Asian Journal of Research in Computer Science

11(4): 35-51, 2021; Article [Link].73758

ISSN: 2581-8260

Scheduling Algorithms Implementation for Real

Time Operating Systems: A Review
Gulistan Ahmead Ismael1*, Azar Abid Salih1, Adel AL-Zebari1, Naaman Omar1,
Karwan Jameel Merceedi1, Abdulraheem Jamil Ahmed1, Nareen O. M. Salim1,
Sheren Sadiq Hasan1, Shakir Fattah Kak1, Ibrahim Mahmood Ibrahim1
and Hajar Maseeh Yasin1
1Duhok Polytechnic University, Duhok, Kurdistan Region, Iraq.

Authors’ contributions

This work was carried out in collaboration among all authors. All authors read and approved the final
manuscript.

Article Information
DOI: 10.9734/AJRCOS/2021/v11i430269
Editor(s):
(1) Prof. M. A. Jayaram, Siddaganga Institute of Technology, India.
Reviewers:
(1) Neelesh Jain, Sagar Institute of Research & Technology, India.
(2) R. Jemima Priyadarsini, Bishop Heber College, India.
Complete Peer review History: [Link]

Received 01 June 2021

Accepted 11 September 2021
Review Article
Published 11 September 2021

ABSTRACT

The term "Real-Time Operating System (RTOS)" refers to systems wherein the time component is
critical. For example, one or more of a computer's peripheral devices send a signal, and the
computer must respond appropriately within a specified period of time. Examples include: the
monitoring system in a hospital care unit, the autopilot in the aircraft, and the safety control system
in the nuclear reactor. Scheduling is a method that ensures that jobs are performed at certain
times. In the real-time systems, accuracy does not only rely on the outcomes of calculation, and
also on the time it takes to provide the results. It must be completed within the specified time frame.
The scheduling strategy is crucial in any real-time system, which is required to prevent overlapping
execution in the system. The paper review classifies several previews works on many
characteristics. Also, strategies utilized for scheduling in real time are examined and their features
compared.

Keywords: Operating system OS; RTOS; real time; scheduling algorism.

_____________________________________________________________________________________________________

*Corresponding author: E-mail: [Link]@[Link];

 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

1. INTRODUCTION resulting in delays [19]. The technique for

processing the time is measured in time
Real-Time Operating System (RTOS) is crucial increments that allow information to be shared
for mechanical and electrical systems that need [20, 21]. Time for processes the priority and
real-time operation, since these systems would coordination duties of event-based systems are
be unable to run properly without it [1]. A missed changed. Interrupts when schedule systems
deadline may have severe repercussions in switch jobs [22]. A number of commercial and
many real-time systems [2]. A number of non-commercial solutions are available [23].
industrial systems use the RTOS, including Embedded systems operating systems for the
process control, avionics, and nuclear power globe today. Each one has its unique set of
plants [3]. Embedded software is used in the features, abilities and advantages. However, they
majority of actual operating systems, which are all offer virtually the same service, in addition to
composed of hardware components that act as downsides in general. The user should have the
controllers for specific operations inside same ability in basic operations [24, 25].
mechanical or electrical systems [4, 5]. The
integrity of the results in Real - time data The remainder of the essay is organized as
Systems is dependent on both the analytical follows: In the Section2, the background theory,
conclusion of the computations and the time in the Section 3 the related work of this survey is
instant at which the conclusion is created [6, 7]. If analyzed. In section 4 discussion, and finally
at least one job in the system has defined timing the conclusion of this work presented in Section
restrictions, i.e., action must be done within a 5.
specified time period, a real time system is the
name given to the system [8]. Meeting a deadline 2. BACKGROUND AND THEORY
is the primary requirement of any real-time
system, and term "deadline" refers to the point in 2.1 Concepts of Scheduling Algorithms
time at which an action must be done. otherwise,
the RTS is said to have failed [9]. Three distinct A RTO (Real Time Operating System) is an OS
forms of Real Time-Systems exist: There are meant to serve in real time applications, often
three types of real-time systems: hard, firm, and without buffer delays, which process data as it
soft [10]. Hard Real Time System: If a deadline is enters [26]. Time requirements for processing
missed, the whole company would collapse; (including OS delays) are measured in 10
some of the deadlines are vital for safety, a good seconds or shorter increments [27]. A time-
example of this is an anti-missile system [11]. based, clearly defined, set timescales system are
System in Soft Real Time The software does not a reliable system [28]. Processing within the
fail if a deadline is missed; but the usefulness of stated restrictions must be carried out or the
the output decreases with time. Performance of system fails. Either you have an event or you
the system will deteriorate. Every interactive share time [29]. Systems driven by events switch
application is an illustration of a soft real time between activities on the basis of priority,
system [12]. Firm Real Time System (RTS): this whereas systems based on clock interruptions
is a system that is almost identical to soft RTS. switches time-sharing systems. Most RTOSs
RTOS has been used in a broad variety of utilize a preventive approach for scheduling [30,
embedded devices for years in automation and 31].
computer science [13]. Such systems have been
created to help regulate embedded algorithms on Software applications running on real-time
military equipment, systems and software for operating systems offer important time
large switching systems [14]. The use of RTOS is requirements for real-time systems [32, 33].
generally performed on hard real-time systems These software tasks must be organized in
[15, 16]. The RTOS has expanded dramatically accordance with software and hardware events.
in recent decades [17]. Applications of these In real-time operating systems certain services
systems are considered robust and always a for control of software tasks (priority-based pre-
designer challenge. These systems have to emption etc.) exist [34]. However, there is a need
ensure that time constraints are met while to plan algorithms in real-time systems in specific
executing complex activities [18]. A ROS is the scenarios [35]. This is particularly necessary
sort of operating system intended to operate in when time-critical software tasks need to be
real time. To manage programs capable of carried out throughout multiple working times and
managing data collected without buffering, a certain schedule [36]. To fulfil the system

