing NFS files reliably Harp Reliability Layer The reali=es of networks and distributed systems

14 In General Important system that doesn t want to worry about failures: (Your applica=on) Layer that does something to handle some failures. The reali=es of networks and distributed systems

15 Failure: what can we do? Suppose you re soon to take an exam, but you re worried about your pencil breaking (let s say it s a 20% chance) Easy solu=on: bring mul=ple (equal) pencils Redundancy: chances that they all break is very low (assuming breaks are independent)

16 Replica=on If something important might fail, keep some backups / spares around ready to stand in For data, this implies it must be copied to mul=ple loca=ons (replicated) Not so simple to implement

17 Tough Ques=ons What type(s) of failures must we survive? How many failures must we survive? How do we find a replica if failure happens? What sort of consistency seman=cs must be maintained between replicas? If failures make the situa=on bad, what are we willing to give up?

18 Brewer s CAP Theorem Consistency Availability Par==on tolerance Pick two*. * h]p:// twelve- years- later- how- the- rules- have- changed

19 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance All machines available for service un=l network par==ons.

20 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance Can t operate, even if online. (As if these two stop- failed.) Any machine that con=nues opera=ng must be in the majority par==on (also applies to stop failures).

21 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance Can t operate, even if online. (As if these two stop- failed.) Generally, to deal with fail- stop failures, need 2N + 1 machines to survive N failures because N+1 cons=tutes a majority. (Here we can survive 2 failures with 2(2) + 1 = 5)

22 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance This case is less well- defined. Can t really build your system such that par==ons are impossible. If you think you re doing this, you probably s=ll have to give up one or the other if a par==on does occur.

23 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance X = 1, Y = 2 X = 1, Y = 2 Kept consistent un=l network par==ons.

24 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance X = 5 X = 1, Y = 2 Y = 9 X = 1, Y = 2 Changes might be no problem. When par==on heals, reconcile changes.

25 Brewer s CAP Theorem Consistency, Availability, Par==on tolerance X = 5 X = 1, Y = 2 X = 7 X = 1, Y = 2 Changes might conflict with one another. Much harder to reconcile changes (oien requires human help).

26 System Classifica=on Reality: systems fall somewhere in a spectrum Some systems even let you tune to your taste ACID (Atomicity, Consistency, Isola=on, Durability) Strongly consistent and conserva=ve BASE (Basically Available, SoI state, Eventually consistent) Willing to tolerate some uncertainty

27 ACID (Favors C in CAP) From database world: data protec=on is key Based on idea of transac=on Sequence of commands that are related Atomicity: transac=on is all or nothing Consistency: transac=on sequence is ordered Isola=on: transac=ons behave as if serial Durability: if transac=on commits, it s safe on disk Be]er to pause than become inconsistent/unsafe

28 BASE (Favors A in CAP) Relaxes the constraints of ACID Basically Available: don t panic on failures SoI state: keep performance hints (eh ) Eventual consistency: data will eventually converge to all replicas if given =me Be]er to diverge than stop serving users

29 Comparison ACID Easier to reason about Safer BASE Scales well, more performance Usually available (more 9 s) Less scalable Lower performance Failures might render system unusable (fewer 9 s) Consistency unclear to users Reconciling diverged state is a hard problem

30 Paper Preview Characterizing Cloud Compu=ng Hardware Reliability Replica=on in the Harp File System The Google File System Paxos Made Simple

Distributed systems Lecture 6: Elec3ons, consensus, and distributed transac3ons. Dr Robert N. M. Watson

Distributed systems Lecture 6: Elec3ons, consensus, and distributed transac3ons. Dr Robert N. M. Watson Distributed systems Lecture 6: Elec3ons, consensus, and distributed transac3ons Dr Robert N. M. Watson 1 Last 3me Saw how we can build ordered mul3cast Messages between processes in a group Need to dis3nguish

More information

Data Management in the Cloud: Limitations and Opportunities. Annies Ductan

Data Management in the Cloud: Limitations and Opportunities. Annies Ductan Data Management in the Cloud: Limitations and Opportunities Annies Ductan Discussion Outline: Introduc)on Overview Vision of Cloud Compu8ng Managing Data in The Cloud Cloud Characteris8cs Data Management

More information

Data Center Evolu.on and the Cloud. Paul A. Strassmann George Mason University November 5, 2008, 7:20 to 10:00 PM

Data Center Evolu.on and the Cloud. Paul A. Strassmann George Mason University November 5, 2008, 7:20 to 10:00 PM Data Center Evolu.on and the Cloud Paul A. Strassmann George Mason University November 5, 2008, 7:20 to 10:00 PM 1 Hardware Evolu.on 2 Where is hardware going? x86 con(nues to move upstream Massive compute

More information

File System Reliability (part 2)

File System Reliability (part 2) File System Reliability (part 2) Main Points Approaches to reliability Careful sequencing of file system opera@ons Copy- on- write (WAFL, ZFS) Journalling (NTFS, linux ext4) Log structure (flash storage)

