dynamic resource allocation and distributed video transcoding in cloud computing

Embed Size (px)

Citation preview

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    1/22

    Optimization of distributed

    video transcoding using mapreduce and dynamic resource

    allocation in cloud computing

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    2/22

    ABSTRACT

    In this paper, we propose a HadoopMap Reduce based Distributed Video

    Transcoding System in a cloud

    computing environment thattranscodes various video codec

    formats into the MPEG-4 video format.

    We will optimize the input files andparallely split the file to store in cloud

    server.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    3/22

    Continued

    The users may access the computingresources by using computer, tablet,

    notebook, smartphone, pad computer

    or other devices. The cloud server provides and

    manages the applications and also

    storing the data remotely in cloud.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    4/22

    Thus, the encoding time to transcode

    large amounts of video content is

    exponentially reduced, facilitating atranscoding function.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    5/22

    Abstract- continued

    For performance evaluation, we focuson measuring the total time to

    transcode a data set into a target data

    set for three sets of experiments. Wehave proposed the solution that how

    video transcoding becomes smart and

    speeds-up due to the efficiency ofcloud computing.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    6/22

    Introduction

    Cloud computing is an Internet basedservices, where we share some of the

    services like software, platform,

    infrastructure, storage, databases tocomputer or other devices on demand

    by the users.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    7/22

    Services are sold on demand, for aminute/hourly basis, services are fully

    managed by the providers and

    consumer need only is a computerand Internet access.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    8/22

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    9/22

    Existing System

    Hadoop-based Distributed VideoTranscoding System in a cloudcomputing environment that transcodesvarious video codec formats into theMPEG-4 video format.

    This system provides various types ofvideo content to heterogeneous devices

    such as smart phones, personalcomputers, television, and pads.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    10/22

    We design and implement the systemusing the MapReduce framework,

    which runs on a Hadoop Distributed

    File System platform, and the mediaprocessing library Xuggler.

    Thus, the encoding time to transcode

    large amounts of video content isexponentially reduced, facilitating a

    transcoding function.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    11/22

    For performance evaluation, we focuson measuring the total time to

    transcode a data set into a target data

    set for three sets of experiments. We also analyze the experimental

    results, providing optimal Hadoop

    Distributed File System andMapReduce options suitable for video

    transcoding.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    12/22

    Disadvantages:

    There are no optimization techniques

    available for cloud resource.

    Transcoding is not effective.

    It will take more time for encoding the

    file.

    Performance is very less.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    13/22

    Proposed System:

    In this paper, we propose a Hadoop MapReduce based Distributed VideoTranscoding System in a cloudcomputing environment that transcodes

    various video codec formats into theMPEG-4 video format.

    The users may access the computing

    resources by using computer, tablet,notebook, smartphone, pad computer orother devices.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    14/22

    The cloud server provides and managesthe applications and also storing the dataremotely in cloud.

    For performance evaluation, we focus onmeasuring the total time to transcode adata set into a target data set for threesets of experiments.

    We have proposed the solution that howvideo transcoding becomes smart andspeeds-up due to the efficiency of cloudcomputing.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    15/22

    Advantages:

    Optimization techniques available for

    cloud resource.

    Transcoding is effective. It is smart and efficient for transcoding the

    file.

    Performance is high.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    16/22

    Hadoop Map Reduce

    Technique Hadoop MapReduce (Hadoop

    Map/Reduce) is a software frameworkfor distributed processing of large datasets on compute clusters of commodity

    hardware. The framework takes care of scheduling

    tasks, monitoring them and re-executing

    any failed tasks. The primary objective of Map/Reduce is

    to split the input data set intoindependent chunks that are processed

    in a completely parallel manner.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    17/22

    The Hadoop MapReduce frameworksorts the outputs of the maps, which

    are then input to the reduce tasks.

    Typically, both the input and the outputof the job are stored in a file system.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    18/22

    System Architecture

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    19/22

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    20/22

    Conclusion

    We proposed a Hadoop Map Reducebased Distributed Video Transcoding

    System in a cloud computing

    environment that transcodes variousvideo codec formats into the MPEG-4

    video format.

    Our system ensures uniform transcoded

    video quality and a fast transcoding

    process by applying HDFS and

    MapReduce, the core techniques in

    cloud computing enabling technologies.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    21/22

    We can further optimize these splits,analyzing what is the optimum amount ofchunks to be generated, which certainlyvary according to the different data types

    (text, images, etc). Performance of Hadoop MapReducejobs can be improved without increasing

    hardware costs, by tuning several keyconfiguration parameters for clusterspecifications, input data size andprocessing complexity.

  • 7/29/2019 dynamic resource allocation and distributed video transcoding in cloud computing

    22/22

    Lot of research work is still going on tooptimize the resources of Cloud

    computing based upon scheduling,

    elasticity and scalability. Future work includes the experiments

    with public Cloud and with different set

    of inputs.