Scheduling and its types

Dr Hitesh Mohapatra
Associate Professor
KIIT University, Bhubaneswar, Odisha, India
Definition
The concept of scheduling in cloud computing refers to the
technique of mapping a set of jobs to a set of virtual
machines (VMs) or allocating VMs to run on the available
resources in order to fulfil users' demands
Cont.
• Scheduling is a method that is used to distribute valuable computing
resources, usually processor time, bandwidth and memory, to the
various processes, threads, data flows and applications that need
them.
• Scheduling is done to balance the load on the system and ensure
equal distribution of resources and give some prioritization according
to set rules.
• This ensures that a computer system is able to serve all requests and
achieve a certain quality of service.
• Scheduling is also known as process scheduling.
Types
Scheduling in a system is done by the aptly named scheduler, which is
mainly concerned with three things:

• Throughput, or how fast it can finish a certain number of tasks from

beginning to end per unit of time
• Latency, which is the turnaround time or the time it takes to finish the task
from the time of request or submission until the finish, which includes the
waiting time before it could be served
• Response time, which is the time it takes for the process or request to be
served, in short, the waiting time
Cont.
• Scheduling is largely based on the factors mentioned in the previous
slide and varies depending on the system and the programming of the
system's or user's preferences and objectives.
• In modern computers such as PCs with large amounts of processing
power and other resources and with the ability to multitask by
running multiple threads or pipelines at once.
• Scheduling is no longer a big issue and most times processes and
applications are given free rein with extra resources, but the
scheduler is still hard at work managing requests.
Types of scheduling include:
• First come, first served — The most straightforward approach and
may be referred to as first in, first out; it simply does what the name
suggests.
• Round robin — Also known as time slicing, since each task is given a
certain amount of time to use resources. This is still on a first-come-
first-served basis.
• Shortest remaining time first — The task which needs the least
amount of time to finish is given priority.
• Priority — Tasks are assigned priorities and are served depending on
that priority. This can lead to the starvation of the least important
tasks as they are always preempted by more important ones.
Scheduling levels
In the cloud computing environment there are two levels of scheduling
algorithms:

• First level: in the host level where a set of policies to distribute VMs in the
host.

• Second level: in VM level where a set of policies to distribute tasks to the VM.
Task scheduling algorithms advantages
• Manage cloud computing performance and QoS.

• Manage the memory and CPU.

• Good scheduling algorithms maximize resource utilization while minimizing the total task execution time.

• Improving fairness for all tasks.

• Increasing the number of successfully completed tasks.

• Scheduling tasks on a real-time system.

• Achieving a high system throughput.

• Improving load balance.

Tasks scheduling algorithms classifications
Tasks scheduling algorithms can be classified
as follows
• Immediate scheduling: when new tasks arrive, they are scheduled to VMs
directly.

• Batch scheduling: tasks are grouped into a batch before being sent; this
type is also called mapping events.

• Static scheduling: is considered very simple compared to dynamic

scheduling; it is based on prior information of the global state of the
system. It does not take into account the current state of VMs and then
divides all traffic equivalently among all VMs in a similar manner such as
round robin (RR) and random scheduling algorithms.
Cont.
• Dynamic scheduling: takes into account the current state of VMs and
does not require prior information on the global state of the system
and distributes the tasks according to the capacity of all available
VMs.

• Preemptive scheduling: each task is interrupted during execution and

can be moved to another resource to complete execution.

• Non-preemptive scheduling: VMs are not reallocated to new tasks

until finishing the execution of the scheduled task.
Task scheduling system in cloud computing
The task scheduling system in cloud computing passes through three levels.

• The first task level: is a set of tasks (Cloudlets) that are sent by cloud users, which
are required for execution.

• The second scheduling level: is responsible for mapping tasks to suitable

resources to get the highest resource utilization with minimum make span. The
make span is the overall completion time for all tasks from the beginning to the
end.

• The third VMs level: is a set of (VMs) which are used to execute the tasks as in
Figure 2.
Scheduling
This level passes through two steps
• The first step is discovering and filtering all the VMs that are
presented in the system and collecting status information related to
them by using a datacenter broker.

• In the second step a suitable VM is selected based on task properties.

Static tasks scheduling algorithms in cloud
computing environment
FCFS: the order of tasks in the task list is based on their arrival time and
then assigned to VMs.
Advantages:

• Most popular and simplest scheduling algorithm.

• Fairer than other simple scheduling algorithms.
• Depend on the FIFO rule in scheduling tasks.
• Less complexity than other scheduling algorithms.
Cont.
Disadvantages of FCFS

• Tasks have high waiting time.