More information

SCALABILITY AND AVAILABILITY

SCALABILITY AND AVAILABILITY SCALABILITY AND AVAILABILITY Real Systems must be Scalable fast enough to handle the expected load and grow easily when the load grows Available available enough of the time Scalable Scale-up increase

More information

DDC Sequencing and Redundancy

DDC Sequencing and Redundancy DDC Sequencing and Redundancy Presenter Sequencing Importance of sequencing Essen%al piece to designing and delivering a successful project Defines how disparate components interact to make up a system

More information

Telephone Related Queries (TeRQ) IETF 85 (Atlanta)

Telephone Related Queries (TeRQ) IETF 85 (Atlanta) Telephone Related Queries (TeRQ) IETF 85 (Atlanta) Telephones and the Internet Our long- term goal: migrate telephone rou?ng and directory services to the Internet ENUM: Deviated significantly from its

More information

Migrating to Hosted Telephony. Your ultimate guide to migrating from on premise to hosted telephony. www.ucandc.com

Migrating to Hosted Telephony. Your ultimate guide to migrating from on premise to hosted telephony. www.ucandc.com Migrating to Hosted Telephony Your ultimate guide to migrating from on premise to hosted telephony Intro What is covered in this guide? A professional and reliable business telephone system is a central

More information

Big Data Storage Options for Hadoop Sam Fineberg, HP Storage

Big Data Storage Options for Hadoop Sam Fineberg, HP Storage Sam Fineberg, HP Storage SNIA Legal Notice The material contained in this tutorial is copyrighted by the SNIA unless otherwise noted. Member companies and individual members may use this material in presentations

More information

Summary of Cloud Compu.ng (CC) from the paper Abovce the Clouds: A Berkeley View of Cloud Compu.ng (Feb. 2009)

Summary of Cloud Compu.ng (CC) from the paper Abovce the Clouds: A Berkeley View of Cloud Compu.ng (Feb. 2009) Summary of Cloud Compu.ng (CC) from the paper Abovce the Clouds: A Berkeley View of Cloud Compu.ng (Feb. 2009) Defini.ons (I) Cloud Compu)ng refers to both the applica)ons delivered as services over the

More information

Data Center 2020. DC planning for the next 5 10 years. Copyright 2004-2013 Experture and Robert Frances Group, all rights reserved

Data Center 2020. DC planning for the next 5 10 years. Copyright 2004-2013 Experture and Robert Frances Group, all rights reserved DC planning for the next 5 10 years Topics to be Discussed Introduc=on Indirect Drivers Technology Direct Drivers Data Center DC Management DC Opera=ons s and Disaster Recovery 2 Introduc=on The future

More information

Storage Architectures for Big Data in the Cloud

Storage Architectures for Big Data in the Cloud Storage Architectures for Big Data in the Cloud Sam Fineberg HP Storage CT Office/ May 2013 Overview Introduction What is big data? Big Data I/O Hadoop/HDFS SAN Distributed FS Cloud Summary Research Areas

More information

Design considera-ons and Guiding Principles for Implemen-ng Cloud Security. William Stearns Security Analyst CloudPassage

Design considera-ons and Guiding Principles for Implemen-ng Cloud Security. William Stearns Security Analyst CloudPassage Design considera-ons and Guiding Principles for Implemen-ng Cloud Security William Stearns Security Analyst CloudPassage In a nutshell How do Cloud Servers differ from Data Center Servers? How do the differences

More information

OS/Run'me and Execu'on Time Produc'vity

OS/Run'me and Execu'on Time Produc'vity OS/Run'me and Execu'on Time Produc'vity Ron Brightwell, Technical Manager Scalable System SoAware Department Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation,

More information

Merit Member Conference 2015 Does Migra+ng to a Virtualized Data Center Make Sense in Higher Educa+on?

Merit Member Conference 2015 Does Migra+ng to a Virtualized Data Center Make Sense in Higher Educa+on? Merit Member Conference 2015 Does Migra+ng to a Virtualized Data Center Make Sense in Higher Educa+on? is underway with a pilot migra8on from a tradi8onal university data center to a scalable virtualized

More information

Transactions and ACID in MongoDB

Transactions and ACID in MongoDB Transactions and ACID in MongoDB Kevin Swingler Contents Recap of ACID transactions in RDBMSs Transactions and ACID in MongoDB 1 Concurrency Databases are almost always accessed by multiple users concurrently

More information

SELLING SHAREPOINT ENGAGEMENTS IN THE CLOUD ERA A GUIDE FOR MICROSOFT SI PARTNERS

SELLING SHAREPOINT ENGAGEMENTS IN THE CLOUD ERA A GUIDE FOR MICROSOFT SI PARTNERS SELLING SHAREPOINT ENGAGEMENTS IN THE CLOUD ERA A GUIDE FOR MICROSOFT SI PARTNERS Sponsored by Microsoft Corporation 1/ Selling SharePoint Online 2/ Selling SharePoint Farms on Windows Azure 3/ Selling

