Intro 2 Cloud Computing

download Intro 2 Cloud Computing

of 48

Transcript of Intro 2 Cloud Computing

  • 8/12/2019 Intro 2 Cloud Computing

    1/48

    Cloud Computing

    The George Washington UniversityWashington DC

    Tarek El-Ghazawi

  • 8/12/2019 Intro 2 Cloud Computing

    2/48

    Observations- What led

    us to Cloud Computing!

  • 8/12/2019 Intro 2 Cloud Computing

    3/48

    3Tarek El-Ghazawi, GWU

    Evolution of Internet Computing

    P u b

    l i s h

    I n

    f o r m

    I n t e r a c t

    I n t e g r a

    t e

    T r a n s a c t

    D i s c o v e r

    ( i n

    t e l l i g e n c e

    )

    A u

    t o m a

    t e ( d i s c o v e r y

    )

    time

    scale

    S o c i a

    l m e

    d i a a n

    d n e

    t w o r k

    i n gSemantic

    discovery

    Data-intensiveHPC, cloudweb deep web

    D a t a m a r k e

    t p l a c e a

    n d a n a

    l y t i c s

    Wipro Chennai 2011

    3

    http://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.weather.com/http://www.weather.com/http://www.weather.com/http://www.weather.com/http://www.weather.com/http://www.weather.com/http://www.weather.com/http://www.weather.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.amazon.com/http://www.weather.com/http://www.treehouses.org/projects/cradlebeach.shtmlhttp://www.treehouses.org/projects/cradlebeach.shtml
  • 8/12/2019 Intro 2 Cloud Computing

    4/48

    4Tarek El-Ghazawi, GWU

    Top Ten Largest Databases

    0

    1000

    2000

    3000

    4000

    5000

    6000

    7000

    LOC CIA Amazon YOUTube ChoicePt Sprint Google AT&T NERSC Climate

    Top ten largest databases (2007)

    Terabytes

    Ref: http://www.focus.com/fyi/operations/10-largest-databases-in-the-world /

    4

  • 8/12/2019 Intro 2 Cloud Computing

    5/48

    5Tarek El-Ghazawi, GWU

    Data Center

    A data center is a facility used for housing a large

    amount of computers that

    store and serve vast amounts of data .

    Source: Google

    Hundreds of thousands of computers

    Peta- and Exa-scale datasets

  • 8/12/2019 Intro 2 Cloud Computing

    6/48

    6Tarek El-Ghazawi, GWU

    Data Center vs SupercomputersScale

    Blue Waters = 40K 8-core serversRoad Runner = 13K cell + 6K AMD serversMS Chicago Data Center = 50 containers = 100K 8-coreservers.

    Network ArchitectureSupercomputers: CLOS Fat Tree infiniband

    Low latency high bandwidthprotocols

    Data Center: IP basedOptimized for Internet Access

    Data StorageSupers: separate data farm

    GPFS or other parallel file system

    DCs: use disk on node +memcache

    Fat tree network

    Standard Data Center Network

  • 8/12/2019 Intro 2 Cloud Computing

    7/487Tarek El-Ghazawi, GWU

    GWU HPCL FacilityMY Own HPCL DataCenter

  • 8/12/2019 Intro 2 Cloud Computing

    8/488Tarek El-Ghazawi, GWU

  • 8/12/2019 Intro 2 Cloud Computing

    9/489Tarek El-Ghazawi, GWU

  • 8/12/2019 Intro 2 Cloud Computing

    10/4810Tarek El-Ghazawi, GWU

  • 8/12/2019 Intro 2 Cloud Computing

    11/4811Tarek El-Ghazawi, GWU

    GEORGE- GW CRAYXE6/XK7

    ~2000 Processor Core

    50+ TF

    Based on16 Core AMD Bulldozer chips

    12-core 64-bit AMD Opteron 6100Series processorsKepler GPGPUs

    64 GB registered ECC DDR3SDRAM per compute node

    1 Gemini routing andcommunications ASIC per twocompute nodes

  • 8/12/2019 Intro 2 Cloud Computing

    12/4812Tarek El-Ghazawi, GWU

  • 8/12/2019 Intro 2 Cloud Computing

    13/4813Tarek El-Ghazawi, GWU

    A Tour in the Google Data Center

    http://www.youtube.com/watch?v=zRwPSFpLX8I

    http://www.youtube.com/watch?v=zRwPSFpLX8Ihttp://www.youtube.com/watch?v=zRwPSFpLX8I
  • 8/12/2019 Intro 2 Cloud Computing