36
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

requirements, different scheduling methods can yet definite fashion utilizing a large number of
be used [37]. For time critical jobs, the flight tiny and diverse sizes of memory allocation/de-
software running on the operating system in real allocation [52, 53].
time, particularly on satellite systems [38].
Real time systems are systems that rely on two
Embedded real-time systems are expanding elements. The first is the logical outcome of
progressively in order to run high-performance, computing and the second is the moment when
multi-core architectural applications [39]. Efficient the results are produced [54, 55]. A process to
task plan models in these systems are extremely finish its execution takes a certain period and the
important to ensure that most of the jobs may be system will fail if the time is over for the process.
planned within their deadline to deliver the Two real-time planning approaches exist: (1)
desired performance [40]. dynamic static (2). The priorities remain the
same for a job in the static algorithm
In the next Internet era, it is greatly wanted to and priorities are set at conception, with
reduce energy consumption of integrated dynamic priority being assigned in due course
computers [41]. One of the most efficient ways [56].
for both dynamic and static processor energy
usage is Minimum Energy Point Tracking [42]. In The Realtime systems are in which the accuracy
the past several years, earlier studies suggested of the output depends not only on the logical
a number of MEPT techniques. Although edge conclusions of the calculations, but on the time of
computing applications generally demand low the output [57]. This indicates that the results
energy tasks with real-time guarantees. The must be generated within the specified time
problem is the time complexity in which to period [58]. The time limit for the system to reply
recognize MEPs and change voltages, which is termed the time limit. Deadline fulfilment in any
frequently prohibits the planning of tasks in real real-time system is an essential characteristic
time [43]. [59]. The ETS enables to complete applications
in real time to fulfil their deadline utilizing the
Robot Operating System (ROS) supports scheduling mechanism [60]. The planning
insulating defects, quicker development and approach is the core of any real-time system that
modularity, and the core reusability, thereby makes decisions regarding the execution of tasks
providing a broad and de facto standard for on the system in order to prevent overlap of any
independent driving systems [44]. The GPUs sort [61].
also allow high-performance computation and
are thus utilized for autonomous driving [45]. As A major feature of the RTOS is the level of
real time processing requirements rise, coherence about its length of time required for an
techniques are being developed for meeting the application to be accepted and completed, the
real-time limit for ROS and GPUs [46, 47]. variability of which called 'jitter.' A 'hard' real-time
Regrettably, algorithms are not under OS (Hard RTOS) is less jittered than a 'soft' real-
investigation, which describe the ROS transport time OS (Soft RTOS) [62]. The late answer in a
(publish/subscribe) model that may be restricted hard RTOS is incorrect, but a late response in a
in execution, resulting to time waiting and a soft RTOS is okay [63]. The main objective for
reduction of the reaction of the system as a design is a soft or hard performance category
whole [48]. In addition, ROS GPU workloads are rather than a high output [64]. An RTOS that
also influenced by the ROS transport model typically or generally meet a time limit is a Soft
since the time is taken to set up the central Real-time OS, while it is a Difficult Real-time OS
processing unit (CPU) [49]. if it meets a time limit deterministically [65]. An
RTOS has a sophisticated programming
Effective time operating systems (RTOS) should algorithm [66]. Scheduler flexibility allows the
encourage the protocols of resource access to orchestration of process priorities by a larger,
limit the maximum delay of priority inversions. computer system, although a real-time OS is
Such protocols must be implemented lightly more often used to address a small number of
since their performance impacts planning [50, applications [67]. Key elements in a real-time OS
51]. are minimum interrupt latency and minimum
thread latency; with a real-time OS, it's more
An essential challenge in soft re-timing system appreciated how quickly or reliably it reacts than
designs is the handling of algorithms that handle how much work it can do in a particular time
a large number of comparable data in a dynamic period [68].

37
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

2.2 Components of RTOS  The Scheduler: This RTOS component

specifies the sequence in which tasks may
A real-time operating system (RTOS), which be done, which is often determined by their
controls system resources in a timely fashion and priority [74].
provides consistent foundation for the production  Symmetric Multiprocessing (SMP): is a
of application code, is described as a program collection of the numerous distinct jobs that
[69]. The RTOS must thus provide the temporal the RTOS may manage in order to do
planning of processes, hardware control and parallel processing [75].
access to RTOS resources. The RTOS kernel is  Function Library: It is a critical component
accessed through an interface known as the of RTOS since it serves as an interface
System Call API [70]. This is a set of RTOS between the kernel and application code.
services functions [71]. Atomic processing must This application allows you to communicate
be performed, which implies the CPU cannot be with the Kernel through a function library,
paused while the kernel resources are changed. guaranteeing that the application produces
Implementation of a system call API function the required results [76].
might be blocked or blocked [72]. The process  Memory Management: This component is
does not halt a non-blocking system call, but a essential to ensure that the system
blocking call suspends the system until a allocates memory to each program, which
requirement has been met. When a process is the RTOS's most crucial function [77].
requires Kernel services, it configures system  Fast dispatch latency: It is the average
call parameters and either conducts a time between a completion of the OS-
particular trap command or sets a kernel event to identified job and the beginning of
force context into kernel space [68]. The kernel processing by the thread in the ready
then checks the reasons for the caller process, queue [78].
carries out the requested action, and  User-defined data objects and classes:
returns control and operational outcome to the RTOS Systems make use of operating
user space process [73]. The following is a list of systems such as Windows or C++, which
the major components of real-time operating must be organized according to the
system purpose for which they are designed [79].

Fig. 1. The components of real time operating system

38
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

2.3 Propertise of Real Time Operating deadline, this is referred to be a hard

System deadline. This category includes systems
that conduct mission-critical tasks such as
The time it takes the OS to receive and execute air traffic control [89].
an application request is the unpredictability of
an actual operating system jitter [80]. Hard RTOS 2.5 Scheduling Criteria
systems have little jitter, while soft RTOS
systems are quite jittery. There are a few For CPU scheduling algorithms, several criteria
important OS features to consider when it comes have been presented; the characteristics used
to data processing speed and jitter [81]: for comparison can make a considerable
difference in selecting which technique is
 Reliability and stability: the operating considered the most effective. Some of the
system is unlikely to break or crash. requirements are below [90, 91]:
 Scalability: this is the operating system's
ability, if more features are introduced, to  CPU Usage: most of the time, the CPU
increase its performance [82]. would use the CPU the most instead of
 Availability: This is an opportunity for the wasting every cycle of the CPU (Ideally I00
OS to actively process its request and not percent of the time). Because of a true
crash when requested. system, CPU utilization should range
 Usability: this is the phase of the between 40% (without charge) and 90%
development of an operating system. (heavily loaded).
 Security: This phase is not subject to an  Performance: this indicates the whole
external attack by the operating system. number of processes per unit or the
complete amount of labor per unit. This
2.4 Scheduling Algorithms may range from l0/second to 1/hour
depending on the particular processes
Scheduling refers to the process of allocating [92].
available CPU resources to processes (s) [83].  Time to turn: the needed amount of time
This is a critical notion in the conception of for a certain procedure, i.e. the gap
multitasking, multiprocessing, and the real time between the time of presentation and
operating systems [84]. Scheduler and completion of the process [93].
dispatcher are used to do this. It is a decision-  Waiting time: total of waiting time for a
making process that involves allocating similar ready queue process waiting to be
resources across several activities and time checked by the CPU.
periods in order to accomplish many goals [85].  Load Average: the average number of
In a homogeneous/heterogeneous organization, processes waiting for CPU access [94].
resources and tasks may take on a variety of
shapes. The jobs have been prioritized; each 3. RELATED Work
assignment is assigned a due date and an early
completion date [86]. In real-time systems, the The following paragraphs summarize the work of
most crucial parameter is the deadline, which is different researchers in the topic of real-time
defined as the time where the results should be processor scheduler from 2018 to 2020.
supplied. As seen below, there are three unique
types of timeline [85, 87]. Teraiya and Shah [95], recommended that the
LST and SJF be implemented in a real-time
 Soft Deadline: is that if findings are still operating system that is not strictly real-time.
valuable after the deadline has past, this is Some algorithms were tested on a periodic work
referred to as a soft deadline. This set, and the following results were achieved.
subcategory includes reservation systems. They measured Success Ratio and Effective
 Firm deadline: This timeline is one that has CPU Utilization under the identical settings and
no usefulness if it is missed. It is OK to compared both algorithms. It is worth noting that
miss a few deadlines on sometimes. These the LST method works well under load but fails
deadlines are often employed in systems miserably under load. SJF is incapable of
that execute critical activities [88]. scheduling particular tasks even when the
 Hard deadline: If a disaster occurs as a system is under stress, yet it functions well when
consequence of failing to meet the the system is overloaded. On a large dataset,