More information

Trevi: Watering down storage hotspots with cool fountain codes. Toby Moncaster University of Cambridge

Trevi: Watering down storage hotspots with cool fountain codes. Toby Moncaster University of Cambridge Trevi: Watering down storage hotspots with cool fountain codes Toby Moncaster University of Cambridge Trevi summary Ø Trevi is a cool new approach to data centre storage Ø based on exis;ng ideas that are

More information

UAB Cyber Security Ini1a1ve

UAB Cyber Security Ini1a1ve UAB Cyber Security Ini1a1ve Purpose of the Cyber Security Ini1a1ve? To provide a secure Compu1ng Environment Individual Mechanisms Single Source for Inventory and Asset Management Current Repor1ng Environment

More information

Update on the Cloud Demonstration Project

Update on the Cloud Demonstration Project Update on the Cloud Demonstration Project Khalil Yazdi and Steven Wallace Spring Member Meeting April 19, 2011 Project Par4cipants BACKGROUND Eleven Universi1es: Caltech, Carnegie Mellon, George Mason,

More information

THE WINDOWS AZURE PROGRAMMING MODEL

THE WINDOWS AZURE PROGRAMMING MODEL THE WINDOWS AZURE PROGRAMMING MODEL DAVID CHAPPELL OCTOBER 2010 SPONSORED BY MICROSOFT CORPORATION CONTENTS Why Create a New Programming Model?... 3 The Three Rules of the Windows Azure Programming Model...

More information

Protec'ng Informa'on Assets - Week 8 - Business Continuity and Disaster Recovery Planning. MIS 5206 Protec/ng Informa/on Assets Greg Senko

Protec'ng Informa'on Assets - Week 8 - Business Continuity and Disaster Recovery Planning. MIS 5206 Protec/ng Informa/on Assets Greg Senko Protec'ng Informa'on Assets - Week 8 - Business Continuity and Disaster Recovery Planning MIS5206 Week 8 In the News Readings In Class Case Study BCP/DRP Test Taking Tip Quiz In the News Discuss items

More information

PIONEER RESEARCH & DEVELOPMENT GROUP

PIONEER RESEARCH & DEVELOPMENT GROUP SURVEY ON RAID Aishwarya Airen 1, Aarsh Pandit 2, Anshul Sogani 3 1,2,3 A.I.T.R, Indore. Abstract RAID stands for Redundant Array of Independent Disk that is a concept which provides an efficient way for

More information

Apache Hadoop. Alexandru Costan

Apache Hadoop. Alexandru Costan 1 Apache Hadoop Alexandru Costan Big Data Landscape No one-size-fits-all solution: SQL, NoSQL, MapReduce, No standard, except Hadoop 2 Outline What is Hadoop? Who uses it? Architecture HDFS MapReduce Open

More information

MTD Keystone s Multiple Service Platforms

MTD Keystone s Multiple Service Platforms MTD s Multiple Service Platforms uses the Microso/ Office pla5orm and is an MS Access applica:on with integra:on to the common Microso/ Office applica:ons, namely Excel, Word, and Outlook. may be installed

More information

Can Cloud Hos+ng Providers Really Replace. Your Cri(cal IT Infrastructure?

Can Cloud Hos+ng Providers Really Replace. Your Cri(cal IT Infrastructure? Can Cloud Hos+ng Providers Really Replace Your Cri(cal IT Infrastructure? Housekeeping Welcome to Align s Webinar Can Cloud Hos+ng Providers Really Replace Your Cri(cal IT Infrastructure? Informa+on for

More information

Accelerating Application Performance on Virtual Machines

Accelerating Application Performance on Virtual Machines Accelerating Application Performance on Virtual Machines...with flash-based caching in the server Published: August 2011 FlashSoft Corporation 155-A W. Moffett Park Dr Sunnyvale, CA 94089 info@flashsoft.com

More information

A survey of big data architectures for handling massive data

A survey of big data architectures for handling massive data CSIT 6910 Independent Project A survey of big data architectures for handling massive data Jordy Domingos - jordydomingos@gmail.com Supervisor : Dr David Rossiter Content Table 1 - Introduction a - Context

More information

Special Relativity and the Problem of Database Scalability

Special Relativity and the Problem of Database Scalability Special Relativity and the Problem of Database Scalability James Starkey NimbusDB, Inc. The problem, some jargon, some physics, a little theory, and then NimbusDB. Problem: Database systems scale badly

More information

How To Protect Virtualized Data From Security Threats

How To Protect Virtualized Data From Security Threats S24 Virtualiza.on Security from the Auditor Perspec.ve Rob Clyde, CEO, Adap.ve Compu.ng; former CTO, Symantec David Lu, Senior Product Manager, Trend Micro Hemma Prafullchandra, CTO/SVP Products, HyTrust