    14/48

    11 football fields

    Google Data CenterThe Dalles, OregonSource: NY Times

    Microsoft Data Center @ San Antonio, Texas

  • 8/12/2019 Intro 2 Cloud Computing

    15/4815Tarek El-Ghazawi, GWU

    What matters is on the inside

    *EPA report t o Congress on Server and Data Center Energy Effici ency, August 2007

    SEAS Seminar SeriesL Barroso and U Holzle, The Datacenter as a Computer, 2009

  • 8/12/2019 Intro 2 Cloud Computing

    16/4816Tarek El-Ghazawi, GWU

    Number GamesMicrosoft San Antonio Data Center:

    $550 mill ion in 2008475,000 square feet, or about 11 acres A large supply of nuclear and wind energy

    Future plans include solar panels8 million gallons of recycled waste water/month

    Energy used by US data centers61 bil lion kWh in 2006$4.5 bil lion1.5% of US electricity consumptionMore than that of all TVs in US

    Equal to that of 5.8 million householdsDoubled from 2000 to 2006Expected to double again by 2011

    *EPA repor t to Congress on Server and Data Center Energy Eff iciency , August 2007

  • 8/12/2019 Intro 2 Cloud Computing

    17/4817Tarek El-Ghazawi, GWU

    Forbes Predictions 2011

    Cloud Adopters Embrace Cloud For BothInnovation and Legacy Optimization

    Replace most new procurement with cloudstrategies.

    Start with private clouds as a stepping stone topublic clouds.

    Get real about security. Move to private cloudsas a back up to public clouds.

    The Bottom Line: Cloud Adoption Provide APath To The Next Generation Enterprise

  • 8/12/2019 Intro 2 Cloud Computing

    18/48

    18Tarek El-Ghazawi, GWU

    Internet 2: 100 Gigabit Network Infrastructure

    18

  • 8/12/2019 Intro 2 Cloud Computing

    19/48

    19Tarek El-Ghazawi, GWU

    Perils of Corporate Computing

    Own information systemsHowever

    Capital investmentHeavy fixed costsRedundant expendituresHigh energy cost, low CPU utilizationDealing with unreliable hardware

    High-levels of overcapacity (Technology and Labor)

    NOT SUSTAINABLE

  • 8/12/2019 Intro 2 Cloud Computing

    20/48

    20Tarek El-Ghazawi, GWU

    Google: CPU Utilization

    Activity profile of a sample of 5,000 Google Servers over a period of 6 months

  • 8/12/2019 Intro 2 Cloud Computing

    21/48

    21Tarek El-Ghazawi, GWU

    Google: Energy Overhead

  • 8/12/2019 Intro 2 Cloud Computing

    22/48

    22Tarek El-Ghazawi, GWU

    Google: Service Disruptions

  • 8/12/2019 Intro 2 Cloud Computing

    23/48

    23Tarek El-Ghazawi, GWU

    Utility Computing

    Computing may someday be organized as a

    public utility, just as the telephone system isorganized as a public utility

    John McCarthy, 1961

  • 8/12/2019 Intro 2 Cloud Computing

    24/48

    24Tarek El-Ghazawi, GWU

    Why Utility Computing Now

    Large data stores

    Fiber networks

    Commodity computing

    Multicore machines

    +

    Huge data setsUtilization/Energy

    Shared people

    > Utility Computing

  • 8/12/2019 Intro 2 Cloud Computing

    25/48

    25Tarek El-Ghazawi, GWU

    Utility Computing

    Let economy of scale prevail

    Outsource all the trouble to someoneelse

    The utility provider will share theoverhead costs among many customers,amortizing the costs

