Comparing Cloud Models

We performed an in-depth analysis of Virtual Machines, Containers and Serverless using two different CPU-bound and IO-bound applications. We can conclude the following observations from our analysis - (i) Serverless is the most convenient for deploying and maintaining an application closely followed by Containers. The deploy- ment is much slower for Virtual Machines. (ii) For dif- ferent types of workloads, Containers give a better per- formance and scalability compared to Virtual Machines and Serverless. (iii) From a monetary standpoint, Server- less and Containers are cheaper compared to Virtual Ma- chines for smaller workloads, but Containers are the best option for larger workloads. Containers and Virtual ma- chine are more suitable for building services which need to be always online so as to avoid the overhead of load- ing the state in Serverless. Our study and the results have helped us understand the various dimensions of deploy- ing applications in the cloud and we will be able to take an informed decision in the future.

Avatar
Palash Chauhan
Masters Student in Computer Science

My research interests include distributed sytems, machine learning and their intersection.

Related