More information

Kaseya Fundamentals Workshop DAY THREE. Developed by Kaseya University. Powered by IT Scholars

Kaseya Fundamentals Workshop DAY THREE. Developed by Kaseya University. Powered by IT Scholars Kaseya Fundamentals Workshop DAY THREE Developed by Kaseya University Powered by IT Scholars Kaseya Version 6.5 Last updated March, 2014 Day Two Overview Day Two Lab Review Patch Management Configura;on

More information

CS 91: Cloud Systems & Datacenter Networks Misc. Topics

CS 91: Cloud Systems & Datacenter Networks Misc. Topics CS 91: Cloud Systems & Datacenter Networks Misc. Topics Announcements EC2 username / passwords Lab today Spanner Enables external consistency, based on wall- clock Gme ConvenGonal Wisdom Don t use clocks

More information

benefit of virtualiza/on? Virtualiza/on An interpreter may not work! Requirements for Virtualiza/on 1/06/15 Which of the following is not a poten/al

benefit of virtualiza/on? Virtualiza/on An interpreter may not work! Requirements for Virtualiza/on 1/06/15 Which of the following is not a poten/al 1/06/15 Benefits of virtualiza/on Virtualiza/on Which of the following is not a poten/al benefit of virtualiza/on? A. cost effec/ve B. applica/on migra/on is easy C. improve applica/on performance D. run

More information

Cassandra A Decentralized Structured Storage System

Cassandra A Decentralized Structured Storage System Cassandra A Decentralized Structured Storage System Avinash Lakshman, Prashant Malik LADIS 2009 Anand Iyer CS 294-110, Fall 2015 Historic Context Early & mid 2000: Web applicaoons grow at tremendous rates

More information

Consistency Trade-offs for SDN Controllers. Colin Dixon, IBM February 5, 2014

Consistency Trade-offs for SDN Controllers. Colin Dixon, IBM February 5, 2014 Consistency Trade-offs for SDN Controllers Colin Dixon, IBM February 5, 2014 The promises of SDN Separa&on of control plane from data plane Logical centraliza&on of control plane Common abstrac&ons for

More information

Project Overview. Collabora'on Mee'ng with Op'mis, 20-21 Sept. 2011, Rome

Project Overview. Collabora'on Mee'ng with Op'mis, 20-21 Sept. 2011, Rome Project Overview Collabora'on Mee'ng with Op'mis, 20-21 Sept. 2011, Rome Cloud-TM at a glance "#$%&'$()!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"#$%&!"'!()*+!!!!!!!!!!!!!!!!!!!,-./01234156!("*+!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!&7"7#7"7!("*+!!!!!!!!!!!!!!!!!!!89:!;62!("$+!

More information

IT as a Service. Transforming IT with the Windows Azure Platform. November 2010

IT as a Service. Transforming IT with the Windows Azure Platform. November 2010 IT as a Service Transforming IT with the Windows Azure Platform November 2010 Version 1.0 11/9/2010 Contents Understanding IT as a Service... 1 Realizing IT as a Service: The Importance of PaaS... 4 What

More information

Experience with Grapevine: The Growth of a Distributed System

Experience with Grapevine: The Growth of a Distributed System : The Growth of a Distributed System Michael Schroeder, Andrew Birrell, and Roger Needham presented by: Ryan O Connor November 4 th, 2009 Grapevine : Grapevine is a distributed, replicated

More information

these three NoSQL databases because I wanted to see a the two different sides of the CAP

these three NoSQL databases because I wanted to see a the two different sides of the CAP Michael Sharp Big Data CS401r Lab 3 For this paper I decided to do research on MongoDB, Cassandra, and Dynamo. I chose these three NoSQL databases because I wanted to see a the two different sides of the

More information

Disaster Recovery Planning and Implementa6on. Chris Russel Director, IT Infrastructure and ISO Compu6ng and Network Services York University

Disaster Recovery Planning and Implementa6on. Chris Russel Director, IT Infrastructure and ISO Compu6ng and Network Services York University Disaster Recovery Planning and Implementa6on Chris Russel Director, IT Infrastructure and ISO Compu6ng and Network Services York University Agenda Background for York s I.T. Disaster Recovery Planning

More information

Hadoop and Map-Reduce. Swati Gore

Hadoop and Map-Reduce. Swati Gore Hadoop and Map-Reduce Swati Gore Contents Why Hadoop? Hadoop Overview Hadoop Architecture Working Description Fault Tolerance Limitations Why Map-Reduce not MPI Distributed sort Why Hadoop? Existing Data

More information

Texas Digital Government Summit. Data Analysis Structured vs. Unstructured Data. Presented By: Dave Larson

Texas Digital Government Summit. Data Analysis Structured vs. Unstructured Data. Presented By: Dave Larson Texas Digital Government Summit Data Analysis Structured vs. Unstructured Data Presented By: Dave Larson Speaker Bio Dave Larson Solu6ons Architect with Freeit Data Solu6ons In the IT industry for over