    You only pay for: the amortized overhead Your real CPU / Storage / Bandwidth

    usage

  • 8/12/2019 Intro 2 Cloud Computing

    26/48

    26Tarek El-Ghazawi, GWU

    Data Intensive Computing

    Data collection too large to transmiteconomically over Internet --- Petabyte datacollectionsComputation produces small data outputcontaining a high density of informationImplemented in CloudsEasy to write programs, fast turn around.MapReduce.

    Map(k1, v1) -> list (k2, v2)Reduce(k2,list(v2)) -> list(v3)

    Hadoop, PIG, HDFS, Hbase

    Sawzall, Google File System, BigTable

  • 8/12/2019 Intro 2 Cloud Computing

    27/48

    27Tarek El-Ghazawi, GWU

    Cloud Challenges Alignment with the needs of the business / user /

    non-computer specialists / community and societyNeed to address the scalability issue: large scaledata, high performance computing, automation,response time, rapid prototyping, and rapid time toproduction

    Need to effectively address (i) ever shorteningcycle of obsolescence, (ii) heterogeneity and (iii)rapid changes in requirements

    Transform data from diverse sources intointelligence and deliver intelligence to rightpeople/user/systems

    What about providing all this in a cost-effectivemanner?

    6/23/2010Wipro Chennai 2011 27

  • 8/12/2019 Intro 2 Cloud Computing

    28/48

    28Tarek El-Ghazawi, GWU

    Cloud Computing- A Formal Perspective

  • 8/12/2019 Intro 2 Cloud Computing

    29/48

    29Tarek El-Ghazawi, GWU

    Cloud Computing

    Cloud computing NIST definition A model for enabling ubiquitous, convenient, on-demand network access to a shared pool ofconfigurable computing resources

    Networks, Servers, Storage, Applications, Services,

    That can be rapidly provisioned and released withminimal management effort or service providerinteraction. This cloud model is composed of fiveessential characteristics, three service models, and fourdeployment models.

    Preston A. Cox: Mobile cloud computing: Devices, trends, issues, and the enablingtechnologies

    29

  • 8/12/2019 Intro 2 Cloud Computing

    30/48

    30Tarek El-Ghazawi, GWU

    The Five Essential Cloud Characteristics

    On-demand self-service A consumer can unilaterally provision computingcapabilities as needed

    Broad network accessCapabilities are available over the network

    Resource poolingResources are pooled to serve multiple consumers

    Rapid elasticityCapabilities can be elastically provisioned and released

    Measured serviceResource usage can be monitored and quantified

  • 8/12/2019 Intro 2 Cloud Computing

    31/48

    31Tarek El-Ghazawi, GWU

    The Three BASIC Cloud Service Model

    SaaS- Software as a Service- Rents software on a subscript ion basis- Service includes software, hardware and support- Users access the service through authorized device- Suitable for a company to outsource hosting of apps

    PaaS Platform as a Service- Vendor offers development environment to application

    developers- Provide develops toolkits, building blocks, payment hooks

    IaaS Infrastructure as a Service- Processing power and storage service

  • 8/12/2019 Intro 2 Cloud Computing

    32/48

    32Tarek El-Ghazawi, GWU

    Cloud Computing Services cont.

  • 8/12/2019 Intro 2 Cloud Computing

    33/48

    33Tarek El-Ghazawi, GWU

    The Four Cloud Deployment Models

    Private cloud: The cloud infrastructure is provisionedfor exclusive use by a single organization

    Community cloud: The cloud infrastructure isprovisioned for exclusive use by a specific communityof consumers from organizations that have sharedconcerns (e.g., mission, security requirements, policy,and compliance considerations)

    Public cloud: The cloud infrastructure is provisionedfor open use by the general public.