• Not give any priority to tasks. That means when we have large tasks in the
begin tasks list, all tasks must wait a long time until the large tasks to finish.

• Resources are not consumed in an optimal manner.

• In order to measure the performance achieved by this method, we will be

testing them and then measuring its impact on (fairness, ET, TWT, and TFT).
When applying FCFS, work mechanism will be
as following

• Dot arrows refer to the first set of tasks scheduling based on their arrival time.
• Dash arrows refer to second set of tasks scheduling based on their arrival time.
• Solid arrows refer to third set of tasks scheduling based on their arrival time.
Example: FCFS
Assume we have six VMs with different properties
based on tasks size:

VM list={VM1,VM2,VM3,VM4,VM5,VM6}.

MIPS of VM list={500,500,1500,1500,2500,2500}.
Execution

VM1={t1⟶t7⟶t12}.
VM2={t2⟶t8⟶t14}.
VM3={t3⟶t9⟶t15}.
VM4={t4⟶t10}.
VM5={t5⟶t11}.
VM6={t6⟶t13}.
SJF
Tasks are sorted based on their priority. Priority is given to tasks based on
task lengths and begins from (the smallest task ≡ highest priority).
Advantages
Wait time is lower than FCFS.
SJF has a minimum average waiting time among all task scheduling
algorithms.

Disadvantages
Unfairness to some tasks when tasks are assigned to VM, due to the long
tasks tending to be left waiting in the task list while small tasks are assigned
to VM.
Taking long execution time and TFT.
SJF work mechanism
When applying SJF, work mechanism will be as follows:
• Assume we have 15 tasks as in Table 1 above. We will be sorting tasks
in the task list, as in Table 3. Tasks are sorted from smallest task to
largest task based on their lengths as in Table 3, and then assigned to
the VMs list sequentially.
Execution: SJF

VM1={t4⟶t6⟶t7}.
VM2={t10⟶t9⟶t1}.
VM3={t11⟶t13⟶t8}.
VM4={t5⟶t14}.
VM5={t12⟶t2}.
VM6={t3⟶t15}.
MAX-MIN
In MAX-MIN tasks are sorted based on the completion time of tasks; long
tasks that take more completion time have the highest priority. Then
assigned to the VM with minimum overall execution time in the VMs list.

Advantages
• Working to exploit the available resources in an efficient manner.
• This algorithm has better performance than the FCFS, SJF, and MIN-MIN
algorithms
Disadvantages
• Increase waiting time for small and medium tasks; if we have six long tasks,
in the MAX-MIN scheduling algorithm they will take priority in six VMs in
the VM list, and short tasks must be waiting until the large tasks finish.
Max-Min
Assume we have 15 tasks as in Table 1 above. We will be sorting tasks
in the task list as in Table 5. Tasks are sorted from largest task to
smallest task based on the highest completion time. They are then
assigned to the VMs with minimum overall execution time in the VMs
list.
Execution: Max-Min

VM6={t1⟶t13⟶t11}.
VM5={t8⟶t9⟶t10}.
VM4={t7⟶t6⟶t4}.
VM3={t15⟶t3}.
VM2={t2⟶t12}.
VM1={t14⟶t5}.
Result Comparison
References
1. Ramotra A, Bala A. Task-Aware Priority Based Scheduling in Cloud Computing [master thesis]. Thapar University; 2013
2. Microsoft Azure website. [Accessed: 01 October 2017]
3. Kumar Garg S, Buyya R. Green Cloud Computing and Environmental Sustainability, Australia: Cloud Computing and Distributed
Systems (CLOUDS) Laboratory Department of Computer Science and Software Engineering, The University of Melbourne; 2012
4. Al-maamari A, Omara F. Task scheduling using PSO algorithm in cloud computing environments. International Journal of Grid
Distribution Computing. 2015;8(5):245-256
5. http://www.pbenson.net/2013/04/the-cloud-defined-part-1-of-8-on-demand-self-service/ [Accessed: 01 October 2017]
6. Endo P, Rodrigues M, Gonçalves G, Kelner J, Sadok D, Curescu C. High availability in clouds: Systematic review and research
challenges. Journal of Cloud Computing Advances, Systems and Applications. 2016
7. http://www.techinmind.com/what-is-cloud-computing-what-are-its-advantages-and-disadvantages/ [Accessed: 01 October 2017]
8. https://siliconangle.com/blog/2016/04/29/survey-sees-rapid-growth-in-enterprise-cloud-adoption/ [Accessed: 01 October 2017]
9. Aladwani, T. (2020). Types of Task Scheduling Algorithms in Cloud Computing Environment. Scheduling Problems - New
Applications and Trends. doi: 10.5772/intechopen.86873