Scribd Logo
Upload

Welcome to Scribd!

  • 6 views

    Final Prev Year Sol

    Uploaded by

    sabah nushra

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Final Prev Year Sol

    Uploaded by

    sabah nushra
    6 views11 pages

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    6 views11 pages

    Final Prev Year Sol

    Uploaded by

    sabah nushra

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 11

    -

    a 2020-21/
    1do Features of elastic cloud computing
    DmseLS: There needs to be many
    users

    to ensure
    using the

    full
    elastic
    resource
    cloud
    utilization.
    computing
    2) Users with small needs: If
    computing any user

    empting
    has needs then he may
    monopolize the entire resource. Thats

    why users need to have small


    computingneeds

    omsoita effeceurcooefload
    unrelated
    to t

    & unrelated so that there is a proper


    work load balance the resources.
    among
    ⑪ Such anexample can be where I
    want to store a fixed amount
    of data that will not be increased
    or decreased. Such an example can
    be
    staring the dataset of an

    event after the event is over,


    only for the purpose of future
    reference. Here the elasticity
    property of ca is not enjoyed.
    Here elasticity is not achieved because
    the dataset is fixed in size. It
    doesnt need elasticity. All the resumes
    that are alloted are sufficient.
    needs to be scaled down
    Nothing
    or scaled up.
    *
    s ame for both

    DVideo Capture: Local


    2) Face Detection: Local
    Face Recognition; aloud
    3)

    Because these devices can

    successfully record a video &


    detect face. But to
    recognize
    the face they must use the

    face database which is


    very
    in size.Thats why for
    large
    this task cloud must be used.
    ·Rambhonytosterine
    e)
    BS
    Shadow e
    master

    &Ask for locations"


    to write

    ② Get replica locations.


    write to elosest replice
    ⑤
    * Request commit to primary.
    ⑨ Primary instructs order of waites to
    secondaries
    ③ secondaries acknowledge.
    ⑦ Primary Ack to client.
    &a nt tee Shadoco
    -> the

    -DJEosin
    I f
    Ast round
    n
    Mapo
    sunh key= movie-name
    value = stream -
    count

    outputs (moviename, stiream-cont

    In part,
    the
    mapping all
    in
    the movies
    servers
    that are streamed, are all mapped
    with the moviename &
    key as

    value as the stream count.


    ence: input movie name, stream count -

    outputotmovie name, sum (stream


    -

    f
    count

    In the reduce part, for each of

    the different movienames we calculate


    the summation of the stream counts.
    2nd round s
    -

    e input (moviename, sum-stream count


    key= moviename

    value = sum-stream count


    -

    Reduc The reducer will


    -
    have to iterate

    through all the


    [veyvalue) pairs
    & Find the one with maze value.
    b) In big data we have
    huge
    amount of data that is not
    processed. So
    they are
    only like
    a code to us. But with proper

    processing we can
    get
    useful

    insights to those codes. Then the


    data will be really useful to us.

    Moreover
    big data technologies
    make it
    possible to organize the
    data so thatthey are understantable
    some hard
    to understand
    not
    just
    2 de.
    T Hybrid Cloud:
    8)

    empromise
    Privacy for the dataset (datacenter
    But
    computing capabilities for all
    of the
    departments. (public cloud
    environmt
    computing
    ⑪ Hybrid cloud can't be opted when-

    The computing capabilities should


    &

    be restricted too - Private doud.

    3) The data & capabilities


    computing available
    both need to be publicly
    for transparency public cloud
    1ancing: is the process of
    redistribution of workload in
    a distributed system
    ensuring no
    T

    resource is overloaded, under-loaded


    or idle.

    b) dcopy?

    *
    Stop source UM.

    keepy all pages over the


    network
    a start destination VM.

    langest downtime. Shortest total migration


    time
    Demand migrations
    ⑪ mmm
    *
    copy over critical as structures.
    A start destination UM

    ↳Page faults trigger network


    copy
    shortest downtime. Longest total migration
    time
    ive
    Precopy s

    over network.
    *
    Steratively copy pages
    *
    keep copying dirtied pages until threshold
    * at threshold,
    stop source VM,
    copy remaining pages, start destination(M
    downtime& total
    Balance between
    migration duration.

    &88 vs of
    big data i

    me huge amount of data


    velocity: billions of ADR pertime unit.
    sma

    vity: audio call, video call,


    text etc.
    message, mms

    veracity
    mus generated directly from the
    users 55 reliable
    value: useful
    -
    info by processing
    The data.
    ⑪ database manage-
    Sola
    ment
    using like
    Big Table
    system HBase

    HBase stores billions of rows of


    detailed call peazeds & performs
    of records in
    fast
    processing
    real time

    You might also like