    Hybrid cloud: The cloud infrastructure is acomposition of two or more distinct cloudinfrastructures (private, community, or public)

  • 8/12/2019 Intro 2 Cloud Computing

    34/48

    34Tarek El-Ghazawi, GWU

    Software Stack

    Mobile (Android), Thin client (Zonbu) Thickclient (Google Chrome)

    Identity, Integration Payments, Mapping,Search, Video Games, Chat

    Peer-to-peer (Bittorrent), Web app (twitter),SaaS (Google Apps, SAP)

    Java Google Web Toolkit , Django, Ruby onRails , .NET

    S3, Nirvanix, Rackspace Cloud Files,Savvis,Full virtualization (GoGrid), Management(RightScale), Compute (EC2), Platform

    (Force.com)

    NIST: Interactions between Actors in Cloud

  • 8/12/2019 Intro 2 Cloud Computing

    35/48

    35Tarek El-Ghazawi, GWU

    NIST: Interactions between Actors in CloudComputing

    35

    CloudConsumer

    Cloud ProviderCloud Broker

    Cloud Auditor

    The communication path between a cloud provider & a cloud consumerThe communication paths for a cloud auditor to collect auditing informationThe communication paths for a cloud broker to provide service to a cloudconsumer

  • 8/12/2019 Intro 2 Cloud Computing

    36/48

    36Tarek El-Ghazawi, GWU

    Windows Azure

    Enterprise-level on-demand capacity builder

    Fabric of cycles and storage available on-request for acost

    You have to use Azure API to work with the

    infrastructure offered by MicrosoftSignificant features: web role, worker role , blobstorage, table and drive-storage

    6/23/2010Wipro Chennai 2011 36

  • 8/12/2019 Intro 2 Cloud Computing

    37/48

    37Tarek El-Ghazawi, GWU

    Google App Engine

    This is more a web interface for a developmentenvironment that offers a one stop facility for design,development and deployment Java and Python-basedapplications in Java, Go and Python.

    Google offers the same reliability, availability andscalability at par with Googles own applications

    Interface is software programming based

    Comprehensive programming platform irrespective of

    the size (small or large)Signature features: templates and appspot, excellentmonitoring and management console

    6/23/2010Wipro Chennai 2011 37

  • 8/12/2019 Intro 2 Cloud Computing

    38/48

    38Tarek El-Ghazawi, GWU

    Amazon EC2 Amazon EC2 is one large complex web service.

    EC2 provided an API for instantiating computinginstances with any of the operating systemssupported.

    It can facilitate computations through Amazon MachineImages (AMIs) for various other models.

    Signature features: S3, Cloud Management Console,MapReduce Cloud, Amazon Machine Image (AMI)

    Excellent distr ibution, load balancing, cloudmonitoring tools

    6/23/2010Wipro Chennai 2011 38

  • 8/12/2019 Intro 2 Cloud Computing

    39/48

    39Tarek El-Ghazawi, GWU 39

    Amazon Elastic Compute Cloud

    a web service that provides resizablecompute capacity in the cloud. It is designedto make web-scale computing easier fordevelopers.

    Create Amazon Machine Image (AMI)

    Upload the AMI into Amazon S3

    Use Amazon EC2 web service to manage

    Pay as you go

  • 8/12/2019 Intro 2 Cloud Computing

    40/48

    40Tarek El-Ghazawi, GWU

    Amazon Simple Storage Service storage for the Internet. It is designed to make web-scale computing easier for developers.