39
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

practical tests were undertaken. Each task set scheduling, adding significant overhead and
has between one and nine operations. The data reducing task execution performance. The
set comprises 7500 task sets. Each process set algorithm's next areas of optimization will aim to
uses between 0.5 and 5 percent of the CPU. It minimize task transfer and context switching.
was validated against a 500-time unit to ensure
that both algorithms were accurate. Nasri et al. [99],The purpose of this study was to
demonstrate how the Non-preemptive
Capodieci et al. [96], showcased the rescheduling may be made more useful by using
development of a sample real-time scheduler for a first-in-first-out (FIFO) scheduling strategy
GPU functions on an embedded System on a combined with a unique reduction tuning
Chip (SoC) based on NVIDIA's cutting-edge process. This methodology allows the FIFO to
Architectures used in self-driving cars. The recreate a given viable scheduler, such as the
scheduling acts as a software segmentation layer one used by CW-EDF, resulting in a high degree
on top of the NVIDIA hypervisor, taking use of of schedule ability and relatively modest runtime
current-generation architectural features like as overheads. Memory overheads are also kept to a
pixel-level and thread-level preemption. We were minimum by using a modest number of offsets
able to construct and test a GPU activity per process. The runtime overhead, memory
scheduler that uses the Earliest Deadline First usage, and schedule ability ratio of the approach
(EDF) method. That provided Using a Constant are all examined using a preliminary result on an
Bandwidth Server, you may isolate your Arduino Uno board.
bandwidth. Using such an architecture (CBS).
Our research focused on alternative Abeni et al. [100], proposed a new hierarchical
programming paradigms for APIs computing, scheduler for Linux that is optimized for
which enabled to describe CPU and GPU container-based virtualization and can be utilized
submittal of a request with more detailed with LXC container that have multiple virtual
scheduling data. A full experimental CPUs. The provided scheduler is constructed by
characterization is offered to demonstrate the changing the real-time control group’s
considerable increase in scheduling of recurrent mechanism in such a way that the SCHED
real time GPU operations. DEADLINE policy is utilized to schedule each
group's real-time run queues. Experiments
Yang et al. [97], a hierarchy scheduling approach demonstrate that a real-time application
for DAG jobs with bounded deadlines was scheduled inside an LXC container using the
presented. Hierarchical scheduling separates the new scheduler operates in the manner expected
scheduling of computing resources from the by previous theoretical CSF analysis. Our control
scheduling of workloads, resulting in a groups scheduler proved to be simpler to setup
significantly more effective resource sharing and produced superior outcomes, while also
solution without regard for task kinds. We having the ability to use less real-time
demonstrate the feasibility of our suggested computing capacity inside the system, so
hierarchical scheduling on a practical platform reducing runs.
with an acceptable runtime overhead.
Additionally, we conduct extensive tests to Baital and Chakrabarti [101], proposed an
determine the schedule ability of our hierarchical improved scheduling algorithm in which random
scheduling system, and the findings indicate that tasks with varying periodicity and execution time
our suggested technique performs admirably. are generated at different time intervals. Energy
consumption is a critical design consideration in
Cao and Bian [98], using a homogeneous real-time systems, particularly battery-powered
multicore CPU, we suggested a DAG job systems. We extended our scheduling work to
scheduling technique. This is a refinement of the heterogeneous multicore systems (HMS)
stretching method. The original method is architectures, in which real-time tasks are
enhanced by adding the ability to pick subtasks distributed to the appropriate cores while still
for conversion and by improving the computation meeting the task deadline. They proved that the
of release time and deadline. This method has a heterogeneous multicore scheduler paradigm
substantially greater success rate than the may be utilized to create commercially
original approach in terms of scheduling. accessible heterogeneous multicore processors.
Nevertheless, owing to the segmentation of They validated the model using generated task
tasks, a high number of task migrations and sets and discovered that our model performs
context switches occur throughout task exceptionally well in all cases and significantly

40
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

reduces the system's energy consumption when Chen et al. [105], The results reported here
compared to some popular and novel scheduling establish the existence of Preemptive, fixed-
techniques. priority real time systems now have a new
scheduling side-channel (RTS). Systems that are
Chen et al. [102], develop a lightweight real time examples of this kind include automobile
scheduling method that is based on job systems, aviation system, electricity generation
duplication, RTSATD, with objective of lowering plants and industrial control systems. Notably,
both the time required to complete and the capturing this timing information is difficult
financial expense associated with cloud-based because to schedule runtime changes, the
big-data workflow processing. RTSATD's system's incorporation of several interrelated
performance is assessed using simulated and activities, and the normal limitations (e.g
real-world processes. Testing findings indicate deadline) inherent in the development of RTS.
that the suggested method beats two current Schedule-Leak methods reveal how to exploit
algorithms in terms of project length (by up to this side-channel successfully. On genuine
28.73 %) and resource consumption (up to 46.31 operating systems, a comprehensive
percent). implementation is shown (in Real-time-Linux and
Free RTOS). Schedule Leak's facts about the
Dauphin et al. [103], On parallel, heterogeneous, time period may considerably help other, more
Non-Uniform Memory Architecture (NUMA) sophisticated attackers in fulfilling their
systems, a hybrid technique for scheduling and objectives.
memory management of periodical dataflow apps
was proven. The task ordering and memory Malik et al. [106], suggested a novel For the
allocation for each Processing Element are proper execution and administration of real-time
distributed and computed concurrently in Odyn. hard and soft activities in embedding IoT
Additionally, He provides a strategy for devices, an adaptable and intelligent scheduling
preventing deadlocks induced by efforts to create mechanism is required. The reposed schedule
buffers in memory that are larger than the method prioritizes the performance of critical
available space. This method reduces the real-time tasks with a high degree of priority
requirement for backtracking, which is a feature activities above the distribution of CPU resources
of dynamic scheduling algorithms. It is based on to potentially hungry, in overloaded instances,
static calculation of exclusion relations among soft real-time processes. This was accomplished
buffers in an application. They demonstrate via the use of two astute deletions: Urgency
Odyn's efficacy on a test bench that simulates Measure (UM) and Failure Measure (FM). By
the interactions of concurrent applications leveraging the available CPU unit for optimal
generated randomly. Additionally, they CPU usage and rapid reaction times, the
demonstrate their technique for avoiding suggested methodology decreases the rate of
deadlocks through a variety of use cases. tasks missed and jobs starved. The findings
demonstrate that the suggested approach
Riasetiawan and Ashari [104], proposed a outperforms the other techniques in terms of task
schedule for the functioning of multiple landslide starvation rate reduction and CPU usage
sensors Separately, data processing is increase.
transferred to an IoT device using FIFO and
Round Robin scheduling. Only the performance Doan and Tanaka [107], suggested a novel way
of FIFO and Round Robin algorithms in of scheduling that is adaptable and intelligent for
scheduling incoming processes in real time on an the effective execution and administration of real-
IoT OS is discussed by the authors. Taking into time hard and soft tasks in embedded IoT
account the waiting time and response time. The devices. The reposed scheduling method
analysis is expected to result in a short response prioritizes the execution of high-priority hard-real-
time and waiting time, allowing for the selection time activities preceding the distribution of CPU
of an appropriate algorithm to complement the resources to potentially hungry, in overloaded
IoT architecture for landslide detection. The FIFO instances, soft real-time processes. This was
and Round Robin algorithms are implemented in accomplished via the use of two intelligent
the Raspberry Pi 3 Model B IoT device via erasures: Urgency-Measure (UM) and Failure-
Raspbian and Arches Ubuntu are two different Measure (FM). By leveraging the available CPU
operating systems. A 64-bit 64-bit ARM Cortex- unit for optimal CPU usage and rapid reaction
A53 64-bit processor operating at 1.2GHz time, the suggested methodology decreases the
powers the Raspberry Pi 3 Model B. rate of tasks missed and jobs starved. The