More information

References. Introduction to Database Systems CSE 444. Motivation. Basic Features. Outline: Database in the Cloud. Outline

References. Introduction to Database Systems CSE 444. Motivation. Basic Features. Outline: Database in the Cloud. Outline References Introduction to Database Systems CSE 444 Lecture 24: Databases as a Service YongChul Kwon Amazon SimpleDB Website Part of the Amazon Web services Google App Engine Datastore Website Part of

More information

Introduction to Database Systems CSE 444

Introduction to Database Systems CSE 444 Introduction to Database Systems CSE 444 Lecture 24: Databases as a Service YongChul Kwon References Amazon SimpleDB Website Part of the Amazon Web services Google App Engine Datastore Website Part of

More information

Strategies for Medical Device So2ware Development Presented By Anthony Giles of Blackwood Embedded Solu;ons And a Case Study by Francis Amoah of Creo

Strategies for Medical Device So2ware Development Presented By Anthony Giles of Blackwood Embedded Solu;ons And a Case Study by Francis Amoah of Creo Strategies for Medical Device So2ware Development Presented By Anthony Giles of Blackwood Embedded Solu;ons And a Case Study by Francis Amoah of Creo Medical Introduc;on Standards 60601-1 in par;cular

More information

Distributed Data Stores

Distributed Data Stores Distributed Data Stores 1 Distributed Persistent State MapReduce addresses distributed processing of aggregation-based queries Persistent state across a large number of machines? Distributed DBMS High

More information

Cloud Storage vs Physical Media What you need to know!

Cloud Storage vs Physical Media What you need to know! Cloud Storage vs Physical Media What you need to know! What You Need to Know about Cloud and Physical Storage 3 Cloud Storage: 4 What is the cloud? 4 Cloud Storage Pros 5 Cloud Storage Cons 6 Physical

More information

How To Build Cloud Storage On Google.Com

How To Build Cloud Storage On Google.Com Building Scalable Cloud Storage Alex Kesselman alx@google.com Agenda Desired System Characteristics Scalability Challenges Google Cloud Storage What does a customer want from a cloud service? Reliability

More information

Why Buy Eaton: Trusted Source to Manage Power

Why Buy Eaton: Trusted Source to Manage Power Why Buy Eaton: Trusted Source to Manage Power If you are responsible for power protec0on for small or medium data centers, Eaton can help you deliver a reliable power source for your business. Target Markets:

More information

Disaster Recovery and Business Continuity What Every Executive Needs to Know

Disaster Recovery and Business Continuity What Every Executive Needs to Know Disaster Recovery and Business Continuity What Every Executive Needs to Know Bruce Campbell & Sandra Evans Contents Why you need DR and BC What constitutes a Disaster? The difference between disaster recovery

More information

SC09 Tutorial M06 Cluster Construc5on Tutorial

SC09 Tutorial M06 Cluster Construc5on Tutorial SC09 Tutorial M06 Cluster Construc5on Tutorial Paul Marshall, Michael Oberg Theron Voran, Ma=hew Woitaszek University of Colorado, Boulder Na5onal Center for Atmospheric Research Advanced topics Parallel

More information

The Pros and Cons of Erasure Coding & Replication vs. RAID in Next-Gen Storage Platforms. Abhijith Shenoy Engineer, Hedvig Inc.

The Pros and Cons of Erasure Coding & Replication vs. RAID in Next-Gen Storage Platforms. Abhijith Shenoy Engineer, Hedvig Inc. The Pros and Cons of Erasure Coding & Replication vs. RAID in Next-Gen Storage Platforms Abhijith Shenoy Engineer, Hedvig Inc. @hedviginc The need for new architectures Business innovation Time-to-market

More information

In Memory Accelerator for MongoDB

In Memory Accelerator for MongoDB In Memory Accelerator for MongoDB Yakov Zhdanov, Director R&D GridGain Systems GridGain: In Memory Computing Leader 5 years in production 100s of customers & users Starts every 10 secs worldwide Over 15,000,000

More information

Map- reduce, Hadoop and The communica3on bo5leneck. Yoav Freund UCSD / Computer Science and Engineering

Map- reduce, Hadoop and The communica3on bo5leneck. Yoav Freund UCSD / Computer Science and Engineering Map- reduce, Hadoop and The communica3on bo5leneck Yoav Freund UCSD / Computer Science and Engineering Plan of the talk Why is Hadoop so popular? HDFS Map Reduce Word Count example using Hadoop streaming

More information

How to Study Mathematics Written by Paul Dawkins

How to Study Mathematics Written by Paul Dawkins How to Study Mathematics Written by Paul Dawkins Before I get into the tips for how to study math let me first say that everyone studies differently and there is no one right way to study for a math class.

