Lecture 9 more cloud applications l.jpg
This presentation is the property of its rightful owner.
Sponsored Links
1 / 48

Lecture 9: More Cloud Applications PowerPoint PPT Presentation

  • Uploaded on
  • Presentation posted in: General

Lecture 9: More Cloud Applications. Xiaowei Yang (Duke University). News: Buffalo as Data Center Mecca. $1.9 billion, at least 200 employees Low-cost electric power, tax incentives, plenty of shovel-ready sites, cool climate. Review. Cloud Computing Elasticity Pay-as-you-go Challenges

Download Presentation

Lecture 9: More Cloud Applications

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript

Lecture 9 more cloud applications l.jpg

Lecture 9: More Cloud Applications

Xiaowei Yang (Duke University)

News buffalo as data center mecca l.jpg

News: Buffalo as Data Center Mecca

  • $1.9 billion, at least 200 employees

  • Low-cost electric power, tax incentives, plenty of shovel-ready sites, cool climate

Review l.jpg


  • Cloud Computing

    • Elasticity

    • Pay-as-you-go

  • Challenges

    • Security: co-residence, inference

    • Performance

      • Coarse-grained sharing

      • Lack of virtualized interface for specialized hardware

Today l.jpg


  • Cloud Applications

    • Execution augmentation for mobile devices

    • Energy saving for mobile

    • Energy saving for desktops

    • Disaster recovery

The case for energy oriented partial desktop migration l.jpg

The Case for Energy-Oriented Partial Desktop Migration

NiltonBila†, Eyal de Lara†, MattiHiltunen, Kaustubh Joshi,

H. Andr´esLagar-Cavillaand M. Satyanarayanan

Motivations l.jpg


  • Offices and homes have many PCs

  • But, they areoften left running idle

    • PCs idle on average 12 hours a day

      • “Skilled in the art of being idle” by Nedevschi et al. in NSDI 2009

    • 60% of desktops remain powered overnight

      • “After-hours power status of office equipment in the USA” by Webber, in Energy 2006

Why is it important l.jpg

Why is it important?

  • Dell Optiplex 745 Desktop

  • Peak power: 280W

  • Idle power: 102.1W

  • Sleep power: 1.2W

  • If we put one to sleep when it is idle, the saving is (102.1-1.2)W.

Why do we leave desktops on l.jpg

Why do we leave desktops on?

  • Applications with always on semantics

    • Skype, IM, email, personal media sharing

  • Interspersed activities with idle periods

    • Lunch break

    • Chatting with colleagues

Related work l.jpg

Related work

  • Full VM migration

    • LiteGreen, USENIX 2010 best paper

    • Encapsulate user session in VM

    • When idle, migrate VM to consolidation server and power down PC

    • When busy, migrate back to user’s PC





Partial vm migration l.jpg

Partial VM migration

  • Idle VM only access partial memory and disk state (working set)

  • Migrate only the working set to a server

    • Potentially a cloud server

    • Cloud provider can further aggregate

Advantages l.jpg


  • Small migration footprint

  • Client

    • Fast migration

    • Low energy cost

  • Network

    • Reduce bandwidth demand

  •   Server

    • More VMs per server

Feasibility study l.jpg

Feasibility Study

  • Can its desktop save energy by sleeping when an VM runs on the cloud?

  • Does the entire domain save energy by migrating idle sessions by sleeping?

Methodology l.jpg


  • Prototyped simple on-demand migration approach with SnowFlock

    • Prepared a VM image, and run the VM

    • After five minutes, used SnowFlock to clone the VM

    • Monitor memory and disk page migration to cloneVM

Setup l.jpg


  • Dell Optiplex 745 Desktop

    • 4GB RAM, 2.66GHz Intel C2D

    • Peak power: 280W

    • Idle power: 102.1W

    • Sleep power: 1.2W

  • VM Image:

    • Debian Linux 5

    • 1GB RAM

    • 12 GB disk

Workloads l.jpg


Memory request pattern l.jpg

Memory Request Pattern

  • Spatial locality

    • Pre-fetching

Page request interval l.jpg

Page Request Interval

  • 98% of request arrive in close succession

Potential sleep intervals l.jpg

Potential Sleep Intervals

Potential sleep intervals19 l.jpg

Potential Sleep Intervals

Potential sleep intervals20 l.jpg

Potential Sleep Intervals

Potential sleep intervals21 l.jpg

Potential Sleep Intervals

Energy savings an hour long trace l.jpg

Energy Savings: an hour-long trace

Hourly energy savings an overnight session l.jpg

Hourly Energy Savings: an overnight session

  • Saves 69% of energy

Memory footprint l.jpg

Memory footprint

  • A cloud node with 4GB of RAM can run ~30 VMs

Domain wide energy savings l.jpg

Domain-wide Energy Savings

Annual energy savings l.jpg

Annual Energy Savings

  • No partial migration

Annual energy savings27 l.jpg

Annual Energy Savings

  • V = 23

Annual savings l.jpg

Annual Savings

Open issues l.jpg

Open issues

  • Can it save cost?

