Scribd Logo
Upload

Welcome to Scribd!

  • 13 views

    IoT Offloading

    Uploaded by

    pandaabinash702

    Copyright:

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

    IoT Offloading

    Uploaded by

    pandaabinash702
    13 views32 pages

    Original Title

    IoT Offloading (1)

    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
    13 views32 pages

    IoT Offloading

    Uploaded by

    pandaabinash702

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 32

    IoT Offloading

    Dr Sourav Kanti Addya


    Iot-fog-cloud architecture
    Need for IFC architecture
    1. cloud computing-based solutions do not always meet the expected quality of service (QoS) and quality of
    experience (QoE) requirements for some classes of IoT applications, especially latency-sensitive ones
    because of the long physical distance between the IoT devices and the remote cloud servers, scarce
    spectrum resources, and intermittent network connectivity.

    2. This has led to the emergence of fog computing, which extends the cloud computing resources (i.e.,
    computing, storage, and networking) closer to the data generation sources (i.e., IoT devices), thereby
    allowing for the prescribed QoS requirements of services and application to be met by enabling the fog
    computing devices (e.g., switches, gateways, and hubs) to process and offload most tasks on behalf of the
    cloud servers in a distributed manner.

    3. IoT-Fog-Cloud (IFC) systems formed by the integration of IoT, fog, and cloud are able to provide
    uninterrupted services and applications with significant QoS improvement along the things-to-cloud
    continuum.some IoT and fog devices can support the process of only one data type such as image, text,
    video, or audio.
    4. modern applications such as artificial intelligence and machine learning algorithms require to the
    computation of complex tasks, which typically include multiple types of input data.

    5. The diverse task requests also have a significant impact on offloading performance in the IFC systems.
    For example, some fog devices are unable to process the entire data of heavy tasks owing to a lack of
    storage and limitation of computational capability.
    Therefore, without an efficient resource allocation policy, a high rate of task request may lead to a high workload
    Difference between Cloud , fog and edge
    What is task offloading?
    The process of transferring resource intensive computational tasks to some external device
    in the network is known as offloading.
    Types of Task Offloading
    ● Full Offloading- Task is completely transferred to the remote server
    ● Binary Offloading - Task is either executed locally or transferred to remote
    server.
    ● Partial Offloading- Task is splitted into parts and can be simultaneously
    executed locally and at remote server. This is type of offloading is usually
    used for larger tasks having subtasks which can be executed parallely.
    Partial Offloading
    •A part of the computation is processed locally while the rest is offloaded to the MEC/Fog.
    •One of the efficient way to reduce the execution delay of applications.
    •Increase the battery life of UE devices.
    Binary offloading on Independent task
    In binary offloading, a binary indicator can be used to represent location where the task
    is executed.
    When the task is executed at local device it can be represented as D1=0, when the
    task is executed at fog node D2=1.
    D1 and D2 are two IoT devices.
    Binary offloading on Dependent task
    In case of dependency between two IoT devices there are four cases:
    Case 1: when two devices tasks are executed at local device.
    D1=0, D2=0

    In this case, IoT device D1 task is executed at the local device and then uploads
    output to the fog node then IoT device D2 will download the result from the fog
    node hereafter the output of IoT device D1 will become input for IoT device D2 will
    execute the task at local device D2. Because of dependency, D2 has to wait until
    it receives the output from D1.
    Case1: when D1=0 and D2=0

    Fig. 1: Case1
    Binary offloading on Dependent task

    Case 2: when D1 executed at fog node and D2 on local device.


    D1=1, D2=0
    In this case, IoT device D1 offloads its task to fog node then D1 execute the task
    on fog node then after its execution IoT device D2 download its results from fog
    node and that output becomes the input to the task of IoT device D2. Hence IoTD2
    executes task on IoT device D1. i.e.,it will execute locally.
    Case2: when D1=1 and D2=0

    Fig.2: Case2
    Binary offloading on Dependent task

    Case 3: when D1 executes at local device and D2 at fog node.


    D1=0, D2=1
    In this case, IoT device D1 executes its task offloads its task to fog node then
    D1 execute the task on fog node then after its execution IoT device D2
    download its results from fog node and that output becomes the input to the
    task of IoT device D2. Hence IoTD2 executes task on IoT device D1. i.e.,it will
    execute locally.
    Case3: when D1=0 and D2=1

    Fig : case3
    Binary offloading on Dependent task

    Case 4: when both device executes at fog node.


    D1=1, D2=1
    In this case, IoT device D1 offloads its task to the fog node and executes the
    task after its execution IoT device D2 offloads its task to the
    same fog node and takes the output of IoT D2 task for the next task execution,
    and that output becomes the input to device D2 task. Hence IoT device D2
    executes at the fog node.
    Case4: when D1=1 and D2=1

    Fig: Case4
    Applications of IoT offloading

    Task offloading scenario in Vehicular Edge Computing


    Applications of IoT offloading

    IoT Smart Healthcare System


    Applications of IoT offloading
    Smart agriculture
    Applications of IoT offloading
    Smart Pollution control
    How to calculate energy and latency
    In a smart home environment. To offload the task to fog node requires energy and
    time. How to calculate the latency and energy consumed during binary offloading
    technique with independent task with five fog nodes.
    Given - Arduino frequency - 16MHz
    Channel frequency - 10 MHz
    Distance - 100m
    Input size - 300 kb
    Output size - 10 kb
    Computational demand of a task - 210 million cycles
    Noise power -
    Max. trans. Power of IoT- 0.5W
    Max. trans. Power of fog node - 1.5W
    Computation power of Fog node- 0.35W
    Fog node Quota - 50
    Capacity of 1 VRU = 1000Ghz
    Local execution time
    Fog node execution time
    Uplink data rate for offloading task i for device j on Fog node k.
    Fog node execution time
    To calculate Channel gain
    Local execution energy
    Energy of IoT device
    Fog node execution time

    Uplink transmission time for task i of device to fog node k


    Fog node execution time
    For downlink data rate we consider fog node transmission power
    Fog node execution time

    ● Computational delay of task i on device j at fog node k


    Fog node execution time

    Total time = transmission time + receiving time + computational time


    Energy consumption at fog node

    Energy consumption at fog node k for task i on device j

    You might also like