41
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

findings demonstrate that the suggested constraints associated with flying time-
approach outperforms the other techniques in constrained UAVs.
terms of task starvation rate reduction and CPU
usage increase. Chen et al. [111], has removed a significant
constraint associated with newly established
D'souza and Rajkumar [108], presented the dependency graph approaches, namely just one
Cycle Tandem static frequency-scaling crucial section per task (DGA). Under terms of
methodology for co-optimization of the CPU and computational complexity, shown that even in
hardware accelerators operating frequencies. extremely limited settings, the multiprocessor
Consider different energy management situations synchronization issue is NP-complete. Suggest a
where the accelerator or CPU frequency may or Provide the approximation ratio(s) for the
may not be configurable, and present the Cycle resulting computation time based on a
Solo family of algorithms for such situations systematic proposed design based on the DGA
based on practical considerations of real-world by leveraging current algorithms designed for job
platforms. Additionally, when multi-core shop scheduling. The results of the evaluation in
processors are utilized in combination with Section 6.2 show that the methodology is quite
hardware accelerators, partitioning solutions effective for real-time task systems based on
should be investigated to lower operation frames.
frequency. Experiments suggest that some
strategies may result in large energy savings. Utkarsh et al. [112], a completely distributed
Additionally, we offer a case study using the model-predictive and computational intelligence
NVIDIA TX2 embedded platform to demonstrate program that enables micro grid devices to
the energy reductions that our suggested function autonomously with little communication
solutions may achieve. exchange, hence eliminating the need for a
central controller. The proposed distributed
Nguyen et al. [109], suggested Real time algorithm's convergence features are evaluated
scheduling of Household Appliances' Quality and compared to those of a state-of-the-art
learning, a well-known value iterative distributed algorithm, and numerical simulations
reinforcement learning methodology, is used to for various situations are conducted to
calculate operational time (RSOTHA-QL). The demonstrate that the proposed distributed
RSOTHA-QL procedure is a two-step procedure. approach may be implemented to real-world
The first step involves Q learning agents micro grids.
engaging with the smart home system with the
purpose of receiving a reward. Additionally, the Lee et al. [113], It is advised that a real time
incentive value is used to arrange the operating schedule study of transiently powered CPUs with
duration of home App in next step, ensure that a NVMs be performed. To begin, quantitatively
energy consumption is kept to a minimum. The analyze the energy harvester's charging and
second phase addresses discontent caused by discharging behaviors and extract the system's
the scheduling of the operating time of the home compute capabilities in the time interval domain.
user's domestic appliances by dividing them in to Then, using real time calculus, we establish
Three categories: 1) deferrable, 2) non- whether or not a particular multi-task workload
deferrable, and3) controlled. It is discovered that can be scheduled using the earliest possible
the operating period of household appliances is deadline first (EDF) or fixed-priority (FP)
effectively planned in order to considerably scheduling rules. Furthermore, the work
minimize energy usage and user discontent. examined how the threshold voltage parameter
selection influences schedulability and then
Khan et al. [110], demonstrated the efficacy of developed a practical threshold selection
our suggested DRL algorithm, for scheduling technique to improve schedulable.
energy harvesting times during UAV-assisted Comprehensive simulations are used to validate
,D2D communication. The terms of EE and the suggested technique's efficacy. In
complexity, offered approaches outperform comparison to the naïve selection approach, the
benchmarks. The energy harvesting time strategy consistently improves schedule ability
scheduling game may be resolving extremely across a range of workloads.
fast by exploiting the benefits of deep learning.
The findings indicate that the DRL method may El Ghor et al. [114], The challenge of real-time
be a viable solution for a real time app, given the scheduling of many processors at the systems
energy storage constraints and the times with an It was investigated if an energy reservoir

42
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

Table 1. Summary of literature review related to Scheduling Algorithms for Real Time Operating Systems

No. Ref. RTOS Type Operating Significant Results Future Work Suggestions
and System
Year
1 [95] Soft RTOS Real time Data Set Includes 7500 task sets. Each procedure is an individual job. In the coming, we may suggest a new method that incorporates LST and
2018 operating systemprocessor load SJF. Load and demand will make it function well.
2 [96] Soft RTOS embedded For quickly executing aperiodic jobs, we give them a null, or small, In the near, we hope to introduce more complex reclamation techniques
2018 System deadline. This allows the aperiodic jobs to preempt already executing jobs. for GPUs' excess bandwidth, which will make more flexible selections of
GPU task budgets possible.
3 [97] Hard A Hierarchical Our hierarchical scheduling implementation was assessed and findings Future directions for this study abound. Let's first look into scheduling
2018 RTOS Scheduling showed it to be promising. many DAG jobs on several virtual platforms under various platforms.
Approach Also, DAG task scheduling models with resource sharing is a worthwhile
consideration. Secondly, we want to use power scheduling techniques for
DAG.
4 [98] Hard Multi-core This method's scheduling success rate is much higher than the original Next up are tasks and context switching for the algorithm.
2020 RTOS Processors approach.
5 [99] Hard Arduino board With respect to code size, FIFO-OT scheduler footprints are likewise less The aggregate minimum amount of offsets for all jobs and extended
2018 RTOS than OE's. systems with release jitter.
6 [100] Soft RTOS Linux Kernel Experimental findings reveal that real-time LXC-based scheduling In the upcoming, we want to experiment with the suggested scheduler in
2019 performed under the new scheduler behaved as expected by current CSF a multi-CPU container-based simultaneous real-time activity setting.
analyses.
7 [101] Hard Multicore systems Our approach works very well in all instances, and greatly decreases The additional feature will help implement dependent task sets.
2019 RTOS energy usage when compared to certain popular and novel schedules.
8 [102] Hard Geo-Distributed The testing findings show that the suggested algorithm offers 28.73% Therefore, fault-tolerant huge data processing scheduling should be one
2020 RTOS Clouds more completion time reduction and 46.31% greater resource usage than of the noteworthy approaches to pursue.
the other alternatives.
9 [103] RTOS Dataflow Authors show the usefulness of Odyn on a test bench that replicates A suggests research project is to calculate the appropriate memory
2019 Applications random program interactions We illustrate the stalemate avoidance assignments for PEs, depending on their performance requirements.
strategy in several scenarios.
12 [106] Hard and IoT Devices The study found that the suggested approach outperformed the other One of the drawbacks of the suggested solution is that it requires time to
2019 Soft Real Embedded techniques, considerably lowering task starvation and boosting CPU collect sufficient training data.
Time Hybrid usage.
13 [107] Soft RTOS Real-time The outcomes from our simulation demonstrate that our technique has In addition, using a hardware accelerator may minimize the runtime
2018 embedded decreased time complexity while still maintaining schedulability, task overhead of the method. This inquiry is in the realm of practice and
systems. preemption, and task migration. segmented near-optimal techniques.