More information

Big Data Management and NoSQL Databases

Big Data Management and NoSQL Databases NDBI040 Big Data Management and NoSQL Databases Lecture 4. Basic Principles Doc. RNDr. Irena Holubova, Ph.D. holubova@ksi.mff.cuni.cz http://www.ksi.mff.cuni.cz/~holubova/ndbi040/ NoSQL Overview Main objective:

More information

Google File System. Web and scalability

Google File System. Web and scalability Google File System Web and scalability The web: - How big is the Web right now? No one knows. - Number of pages that are crawled: o 100,000 pages in 1994 o 8 million pages in 2005 - Crawlable pages might

More information

Session 11 : (additional) Cloud Computing Advantages and Disadvantages

Session 11 : (additional) Cloud Computing Advantages and Disadvantages INFORMATION STRATEGY Session 11 : (additional) Cloud Computing Advantages and Disadvantages Tharaka Tennekoon B.Sc (Hons) Computing, MBA (PIM - USJ) POST GRADUATE DIPLOMA IN BUSINESS AND FINANCE 2014 Cloud

More information

Migra1ng to the Cloud

Migra1ng to the Cloud Migra1ng to the Cloud Barry P. Sheward LM Fellow barry.p.sheward@lmco.com June 9, 2014 2014 Lockheed Mar1n. All Rights Reserved. 1 About this presenta/on The presenta1on will cover a brief introduc1on

More information

The 3 questions to ask yourself about BIG DATA

The 3 questions to ask yourself about BIG DATA The 3 questions to ask yourself about BIG DATA Do you have a big data problem? Companies looking to tackle big data problems are embarking on a journey that is full of hype, buzz, confusion, and misinformation.

More information

CS 4604: Introduc0on to Database Management Systems

CS 4604: Introduc0on to Database Management Systems CS 4604: Introduc0on to Database Management Systems B. Aditya Prakash Lecture #1: Introduc/on Many slides based on material by Profs. Murali, Ramakrishnan and Faloutsos Course Informa0on Instructor B.

More information

WINDOWS AZURE EXECUTION MODELS

WINDOWS AZURE EXECUTION MODELS WINDOWS AZURE EXECUTION MODELS Windows Azure provides three different execution models for running applications: Virtual Machines, Web Sites, and Cloud Services. Each one provides a different set of services,

More information

INSTALLING GPS TRACKING DEVICES: Which Method is Best?

INSTALLING GPS TRACKING DEVICES: Which Method is Best? INSTALLING GPS TRACKING DEVICES: Which Method is Best? In any fleet tracking implementation, the most important step is to ensure devices are installed correctly so your business can start seeing the benefits

More information

NERSC Archival Storage: Best Practices

NERSC Archival Storage: Best Practices NERSC Archival Storage: Best Practices Lisa Gerhardt! NERSC User Services! Nick Balthaser! NERSC Storage Systems! Joint Facilities User Forum on Data Intensive Computing! June 18, 2014 Agenda Introduc)on

More information

Mobility in the Modern Factory. Discussion of Mobile Adop7on for the Factories of the Future

Mobility in the Modern Factory. Discussion of Mobile Adop7on for the Factories of the Future Mobility in the Modern Factory Discussion of Mobile Adop7on for the Factories of the Future Talking Points History Lesson The Reasons for Going Mobile Mobile Infrastructure Mobile Device Security BYOD

More information

Using RDBMS, NoSQL or Hadoop?

Using RDBMS, NoSQL or Hadoop? Using RDBMS, NoSQL or Hadoop? DOAG Conference 2015 Jean- Pierre Dijcks Big Data Product Management Server Technologies Copyright 2014 Oracle and/or its affiliates. All rights reserved. Data Ingest 2 Ingest

More information

Big Data, Deep Learning and Other Allegories: Scalability and Fault- tolerance of Parallel and Distributed Infrastructures.

Big Data, Deep Learning and Other Allegories: Scalability and Fault- tolerance of Parallel and Distributed Infrastructures. Big Data, Deep Learning and Other Allegories: Scalability and Fault- tolerance of Parallel and Distributed Infrastructures Professor of Computer Science UC Santa Barbara Divy Agrawal Research Director,

More information

Introduction to Datacenters & the Cloud

Introduction to Datacenters & the Cloud Introduction to Datacenters & the Cloud Introduction to Storage in the Cloud Alex M. Hurd Clarkson Open Source Institute April 14, 2015 Alex M. Hurd (COSI) Introduction to Datacenters & the Cloud April

More information

Massive Data Storage

Massive Data Storage Massive Data Storage Storage on the "Cloud" and the Google File System paper by: Sanjay Ghemawat, Howard Gobioff, and Shun-Tak Leung presentation by: Joshua Michalczak COP 4810 - Topics in Computer Science

More information

Chapter 3. Database Architectures and the Web Transparencies

Chapter 3. Database Architectures and the Web Transparencies Week 2: Chapter 3 Chapter 3 Database Architectures and the Web Transparencies Database Environment - Objec

More information

HDFS and Availability Data Retragement

HDFS and Availability Data Retragement 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