    Write, read, and delete objects

    Unlimited objects

    Authorization mechanisms

    REST and SOAP interfaces

    HTTP/BitTorrent protocol

    40

  • 8/12/2019 Intro 2 Cloud Computing

    41/48

    41Tarek El-Ghazawi, GWU

    Amazon Pricing

    Compute

    $0.10 - Small Instance (Default) 1.7 GB of memory, 1 EC2 ComputeUnit (1 virtual core with 1 EC2 Compute Unit), 160 GB of instancestorage, 32-bit platform$0.40 - Large Instance$0.80 - Extra Large Instance

    Data Transfer$0.100 per GB - all data transfer in $0.170 per GB - first 10 TB / monthdata transfer out$0.130 per GB - next 40 TB / month data transfer out

    $0.110 per GB - next 100 TB / month data transfer out$0.100 per GB - data transfer out / month over 150 TB

    Looks inexpensive, but really?41

  • 8/12/2019 Intro 2 Cloud Computing

    42/48

    42Tarek El-Ghazawi, GWU

    Putting Numbers Together

    EC21K instance hours, 1TB data in & out = $37010K instance hours, 1TB data in & out = $1,270100K instance hours, 1TB data in & out = $100,270

    S310TB storage, 100GB data in &out = 1,527.00

    100TB storage, 1 TB data in &out = 15,270.001PB storage, 10 TB data in &out = 152,700.00

    42

  • 8/12/2019 Intro 2 Cloud Computing

    43/48

    43Tarek El-Ghazawi, GWU

    Downtime

    7.1. Downtime and Service Suspensions. In additionto our rights to terminate or suspend Services to youas described in Section 3 above, you acknowledgethat: (i) your access to and use of the Services may besuspended for the duration of any unanticipated or

    unscheduled downtime or unavailability of any portionor all of the Services for any reason, including as aresult of power outages, system failures or otherinterruptions ; and (ii) we shall also be entitled, withoutany liability to you, to suspend access to any portionor all of the Services at any time , on a Service-widebasis

    43

  • 8/12/2019 Intro 2 Cloud Computing

    44/48

    44Tarek El-Ghazawi, GWU

    Security

    7.2. Security. We strive to keep Your Content secure,but cannot guarantee that we will be successful atdoing so , given the nature of the Internet. Accordingly,without limitation to Section 4.3 above and Section11.5 below, you acknowledge that you bear sole

    responsibility for adequate security, protection andbackup of Your Content. We strongly encourage you,where available and appropriate, to use encryptiontechnology to protect Your Content from unauthorizedaccess and to routinely archive Your Content. We willhave no liability to you for any unauthorized access oruse, corruption, deletion, destruction or loss of any ofYour Content.

    44

  • 8/12/2019 Intro 2 Cloud Computing

    45/48

    45Tarek El-Ghazawi, GWU

    Amazon S3 SLA

    45

  • 8/12/2019 Intro 2 Cloud Computing

    46/48

    46Tarek El-Ghazawi, GWU

    Open Cloud Computing Interface InfrastructureEC2 API

    Simple Storage Service (S3) API

    Windows Azure Storage Service REST APIs

    Windows Azure Service Management REST APIs

    Deltacloud API

    Rackspace Cloud Servers API

    Rackspace Cloud Files API

    Cloud Data Management Interface

    vCloud API

    GlobusOnline REST API 46

    Cloud Interoperability Standards

  • 8/12/2019 Intro 2 Cloud Computing

    47/48

    47Tarek El-Ghazawi, GWU

    Challenges

    Alignment with the needs of the business / user / non-computer specialists / community and society

    Need to address the scalabili ty issue: large scale data,high performance computing, automation, responsetime, rapid prototyping, and rapid t ime to production

    Need to effectively address (i) ever shortening cycle ofobsolescence, (ii) heterogeneity and (iii) rapid changes inrequirements

    Transform data from diverse sources into intelligenceand deliver intelligence to right people/user/systems

    What about providing all this in a cost-effective manner?

    6/23/2010Wipro Chennai 2011 47

    C l i

  • 8/12/2019 Intro 2 Cloud Computing

    48/48

    Conclusions

    Economics and technology advances side by sidedrive our IT transformationOne such transformation is utility computing and thecloud which is driven by

    Necessity of having excess capacity to deal withprocessingRising complexity of keeping and maintaining ITcapabili ties in houseThe explosion of data and its distribution across the

    world