43
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

No. Ref. RTOS Type Operating Significant Results Future Work Suggestions
and System
Year
14 [108] Hard Cyber-physical Lastly, we implement the Cycle Solo and Cycle Tandem algorithms to full- Multi-core systems are where the rate of each processor core may be
2018 RTOS systems partitioned multi-core CPUs, where all CPU cores should be tuned to the separately adjusted in the next. Systems which feature numerous
same rate. accelerators may also be considered.
15 [109] Soft RTOS neural networks The findings show that the DRL method may be applicable for real-time In the approach, we will handle more complex challenges by
2020 applications even if energy storage and flight duration are limited. simultaneously optimizing power allocation, trajectory planning, and
numerous UAV scenarios.
16 [110] Hard Smart Home The suggested scheduling technique improves on the LST based Future research will use deep Q knowledge and artificial neural networks
2018 RTOS Appliances scheduling when it comes to decreasing energy use and lowering to simulate a complex multiple-home user situation.
unhappiness for the house us.
17 [111] Soft RTOS Multicore systems Using created task sets, the model was verified and shown to perform Our next projects will be focused on validating the scheduling model,
2020 brilliantly and dramatically lower energy usage in comparison to a few ensuring it is fault tolerant and reliability conscious.
popular and novel scheduling approaches.
18 [112] Soft RTOS Network of Smart in the simulation game, there are huge benefits to reactive power trading, Not mentioned
2018 Micro grids including dynamic price adjustments between micro grids.
19 [113] Hard Wireless Sensor This article examines the schedulability of transiently supplied CPUs with Additionally, minimizing the threshold voltage searching overhead without
2020 RTOS Network NVMs in real time. Quantified analysing the behaviour of the energy approximation remains a future studies objective.
harvester across time intervals permits real-time multi-task execution time
analysis using the RTC.
20 [114] Hard Ambient energy Our method delivers substantial performance improvement when The next study will concentrate on an evaluation of the performance of
2018 RTOS compared to EDF. EH-RA comparison to EDF, which uses the bin-packing methodology.

44
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

could be recharged by an external source.. We interval or exact instant in time. This

are particularly interested in energy-efficient characteristic is very important in many
partitioning for periodic real-time applications on applications where any delay in implementing the
a homogenous multi-core architecture, taking program or any delay in the reaction time results
into account both timing and energy needs. in adverse effects for the system, such as
Assume that the optimum scheduler is employed automated control systems in industrial
on each core of the design, namely the Earliest equipment and different transportation modes. In
Deadline - Harvesting (ED-H). The purpose is to this review paper about real time operating
provide realistic to achieve the intended absence system a total of 20 papers we used in a related
of both energy famine and deadline violation, a works from years 2018 to 2020 based of the
segmentation method based on the real-time findings the number of the papers about RTOS
execution of tasks given to sensor nodes utilizing increased from 2020. Most of the articles about
actual energy recovery data was developed. To the hard RTOS type, and many of the related
accomplish this, they presented an Energy work used in embedded systems.
Harvesting Reasonable allocation (EH-RA)
method that is equivalent to the conventional bin- DISCLAIMER
packing methodology in terms of both temporal
limitations and energy awareness. Experimental The products used for this research are
data indicate that our technique can significantly commonly and predominantly use products in our
outperform EDF in terms of performance. area of research and country. There is absolutely
no conflict of interest between the authors and
4. DISCUSSION producers of the products because we do not
intend to use these products as an avenue for
Additionally, it is necessary to emphasize that in any litigation but for the advancement of
some real-world applications, end-to-end time knowledge. Also, the research was not funded by
limitations apply to calculations that span many the producing company rather it was funded by
processing locations. We will present some of personal efforts of the authors.
the most prominent programming algorithms
used in programming for the CPU. They are not
all suited for usage in embedded systems in real- COMPETING INTERESTS
time. Currently, non-preventative planning, round
timing and preemptive priority planning are the Authors have declared that no competing
most often used Algorithms in real RTOS. The interests exist.
findings reveal that an overview of the linked
research is provided in the table below. The table REFERENCES
gives the reference name and the corresponding
year of the investigations (2018, 2019 and 2020), 1. Sadeeq MA, Zeebaree S. Energy
the kind of RTOS system, the system used by management for internet of things via
them and the results obtained from research and distributed systems. Journal of Applied
recommendations or future study where these Science and Technology Trends.
are conducted. 2021;2:59-71.
2. Hambarde P, Varma R, Jha S. The survey
5. CONCLUSION of real time operating system: RTOS. 2014
International Conference on Electronic
Real-time operating system enables real-time Systems, Signal Processing and
apps achieve their deadline through scheduling, Computing Technologies. 2014;34-39.
the scheduling approach is it the beating hearts 3. Abdullah RM, Ameen SY, Ahmed DM, Kak
of every Real-Time System, which controls every SF, Yasin HM, Ibrahim IM, et al.
job execution to prevent overlap. Real-time Paralinguistic Speech Processing: An
operating systems are fundamental to the Overview. Asian Journal of Research in
functioning of numerous technologies and Computer Science. 2021;34-46.
technologies that are crucial to our everyday 4. Ibrahim IM, Ameen SY, Yasin HM, Omar
lives. Additionally, these systems have the N, Kak SF, Rashid ZN, et al. Web server
capacity to execute the software necessary and performance improvement using dynamic
to give temporal precision, which means that load balancing techniques: A review. Asian
they are able to assure that the termination of a Journal of Research in Computer Science.
program or task execution is related to a time 2021;47-62.