More information

Understanding NoSQL on Microsoft Azure

Understanding NoSQL on Microsoft Azure David Chappell Understanding NoSQL on Microsoft Azure Sponsored by Microsoft Corporation Copyright 2014 Chappell & Associates Contents Data on Azure: The Big Picture... 3 Relational Technology: A Quick

More information

Neil Stobart Cloudian Inc. CLOUDIAN HYPERSTORE Smart Data Storage

Neil Stobart Cloudian Inc. CLOUDIAN HYPERSTORE Smart Data Storage Neil Stobart Cloudian Inc. CLOUDIAN HYPERSTORE Smart Data Storage Storage is changing forever Scale Up / Terabytes Flash host/array Tradi/onal SAN/NAS Scalability / Big Data Object Storage Scale Out /

More information

Unitt www.unitt.com. Zero Data Loss Service (ZDLS) The ultimate weapon against data loss

Unitt www.unitt.com. Zero Data Loss Service (ZDLS) The ultimate weapon against data loss Zero Data Loss Service (ZDLS) The ultimate weapon against data loss The ultimate protection for your business-critical data In the past, ultimate data protection was a costly issue, if not an impossible

More information

Hunk & Elas=c MapReduce: Big Data Analy=cs on AWS

Hunk & Elas=c MapReduce: Big Data Analy=cs on AWS Copyright 2014 Splunk Inc. Hunk & Elas=c MapReduce: Big Data Analy=cs on AWS Dritan Bi=ncka BD Solu=ons Architecture Disclaimer During the course of this presenta=on, we may make forward looking statements

More information

Cloudian The Storage Evolution to the Cloud.. Cloudian Inc. Pre Sales Engineering

Cloudian The Storage Evolution to the Cloud.. Cloudian Inc. Pre Sales Engineering Cloudian The Storage Evolution to the Cloud.. Cloudian Inc. Pre Sales Engineering Agenda Industry Trends Cloud Storage Evolu4on of Storage Architectures Storage Connec4vity redefined S3 Cloud Storage Use

More information

Data Privacy and Data Security in Telemedicine Applica5ons. Patrick Harpes www.monitor it.lu

Data Privacy and Data Security in Telemedicine Applica5ons. Patrick Harpes www.monitor it.lu Data Privacy and Data Security in Telemedicine Applica5ons Patrick Harpes www.monitor it.lu Agenda Right to privacy Data/Informa@on security Data security measures Risks using telemedicine Composi@on of

More information

Big Data Storage, Management and challenges. Ahmed Ali-Eldin

Big Data Storage, Management and challenges. Ahmed Ali-Eldin Big Data Storage, Management and challenges Ahmed Ali-Eldin (Ambitious) Plan What is Big Data? And Why talk about Big Data? How to store Big Data? BigTables (Google) Dynamo (Amazon) How to process Big

More information

How To Develop An Application

How To Develop An Application What is Application Lifecycle Management? David Chappell Sponsored by Microsoft Corporation Copyright 2014 Chappell & Associates Defining application lifecycle management (ALM) isn t easy. Different people

More information

An introduction to disaster recovery. And how DrAAS from I.R.I.S. Ondit can help!

An introduction to disaster recovery. And how DrAAS from I.R.I.S. Ondit can help! An introduction to disaster recovery And how DrAAS from I.R.I.S. Ondit can help! Events That Impact Information Availability Events that require a data center move: Fewer than 1% of occurrences Natural

More information

Solidifying The Cloud

Solidifying The Cloud Solidifying The Cloud How to back up the Internet Raymond Blum Staff Site Reliability Engineer, International Man of Mystery, Google Lessons Learned Backing Up Google Ensuring durability and integrity

More information

High School Juniors Views on Free Enterprise and Entrepreneurship: A Na<onal Survey

High School Juniors Views on Free Enterprise and Entrepreneurship: A Na<onal Survey High School Juniors Views on Free Enterprise and Entrepreneurship: A Na

More information

REST (Representa.onal State Transfer) Ingegneria del So-ware e Lab. Università di Modena e Reggio Emilia Do<. Marzio Franzini

REST (Representa.onal State Transfer) Ingegneria del So-ware e Lab. Università di Modena e Reggio Emilia Do<. Marzio Franzini REST (Representa.onal State Transfer) Ingegneria del So-ware e Lab. Università di Modena e Reggio Emilia Do

More information

Data Warehouses and NoSQL Sharing Administra6ve Informa6on

Data Warehouses and NoSQL Sharing Administra6ve Informa6on Data Warehouses and NoSQL Sharing Administra6ve Informa6on Carmen Barandela So-ware Engineer CERN / GS AIS October 24 28, 2011 JINR/CERN Grid and Management Informa6on Systems Agenda Data Warehouses in

More information

Offensive & Defensive & Forensic Techniques for Determining Web User Iden<ty

Offensive & Defensive & Forensic Techniques for Determining Web User Iden<ty Offensive & Defensive & Forensic Techniques for Determining Web User Iden

