Elastic Scaling: Dynamic Resource Scaling in

Logistics through Microservices Architecture

Team Members:
Project Guide: 1. 21I208-Bhoomika S
Dr.Hema Priya N 2. 21I216-Gowtham M
3. 21I259- Tharun Srikanth L R
4. 21I261- Varshitha Bandaru
5. 21I265-Yogharaj A R
 AGENDA
• Abstract
• Problem statement
• Requirements
• Limitations of existing solutions
• Proposed methodology
• Timeline
• Frameworks : Docker and Kubernetes
• References
Abstract
• In today's dynamic logistics landscape, efficient resource allocation stands as
a cornerstone for operational success.
• "Elastic Scaling" is an innovative venture aimed at revolutionizing the
logistics industry by capitalizing on the inherent elasticity of microservices,
fortified by cutting-edge containerization tools like Docker and robust
orchestration platforms such as Kubernetes.
• This pioneering initiative redefines traditional resource allocation
methodologies by enabling the seamless auto-scaling of individual services
in response to the ever-fluctuating demands within logistics operations.
• The primary objective of Elastic Scaling is to orchestrate a paradigm shift,
optimizing resource utilization through the inherent flexibility of
microservices. By breaking down the logistics system into manageable,
independent services, this project lays the groundwork for precise workload
distribution, ensuring efficient handling of varying tasks and resource
demands.
• Leveraging containerization technology, Elastic Scaling
empowers logistics enterprises to dynamically adjust
resources, allowing for real-time adaptation to changes
in demand patterns.
• The significance of Elastic Scaling lies in its potential to
revolutionize logistics management paradigms. By
embracing microservices-based elastic scaling, this
initiative promises not only enhanced operational
efficiency but also scalability and responsiveness in a
continuously evolving industry landscape.
• Elastic Scaling aims to set a new standard in logistics
resource allocation, ushering in an era of agility, cost-
effectiveness, and unparalleled competitiveness for
logistics companies worldwide.
Problem Statement
Problem Domain: Cloud Computing
Elastic Scaling:
Resource allocation and load balancing for a logistics company using microservices
can be employed using several strategies to ensure efficient utilization of
resources and services.

Leverage the elastic nature of microservices to scale individual services based on

demand. Containers (e.g., Docker) and container orchestration tools (e.g., Kubernetes)
can facilitate auto-scaling to handle varying loads.

Microservices:
Microservices are collections of smaller, independently deployable services. In a
distributed architecture, each microservice works to accomplish a single feature or
business logic.
Load Balancing

IMAGE SOURCE: https://www.cloudns.net/blog/load-balancing/

Requirements
• Hardware:
 2 or more Server Systems or remote servers
 Multiple client machines

• Frameworks :
 Docker
 Kubernetes
 Limitations of existing solutions
 Round Robin (Issue when server fails)
 Sticky Round Robin (Issue with new user)
 Weighted Round Robin ( Changing weights is
critical)
 Linear Hashing (Cannot be implemented for
any number of servers)
 Least Connections (Indifferent for heavy and
normal request)
 Least Time (Monitoring overhead)
Proposed Methodology
A combination of weighted round robin, linear
hashing and least connections.
When request is received from the client , the
load balancer will fetch the server with least
connection and high weight and upon tie
between servers , hashing algorithm to
randomly route request to a server to ensure
homogeneous allocation of requests.
FRAMEWORKS : DOCKER AND KUBERNETES

• L i g h t w e i g h t , open, secure
platform
W h a t Is
D o c ke r ? • S i m p l i f y building, shipping,
running a p p s
• R u n s natively o n L i n u x or
Windows Server
• R u n s o n W i n d o w s or
Mac D e v e l o p m e n t
m a c h i n e s (with a virtual
machine)
• Relies o n " i m a g e s " a n d
"containers"
Kubernetes
Kubernetes is a tool that helps us to run and manage applications in
containers. It was developed by Google Lab in 2014, and it is also known as
k8s. It is an open-source container orchestration platform that automates the
deployment, management, and scaling of container-based applications in
different kinds of environments like physical, virtual, and cloud-native
computing foundations. Containers are isolated from each other so that
multiple containers can run on the same machine without interruption.
Why Kubernetes?

There are several reasons to learn Kubernetes like easy scaling of applications,
self-healing, portability, and automation. It is very helpful for running
microservices and distributed systems.
Features of Kubernetes:
• Consistent Development, management, and deployment
• Container-based infrastructure
• Utilization of resources in higher density
• Each component is like a separate unit
• Application-centric infrastructure
• Auto scalability
• Consistency is maintained across testing and development
 Kubernetes vs Docker

Kubernetes Docker

Docker is a tool that is used to automate

Kubernetes is an open-source platform the deployment of applications in
used for maintaining and deploying a group lightweight containers so that applications
of containers can work efficiently in different
environments.

In practice, Kubernetes is most commonly With Docker, multiple containers run on the
used alongside Docker for better control same hardware much more efficiently than
and implantation of containerized the VM environment & productivity of
applications. Docker is extremely high
References
 B. Tang, F. Guo, B. Cao, M. Tang and K. Li, "Cost-Aware
Deployment of Microservices for IoT Applications in Mobile
Edge Computing Environment," in IEEE Transactions on
Network and Service Management, vol. 20, no. 3, pp. 3119-
3134, Sept. 2023, doi: 10.1109/TNSM.2022.3232503.

Z. Ding, S. Wang and C. Jiang, "Kubernetes-Oriented
Microservice Placement With Dynamic Resource
Allocation," in IEEE Transactions on Cloud Computing, vol.
11, no. 2, pp. 1777-1793, 1 April-June 2023, doi:
10.1109/TCC.2022.3161900.
THANK YOU