45
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

5. Omer MA, Zeebaree SR, Sadeeq MA, Machine learning-based diabetic

Salim BW, Mohsin Sx, Rashid ZN, et al. retinopathy early detection and
Efficiency of malware detection in android classification systems-A survey. 2021 1st
system: A survey. Asian Journal of Babylon International Conference on
Research in Computer Science. 2021;59- Information Technology and Science
69. (BICITS). 2021;16-21.
6. Donga J, Holia M. An analysis of 16. Sadeeq MM, Abdulkareem NM, Zeebaree
scheduling algorithms in real-time SR, Ahmed DM, Sami AS, Zebari RR. IoT
operating system. International Conference and Cloud computing issues, challenges
on Inventive Computation Technologies. and opportunities: A review. Qubahan
2019;374-381. Academic Journal. 2021;1:1-7.
7. Maulud DH, Zeebaree SR, Jacksi K, 17. Yahia HS, Zeebaree SR, Sadeeq MA,
Sadeeq MAM, Sharif KH. State of art for Salim NO, Kak SF, Adel AZ, et al.
semantic analysis of natural language Comprehensive survey for cloud
processing. Qubahan Academic Journal. computing based nature-inspired
2021;1:21-28. algorithms optimization scheduling. Asian
8. Ahmed DM, Ameen SY, Omar N, Kak SF, Journal of Research in Computer Science.
Rashid ZN, Yasin HM, et al. A state of art 2021;1-16.
for survey of combined iris and fingerprint 18. Turci LdO. Real-time operating system
recognition systems. Asian Journal of freertos application for fire alarm project in
Research in Computer Science. 2021;18- reduced scale. International Journal of
33. Computing and Digital Systems.
9. Brucker P, Heitmann S, Hurink J, Nieberg 2017;6:197-204.
T. Job-shop scheduling with limited 19. Hasan DA, Hussan BK, Zeebaree SR,
capacity buffers. OR spectrum. Ahmed DM, Kareem OS, Sadeeq MA. The
2006;28:151-176. impact of test case generation methods on
10. Ibrahim BR, Khalifa FM, Zeebaree SR, the software performance: A review.
Othman NA, Alkhayyat A, Zebari RR, et al. International Journal of Science and
Embedded system for eye blink detection Business. 2021;5:33-44.
using machine learning technique. 2021 20. Ageed ZS, Zeebaree SR, Sadeeq MM,
1st Babylon International Conference on Kak SF, Rashid ZN, Salih AA, et al. A
Information Technology and Science survey of data mining implementation in
(BICITS). 2021;58-62. smart city applications. Qubahan
11. Maulud DH, Ameen SY, Omar N, Kak SF, Academic Journal. 2021;1:91-99.
Rashid ZN, Yasin HM, et al. Review on 21. Jacksi K, Ibrahim RK, Zeebaree SR,
natural language processing based on Zebari RR, Sadeeq MA. Clustering
different techniques. Asian Journal of documents based on semantic similarity
Research in Computer Science. 2021;1- using HAC and K-mean algorithms. 2020
17. International Conference on Advanced
12. Salih AA, Ameen SY, Zeebaree SR, Science and Engineering (ICOASE).
Sadeeq MA, Kak SF, Omar N, et al. Deep 2020;205-210.
learning approaches for intrusion 22. Ageed ZS, Zeebaree SR, Sadeeq MA,
detection. Asian Journal of Research in Abdulrazzaq MB, Salim BW, Salih AA, et
Computer Science. 2021;50-64. al. A state of art survey for intelligent
13. Hassan RJ, Zeebaree SR, Ameen SY, Kak energy monitoring systems," Asian Journal
SF, Sadeeq MA, Ageed ZS, et al. State of of Research in Computer Science.
art survey for iot effects on smart city 2021;46-61.
technology: Challenges, opportunities, and 23. Jijo BT, Zeebaree SR, Zebari RR, Sadeeq
solutions. Asian Journal of Research in MA, Sallow AB, Mohsin S, et al. A
Computer Science. 2021;32-48. comprehensive survey of 5G mm-wave
14. Zebari S, Yaseen NO. Effects of parallel technology design challenges. Asian
processing implementation on balanced Journal of Research in Computer Science.
load-division depending on distributed 2021;1-20.
memory systems. J. Univ. Anbar Pure Sci. 24. Arsinte R. Real time operating system
2011;5:50-56. options in connected embedded equipment
15. Hasan DA, Zeebaree SR, Sadeeq MA, for distributed data acquisition. Carpathian
Shukur HM, Zebari RR, Alkhayyat AH.

46
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

Journal of Electronic and Computer tracker. SMS reminder system: Design and
Engineering. 2018;11:35-58. implementation; 2020.
25. Sadeeq MA, Abdulazeez AM. Neural 36. Malallah H, Zeebaree SR, Zebari RR,
networks architectures design, and Sadeeq MA, Ageed ZS, Ibrahim IM, et al.
applications: A review. 2020 International A comprehensive study of kernel (issues
Conference on Advanced Science and and concepts) in different operating
Engineering (ICOASE). 2020;199-204. systems. Asian Journal of Research in
26. Ageed ZS, Ibrahim RK, Sadeeq M. Unified Computer Science. 2021;16-31.
ontology implementation of cloud 37. Sadeeq MA, Zeebaree SR, Qashi R,
computing for distributed systems. Current Ahmed SH, Jacksi K. Internet of things
Journal of Applied Science and security: A survey. in 2018 International
Technology. 2020;82-97. Conference on Advanced Science and
27. Zeebaree S, Ameen S, Sadeeq M. Social Engineering (ICOASE). 2018;162-166.
media networks security threats, risks and 38. Ismael HR, Ameen SY, Kak SF, Yasin HM,
recommendation: A case study in the Ibrahim IM, Ahmed AM, et al. Reliable
kurdistan region. International Journal of communications for vehicular networks.
Innovation, Creativity and Change. Asian Journal of Research in Computer
2020;13:349-365. Science. 2021;33-49.
28. Sulaiman MA, Sadeeq M, Abdulraheem 39. Yasin HM, Zeebaree SR, Sadeeq MA,
AS, Abdulla AI. Analyzation study for Ameen SY, Ibrahim IM, Zebari RR, et al.
gamification examination fields. Technol. IoT and ICT based smart water
Rep. Kansai Univ. 2020;62:2319-2328. management, monitoring and controlling
29. Sadeeq M, Abdulla AI, Abdulraheem AS, system: A review. Asian Journal of
Ageed ZS. Impact of electronic commerce Research in Computer Science. 2021;42-
on enterprise business. Technol. Rep. 56.
Kansai Univ. 2020;62:2365-2378. 40. Abdulla AI, Abdulraheem AS, Salih AA,
30. Salih A, Zeebaree ST, Ameen S, Alkhyyat Sadeeq M, Ahmed AJ, Ferzor BM, et al.
A, Shukur HM. A survey on the role of Internet of things and smart home security.
artificial intelligence, machine learning and Technol. Rep. Kansai Univ. 2020;62:2465-
deep learning for cybersecurity attack 2476.
detection. 2021 7th International 41. Abdulraheem AS, Salih AA, Abdulla AI,
Engineering Conference “Research & Sadeeq M, Salim N, Abdullah H, et al.
Innovation amid Global Pandemic"(IEC). Home automation system based on IoT;
2021;61-66. 2020.
31. Alzakholi O, Shukur H, Zebari R, Abas S, 42. Salih AA, Zeebaree S, Abdulraheem AS,
Sadeeq M. Comparison among cloud Zebari RR, Sadeeq M, Ahmed OM.
technologies and cloud performance. Evolution of mobile wireless
Journal of Applied Science and communication to 5G revolution.
Technology Trends. 2020;1:40-47. Technology Reports of Kansai University.
32. Abdullah DM, Ameen SY, Omar N, Salih 2020;62:2139-2151.
AA, Ahmed DM, Kak SF, et al. Secure data 43. Dino HI, Zeebaree S, Salih AA, Zebari RR,
transfer over internet using image Ageed ZS, Shukur HM, et al. Impact of
steganography. Asian Journal of Research process execution and physical memory-
in Computer Science. 2021;33-52. spaces on OS performance. Technology
33. Ageed Z, Mahmood MR, Sadeeq M, Reports of Kansai University.
Abdulrazzaq MB, Dino H. Cloud computing 2020;62:2391-2401.
resources impacts on heavy-load parallel 44. Hamdi SJ, Ibrahim IM, Omar N, Ahmed
processing approaches. IOSR Journal of OM, Rashid ZN, Ahmed AM, et al. A
Computer Engineering (IOSR-JCE). comprehensive study of malware detection
2020;22:30-41. in android operating systems.
34. Kareem FQ, Ameen SY, Salih AA, Ahmed 45. Ibrahim IM. Task scheduling algorithms in
DM, Kak SF, Yasin HM, et al. SQL cloud computing: A review. Turkish Journal
injection attacks prevention system of Computer and Mathematics Education
technology. Asian Journal of Research in (TURCOMAT). 2021;12:1041-1053.
Computer Science. 2021;13-32. 46. Ageed ZS, Ahmed AM, Omar N, Kak SF,
35. Sallow A, Zeebaree S, Zebari R, Mahmood Ibrahim IM, Yasin HM, et al. A state of art
M, Abdulrazzaq M, Sadeeq M. Vaccine survey of nano technology:

47
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

Implementation, Challenges, and future 57. Zeebaree A, Adel A, Jacksi K, Selamat A.

trends. Designing an ontology of E-learning
47. Abdulazeez AM, Zeebaree SR, Sadeeq system for duhok polytechnic university
MA. Design and implementation of using protégé OWL tool. J Adv Res Dyn
electronic student affairs system. Control Syst. 2019;11:24-37.
Academic Journal of Nawroz University. 58. Jader OH, Zeebaree S, Zebari RR. A state
2018;7:66-73. of art survey for web server performance
48. Abdulqadir MM, Salih AA, Ahmed OM, measurement and load balancing
Hasan DA, Haji LM, Ahmed SH, et al. A mechanisms. International Journal of
comprehensive study of caching effects on Scientific & Technology Research.
fog computing performance. 2019;8:535-543.
49. Yazdeen AA, Zeebaree SR, Sadeeq MM, 59. Adel AZ, Zebari S, Jacksi K. Football
Kak SF, Ahmed OM, Zebari RR. FPGA ontology construction using oriented
implementations for data encryption and programming. Journal of Applied Science
decryption via concurrent and parallel and Technology Trends. 2020;1:24-30.
computation: A review. Qubahan 60. Zeebaree S, Zebari RR, Jacksi K, Hasan
Academic Journal. 2021;1:8-16. DA. Security approaches for integrated
50. Ageed ZS, Zeebaree SR, Sadeeq MM, enterprise systems performance: A
Kak SF, Yahia HS, Mahmood MR, et al. Review. Int. J. Sci. Technol. Res. 2019;8.
Comprehensive survey of big data mining 61. Selamat SAAZA. Electronic learning
approaches in cloud systems. Qubahan management system based on semantic
Academic Journal. 2021;1:29-38. web technology: A review. Int. J. Adv.
51. Zeebaree S, Zebari RR, Jacksi K. Electron. Comput. Sci. 2017;4:1-6.
Performance analysis of IIS10. 0 and 62. Abdullah RM, Abdulazeez AM, Al-Zebari A.
Apache2 Cluster-based Web Servers Machine learning algorithm of intrusion
under SYN DDoS Attack. TEST detection system. Asian Journal of
Engineering & Management. Research in Computer Science. 2021;1-
2020;83:5854-5863. 12.
52. Abdulrahman LM, Zeebaree SR, Kak SF, 63. Zebari IM, Zeebaree SR, Yasin HM. Real
Sadeeq MA, Adel AZ, BW. Salim, et al. A time video streaming from multi-source
state of art for smart gateways issues and using client-server for video distribution," in
modification. Asian Journal of Research in 2019 4th Scientific International
Computer Science. 2021;1-13. Conference Najaf (SICN). 2019;109-
53. Sallow AB, Sadeeq M, Zebari RR, 114.
Abdulrazzaq MB, Mahmood MR, Shukur 64. Shukur H, Zeebaree SR, Ahmed AJ,
HM, et al. An investigation for mobile Zebari RR, Ahmed O, Tahir BSA, et al. A
malware behavioral and detection state of art survey for concurrent
techniques based on android platform. computation and clustering of parallel
IOSR Journal of Computer Engineering computing for distributed systems. Journal
(IOSR-JCE). 2020;22:14-20. of Applied Science and Technology
54. Abdulqadir HR, Zeebaree SR, Shukur HM, Trends. 2020;1:148-154.
Sadeeq MM, Salim BW, Salih AA, et al. A 65. Ibrahim BR, Zeebaree SR, Hussan BK.
study of moving from cloud computing to Performance measurement for distributed
fog computing. Qubahan Academic systems using 2TA and 3TA based on
Journal. 2021;1:60-70. OPNET principles. Science Journal of
55. Dino H, Abdulrazzaq MB, Zeebaree S, University of Zakho. 2019;7:65-69.
Sallow AB, Zebari RR, Shukur HM, et al. 66. Tahir B, Ali Saktioto J, Fadhali M, Rahman
Facial expression recognition based on R, Ahmed A. A study of FBG sensor and
hybrid feature extraction techniques with electrical strain gauge for strain
different classifiers. TEST Engineering & measurements. Journal of optoelectronics
Management. 2020;83:22319-22329. and advanced materials. 2008;10:2564-
56. AL-Zebari A, Zeebaree S, Jacksi K, 2568.
Selamat A. ELMS–DPU ontology 67. Zeebaree SR, Sallow AB, Hussan BK, Ali
visualization with Protégé VOWL and Web SM. Design and simulation of high-speed
VOWL. Journal of Advanced Research in parallel/sequential simplified DES code
Dynamic and Control Systems. breaking based on FPGA. 2019
2019;11:478-85. International Conference on Advanced

48
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

Science and Engineering (ICOASE). Systems and Software Engineering.

2019;76-81. 2013;9:29-43.
68. Harki N, Ahmed A, Haji L. CPU scheduling 80. Salim NO, Abdulazeez AM. Human
techniques: A review on novel approaches diseases detection based on machine
strategy and performance assessment. learning algorithms: A review. International
Journal of Applied Science and Journal of Science and Business.
Technology Trends. 2020;1:48-55. 2021;5:102-113.
69. Ahmed A, Ahmed O. Correlation pattern 81. Khan S. Real-Time operating system
among morphological and biochemical (RTOS) with Different application: A
traits in relation to tillering capacity in systematic mapping. European Journal of
sugarcane (Saccharum Spp). Acad J Plant Engineering and Technology Research.
Sci. 2012;5:119-122. 2021;6:100-103.
70. Zebari DA, Haron H, Zeebaree SR, 82. Yasin HM, Zeebaree SR, Zebari IM.
Zeebaree DQ. Multi-Level of DNA Arduino based automatic irrigation system:
Encryption Technique Based on DNA Monitoring and SMS controlling. 2019 4th
Arithmetic and Biological Operations. 2018 Scientific International Conference Najaf
International Conference on Advanced (SICN). 2019;109-114.
Science and Engineering (ICOASE). 83. Zeebaree S, Yasin HM. Arduino based
2018;312-317. remote controlling for home: Power saving,
71. Ahmed AJ, Mohammed FH, Majedkan NA. security and protection. International
An evaluation study of an E-learning Journal of Scientific & Engineering
course at the Duhok Polytechnic Research. 2014;5:266-272.
University: A case study. Journal of Cases 84. Salim NO, Zeebaree SR, Sadeeq MA,
on Information Technology (JCIT). Radie A, Shukur HM, Rashid ZN. Study for
2022;24:1-11. food recognition system using deep
72. Ahmed O, Geraldes R, Ahmed A, DeLuca learning. Journal of Physics: Conference
G, Palace J. Multiple sclerosis and the risk Series. 2021;012014.
of venous thrombosis: a systematic review. 85. Khera I, Kakkar A. Comparative study of
MUltiple Sclerosis Journal. 2017;757-758. scheduling algorithms for real time
73. Juven V. Lightweight Event-driven real- environment. International Journal of
time operating system for resource Computer Applications. 2012;44:5-8.
constrained connectivity; 2017. 86. Salim NO, Abdulazeez AM. Science and
74. Gracioli G, Fröhlich AA, Pellizzoni R, business. International Journal. 5:102-113.
Fischmeister S. Implementation and 87. Du D. Scheduling Algorithms: JSTOR;
evaluation of global and partitioned 2008.
scheduling in a real-time OS. Real-Time 88. Eesa AS, Sadiq S, Hassan M, Orman Z.
Systems. 2013;49:669-714. Rule generation based on modified
75. Holman P, Anderson JH. Adapting Pfair cuttlefish algorithm for intrusion detection
scheduling for symmetric multiprocessors. system. Uludağ University Journal of The
Journal of Embedded Computing. Faculty of Engineering. 2021;26:253-268.
2005;1:543-564. 89. Eesa AS. Optimization algorithms for
76. Chapin SJ, Spafford EH. Support for intrusion detection system: A review.
implementing scheduling algorithms using International Journal of Research-
MESSIAHS. Scientific Programming. GRANTHAALAYAH. 2020;8:217-225.
1994;3:325-340. 90. Mahmud N, Afrin S, Rahman F,
77. Kato S, Ishikawa Y, Rajkumar RR. CPU Monirujjaman M. An application based
scheduling and memory management for improved round robin CPU scheduling for
interactive real-time applications. Real- real time operating system. Brac
Time Systems. 2011;47:454-488. University; 2017.
78. Yousefi H, Malekimajd M, Ashouri M, 91. Haji SH, Zeebaree SR, Saeed RH, Ameen
Movaghar A. Fast aggregation scheduling SY, Shukur HM, Omar N, et al.
in wireless sensor networks. IEEE Comparison of software defined
Transactions on Wireless networking with traditional networking.
Communications. 2015;14:3402-3414. Asian Journal of Research in Computer
79. Bjørk J, de Boer FS, Johnsen EB, Schlatte Science. 2021;1-18.
R, Tarifa SLT. User-defined schedulers for 92. Zeebaree S, Zebari I. Multilevel
real-time concurrent objects. Innovations in client/server peer-to-peer video