More information

How the ersa Problem became the ersa Solu3on. Why a network and network security is impera3ve for ersa s NeCTAR cloud. Paul Bartczak Infrastructure

How the ersa Problem became the ersa Solu3on. Why a network and network security is impera3ve for ersa s NeCTAR cloud. Paul Bartczak Infrastructure How the ersa Problem became the ersa Solu3on. Why a network and network security is impera3ve for ersa s NeCTAR cloud. Paul Bartczak Infrastructure Manager About ersa eresearch SA is a collabora3ve joint

More information

Clouds Burst Horizon. J.Crowcro(/S.Hand, University of Cambridge

Clouds Burst Horizon. J.Crowcro(/S.Hand, University of Cambridge Clouds Burst Horizon J.Crowcro(/S.Hand, University of Cambridge Cloud Compu1ng? Q: What is Cloud Compu1ng, exactly? A: It depends but a rough defini1on might be on- demand internet- based compu1ng i.e.

More information

DISTRIBUTED SYSTEMS [COMP9243] Lecture 9a: Cloud Computing WHAT IS CLOUD COMPUTING? 2

DISTRIBUTED SYSTEMS [COMP9243] Lecture 9a: Cloud Computing WHAT IS CLOUD COMPUTING? 2 DISTRIBUTED SYSTEMS [COMP9243] Lecture 9a: Cloud Computing Slide 1 Slide 3 A style of computing in which dynamically scalable and often virtualized resources are provided as a service over the Internet.

More information

Philosophy of Network Opera3ng Systems and Intent APIs

Philosophy of Network Opera3ng Systems and Intent APIs Philosophy of Network Opera3ng Systems and Intent APIs Inder Monga CTO, ESnet Technology Exchange, October 2015 Cleveland, Ohio Agenda Background ESnet s Network Opera3ng System What is your Intent? Summary

More information

bigdata Managing Scale in Ontological Systems

bigdata Managing Scale in Ontological Systems Managing Scale in Ontological Systems 1 This presentation offers a brief look scale in ontological (semantic) systems, tradeoffs in expressivity and data scale, and both information and systems architectural

More information

Eventually Consistent

Eventually Consistent Historical Perspective In an ideal world there would be only one consistency model: when an update is made all observers would see that update. The first time this surfaced as difficult to achieve was

More information

Cloud Computing at Google. Architecture

Cloud Computing at Google. Architecture Cloud Computing at Google Google File System Web Systems and Algorithms Google Chris Brooks Department of Computer Science University of San Francisco Google has developed a layered system to handle webscale

More information

Copyright www.agileload.com 1

Copyright www.agileload.com 1 Copyright www.agileload.com 1 INTRODUCTION Performance testing is a complex activity where dozens of factors contribute to its success and effective usage of all those factors is necessary to get the accurate

More information

Outline. MCSA: Server Virtualization

Outline. MCSA: Server Virtualization MCSA: Server Virtualization Description Get hands-on instruction and practice implementing Microsoft Server Virtualization with Windows Server 2012 R2 Hyper-V and System Center 2012 R2 Virtual Machine

More information

WINDOWS AZURE DATA MANAGEMENT

WINDOWS AZURE DATA MANAGEMENT David Chappell October 2012 WINDOWS AZURE DATA MANAGEMENT CHOOSING THE RIGHT TECHNOLOGY Sponsored by Microsoft Corporation Copyright 2012 Chappell & Associates Contents Windows Azure Data Management: A

More information

Informa*on Management

Informa*on Management Informa*on Management Deepak Mohan SVP, Informa3on Management Group 1 Symantec Informa*on Management Strategy Protect Completely Dedupe Everywhere Delete Confidently Discover Efficiently Backup, archive

More information

Licensing++ for Clouds. Mark Perry

Licensing++ for Clouds. Mark Perry Licensing++ for Clouds Mark Perry Plan* 1. Cloud? 2. Survey 3. Some ques@ons 4. Some ideas 5. Some sugges@ons (that would be you) * Plan 9 future events such as these will affect you in the future Clouds

More information

High availability on the Catalyst Cloud

High availability on the Catalyst Cloud White paper High availability on the Catalyst Cloud Features and techniques to improve availability, resiliency and business continuity of web applications hosted on the Catalyst Cloud 3 February 2016

More information

Do Relational Databases Belong in the Cloud? Michael Stiefel www.reliablesoftware.com development@reliablesoftware.com

Do Relational Databases Belong in the Cloud? Michael Stiefel www.reliablesoftware.com development@reliablesoftware.com Do Relational Databases Belong in the Cloud? Michael Stiefel www.reliablesoftware.com development@reliablesoftware.com How do you model data in the cloud? Relational Model A query operation on a relation

More information

Distributed Storage Systems

Distributed Storage Systems Distributed Storage Systems John Leach john@brightbox.com twitter @johnleach Brightbox Cloud http://brightbox.com Our requirements Bright box has multiple zones (data centres) Should tolerate a zone failure

More information