Motivation

An ill workman quarrels with his tools

When developers are building an application, they should be choosing the best available tools. However, mostly the developers are bound to choose the tools they best know how to use. Although the application is built and delivered successfully, they could never go back to see if the results could have been better had they chosen a different set of tools.

Our objective is to take the tools out of the equation. Saral helps developers to be able to build an application in any of the desired technologies. They can compare the results and choose to deliver the one that works best.

Distributed systems fail more often than not

We do not have the right tools to build applications natively for distributed systems. We still build applications to run on one machine and then make them work in a distributed environment by including - scalability and resiliency.

In data processing, Apache Spark has been a successful experiment in natively building distributed data processing applications. In ML/AI pytorch lightning also tries to address the same concerns.

Our objective is to replicate the success of apache spark and pytorch lightning with Project Saral.

SDLC process

1. Plan and design
2. Implementation
3. Testing
4. Optimization
5. Deployment
6. Monitoring
7. User Trials

Saral helps the developer by taking care of point 2 to pint 6. So that they can spend more of their energies on 1 and 6, i.e., design and user trials.