    • Network

    • Cloud Rental

  • Frequent power cycling reduces hw life expectancy and limits power savings

    • Reduce number of sleep cycles and increase sleep duration

    • Predict page access patterns and prefetch

    • Leverage content addressable memory

  • Fast reintegration

    • Big Q: Can it be fast enough so that a user does not suffer a long delay?

  • Policies

    • When to migrate/re-integrate?

    • When does the desktop go to sleep?

    • On re-integration, should state be maintained in the cloud? For how long?

Disaster recovery as a cloud service economic benefits deployment challenges l.jpg

Disaster Recovery as a Cloud Service: Economic Benefits & Deployment Challenges

Timothy Wood and Emmanuel Cecchet, University of Massachusetts Amherst; K.K. Ramakrishnan, AT&T Labs—Research; PrashantShenoy, University of Massachusetts Amherst; Jacobus van derMerwe, AT&T Labs—Research; ArunVenkataramani, University of Massachusetts Amherst

Datacenter disasters l.jpg

Datacenter Disasters

  • Disasters cause expensive application downtime

  • Truck crash shuts down Amazon EC2 site center (May 2010)

  • Lightning strikes EC2 data (May 2009)

  • Comcast Down: Hunter shoots cable (2008)

  • Squirrels bring down NASDAQ exchange (1987 and 1994)

Dr fits in the cloud l.jpg

DR Fits in the Cloud

  • Customer: pay-as-you-go and elasticity

    • Normal is cheap (fewer resources for backup than normal operations)

    • Rapidly scale up resources after disaster is detected

  • Provider: high degree of multiplexing

    • Customers will not fail at once

    • Can offer extra services like disaster detection

What is disaster recovery l.jpg

What is disaster recovery

  • Use DR services to prevent lengthy service disruptions

  • Data backups + failover mechanism

    • Periodically replicate state

    • Switch to backup site after disaster

Dr metrics l.jpg

DR Metrics

  • Recovery Point Objective (RPO): the most recent backup time prior to any failure

  • Recovery Time Objective (RTO): how long it can take for an application to come back online after a failure occurs

    • Time to detect failure

    • Provision servers

    • Initialize applications

    • Configure networks to connect

Slide36 l.jpg

  • Performance

    • Have a minimal impact on the performance of each application being protected under failure-free operation

    • How can DR impact performance?

  • Consistency

    • The application can be restored to a consistent state

  • Geographic separation

    • Challenge: increasing network latency

Dr mechanisms l.jpg

DR Mechanisms

  • Hot Backup Site

    • Provides a set of mirrored stand-by servers that are always available

    • Minimal RTO and RPO

    • Use synchronous replication to prevent any data loss

Warm backup site l.jpg

Warm backup Site

  • Cheaply synchronize state during normal operations

  • Obtain resources on demand after failure

  • Short delay to resource provision and applications

Cost analysis study l.jpg

Cost analysis study

  • Compare DR in Colocation center to Cloud

  • Colocation

    • pays for servers and space at all times

  • Cloud DR

    • Pays for resources as they are used

Case study 1 l.jpg

Case Study 1

  • RUBiS: an ebay-like multi-tier web application

    • Three front ends

    • One database server

    • Only database state is replicated

Cost analysis l.jpg

Cost analysis

  • 99% Uptime cost (3 days of disaster per year)

Case 2 data warehouse l.jpg

Case 2: Data Warehouse

  • Post-disaster expensive due to high powered VM instance

  • Overall cheaper because 99% Uptime

Rpo vs cost tradeoff l.jpg

RPO vs Cost Tradeoff

  • Flexible

  • Colo has a fixed cost regardless of RPO requirements

Cost analysis summary l.jpg

Cost Analysis Summary

  • Cloud DR’s benefits depend on

    • Type of resources to run application

    • Variation between normal and post-disaster costs

    • RPO and RTO requirements

    • Uptime

  • Cloud is better if post-disaster cost much higher than normal mode

Provider challenges l.jpg

Provider Challenges

  • How to maximize revenue?

    • Makes money from storage in normal case

    • But must pay for servers and keep them available for DR

    • Possible solutions

      • Spot instances (EC2 uses them)

      • Higher prices for higher priority resources

  • Correlated failures

    • Large disasters may affect many

    • Possible solutions

      • Decide provision using a risk model

      • Spread out customers

Mechanisms needed for cloud dr l.jpg

Mechanisms Needed for Cloud DR

  • Network reconfiguration

    • Application must be brought up online after moved to a backup site

    • May require setting up a private business network

  • Security and Isolation

  • VM migration and cloning

    • Restore an application after a disaster is handled

    • Cloud provider does not support VM migration in and out cloud yet

Summary l.jpg


  • Cloud based disaster recovery

    • Can reduce cost

      • Up to 85% from a case study

    • Flexible tradeoff between cost and RPO

Forecast l.jpg


  • Next lecture

    • Another cloud application for group collaboration

  • Monday is in fall break

  • Next Wednesday

    • Midterm

    • http://www.cs.duke.edu/courses/fall10/cps296.2/syllabus.html

  • Login