Deadline-Aware Task Scheduling in a Tiered IoT
Infrastructure
Jianhua Fan, Xianglin Wei, Tongxiang Wang
Nanjing Telecommunications Technology Institute
Nanjing, China
fjh7659@163.com, wei_xianglin@163.com
Tian Lan, Suresh Subramaniam
The George Washington University
Washington DC, US
tlan@email.gwu.edu, suresh@email.gwu.edu
Abstract—With the proliferation of the Internet of Things
(IoT), the current “cloud-only” architectures cannot efficiently
handle IoT’s data processing and communications needs, while
providing satisfactory service latency to support emerging
mobile applications on the horizon that require almost real-
time responses. fog computing is introduced as a new
computing paradigm that distributes computation,
communication, control, and storage closer to the end users
along the “cloud-to-things” continuum. In this paper, we
present a deadline-aware task scheduling mechanism for fog
computing in a tiered IoT infrastructure, where service
providers exploit the collaboration between their own fog
nodes and the rented cloud resources to efficiently execute
users’ offloaded tasks, at large geographical scale. We first
formulate the task-scheduling problem in such a cloud-fog
environment as a multi-dimensional 0-1 knapsack problem
that is NP-hard, and then propose an efficient algorithmic
solution based on ant colony optimization heuristic. The main
objective is to maximize the profits of fog service provider
while meeting the tasks’ deadline constraint. Extensive
experimental results show that our proposed optimization and
solution significantly improves the system performance
compared with existing heuristics.
I. INTRODUCTION
As smart cities, smart transportation, and smart homes
are transitioning toward the real world, cloud computing,
which employs computation resources located far away from
the end users and relies on robust network infrastructure,
falls short on providing low-latency guarantees to deadline-
sensitive applications that are very common in Internet of
Things (IoT) scenarios [1].
To address this issue, fog computing (also known as
Edge Computing) extends the cloud computing paradigm to
the edge of the network, in close proximity to end devices
that serve as both data generator and consumer. Many
characteristics of fog computing make it the appropriate
platform to support critical IoT services and applications. By
placing fog resources (i.e., local computing infrastructures)
within one-hop of the IoT devices, fog computing can bring
a number of key advantages including: a) Low service
latency and location awareness; b) Wide-spread geographical
distribution and coverage; c) Mobility; d) Scalability with
respect to system/network scale, e) Ubiquitous wireless
access, f) Strong support for real-time and ultra-low-latency
applications, g) Heterogeneity and diversity [2]. A typical
architecture of fog computing is shown in Fig. 1. The
cloudlet entities act as the local fog computing platforms,
connecting to the remote datacenters through the Internet.
IoT devices, residing on the network edge, connect to the
cloudlet through a wireless network (e.g., cellular or WiFi)
and can offload their computation tasks to the cloudlet or the
datacenters, if the desired application deadlines can be met.
Figure 1. Architecture of a tiered IoT system including cloudlets for edge
computing, datacenters for cloud computing, and IoT devices.
Harnessing both fog and cloud computing can enable
more agile and context-aware services since the cloudlet
entities and the datacenter usually have much more resources
than IoT devices and are able to more rapidly execute
computation-intensive tasks. Moreover, placing computing
resources at the edge of the network allows fog nodes to
efficiently process latency-sensitive tasks in a timely manner,
while large-scale and latency-tolerant tasks can still be
efficiently processed by the cloud that is equipped with more
computing power. Many IoT applications such as big data
analytics [3] require the interplay and cooperation between
the edge (fog) and the core (cloud), calling for a joint
optimization of both fog and cloud resources in the network.
In this paper, we present a deadline-constrained task
scheduling framework for IoT systems with a joint fog and
cloud computing architecture. Our goal is to maximize the
total net profit received by service providers through task
scheduling and placement, while meeting application
deadline requirements and satisfying resource capacity
constraints in both fog and cloud networks. A fog service
provider can exploit the collaboration between its own fog
nodes (including cloudlet and participating mobile user
devices) and the rented cloud resources for efficiently
978-1-5090-5019-2/17/$31.00 ©2017 IEEE