49
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

broadcasting system. International Journal 103. Dauphin B, Pacalet R, Enrici A, Apvrille L.

of Scientific & Engineering Research. Odyn: Deadlock Prevention and Hybrid
2014;5:260-265. Scheduling Algorithm for Real-Time
93. Kareem FQ, Zeebaree SR, Dino HI, Dataflow Applications. 2019 22nd
Sadeeq MA, Rashid ZN, Hasan DA, et al. Euromicro Conference on Digital System
A survey of optical fiber communications: Design (DSD). 2019;88-95.
Challenges and processing time 104. Riasetiawan M, Ashari A. Performance
influences. Asian Journal of Research in analysis of FIFO and Round robin
Computer Science. 2021;48-58. scheduling process algorithm in IoT
94. Abdullah SMSA, Ameen SYA, Sadeeq MA, Operating system for collecting landslide
Zeebaree S. Multimodal emotion data. 2020 International Conference on
recognition using deep learning. Journal of Data Science, Artificial Intelligence, and
Applied Science and Technology Trends. Business Analytics (DATABIA). 2020;63-
2021;2:52-58. 68.
95. Teraiya J, Shah A. Comparative study of 105. Chen CY, Mohan S, Pellizzoni R, Bobba
LST and SJF scheduling algorithm in soft RB, Kiyavash N. A novel side-channel in
real-time system with its implementation real-time schedulers. 2019 IEEE Real-
and analysis. 2018 International Time and Embedded Technology and
Conference on Advances in Computing, Applications Symposium (RTAS). 2019;90-
Communications and Informatics 102.
(ICACCI). 2018;706-711. 106. Malik S, Ahmad S, Ullah I, Park DH, Kim
96. Capodieci N, Cavicchioli R, Bertogna M, D. An adaptive emergency first intelligent
Paramakuru A. Deadline-based scheduling scheduling algorithm for efficient task
for gpu with preemption support. 2018 management and scheduling in hybrid of
IEEE Real-Time Systems Symposium hard real-time and soft real-time
(RTSS). 2018;119-130. embedded IoT systems. Sustainability.
97. Yang T, Deng Q, Sun L. Building real-time 2019;11:2192.
parallel task systems on multi-cores: A 107. Doan D, Tanaka K. A novel task-to-
hierarchical scheduling approach. Journal processor assignment approach for
of Systems Architecture. 2019;92:1-11. optimal multiprocessor real-time
98. Cao S, Bian J. Improved DAG tasks scheduling. 2018 IEEE 12th International
stretching algorithm based on multi-core Symposium on Embedded Multicore/Many-
processors. 2020 IEEE 11th International core Systems-on-Chip (MCSoC).
Conference on Software Engineering and 2018;101-108.
Service Science (ICSESS). 2020;18-21. 108. D'souza S, Rajkumar R. Cycletandem:
99. Nasri M, Davis RI, Brandenburg BB. FIFO Energy-saving scheduling for real-time
with offsets: High schedulability with low systems with hardware accelerators. 2018
overheads. in 2018 IEEE Real-Time and IEEE Real-Time Systems Symposium
Embedded Technology and Applications (RTSS). 2018;94-106.
Symposium (RTAS). 2018;271-282. 109. Nguyen KK, Vien NA, Nguyen LD, Le MT,
100. Abeni L, Balsini A, Cucinotta T. Container- Hanzo L, Duong TQ. Real-time energy
based real-time scheduling in the linux harvesting aided scheduling in UAV-
kernel. ACM SIGBED Review. 2019;16:33- assisted D2D networks relying on deep
38. reinforcement learning. IEEE Access.
101. Baital K, Chakrabarti A. Various 2020;9:3638-3648.
approaches for high throughput and 110. Khan M, Seo J, Kim D. Real-time
energy efficient scheduling of real-time scheduling of operational time for smart
tasks in multicore systems. 2019 IEEE home appliances based on reinforcement
International Symposium on Smart learning. IEEE Access. 2020;8:116520-
Electronic Systems (iSES)(Formerly iNiS). 116534.
2019;402-405. 111. Chen JJ, Shi J, von der Brüggen G, Ueter
102. Chen H, Wen J, Pedrycz W, Wu G. Big N. Scheduling of real-time tasks with
data processing workflows oriented real- multiple critical sections in multiprocessor
time scheduling algorithm using task- systems. arXiv preprint arXiv:2007.08302;
duplication in geo-distributed clouds. IEEE 2020.
Transactions on Big Data. 2018;6:131- 112. Utkarsh K, Srinivasan D, Trivedi A, Zhang
144. W, Reindl T. Distributed model-predictive

50
 Ismael et al.; AJRCOS, 11(4): 35-51, 2021; Article [Link].73758

real-time optimal operation of a network of 114. El Ghor H, Chetto M, El Osta R.

smart microgrids," IEEE Transactions on Multiprocessor Real-Time Scheduling for
Smart Grid. 2018;10:2833-2845. Wireless Sensors Powered by Renewable
113. Lee D, Jung H, Yang H. Real-time Energy Sources. 2018 IEEE/ACS 15th
schedulability analysis and enhancement International Conference on Computer
of transiently powered processors with Systems and Applications (AICCSA).
nvms. IEEE Transactions on Computers. 2018;1-6.
2020;70:372-383.
_________________________________________________________________________________
© 2021 Ismael et al.; This is an Open Access article distributed under the terms of the Creative Commons Attribution License
([Link] which permits unrestricted use, distribution, and reproduction in any medium,
provided the original work is properly cited.

Peer-review history:
The peer review history for this paper can be accessed here:
[Link]

51