Microservices vs Monolithic Advantages with real time Metrics.pptx

Micro Serivces vs Monolithic

1. Distributed Transactions and Evaluation of Performance of Microservices Balakrishna Pendiyala

2. Monolithic

3. Drawbacks Understanding. When a monolithic application scales up, it becomes too complicated to understand. Also, a complex system of code within one application is hard to manage. Making changes. It is harder to implement changes in such a large and complex application with highly tight coupling. Any code change affects the whole system so it has to be thoroughly coordinated. This makes the overall development process much longer. Scalability. You cannot scale components independently, only the whole application. New technology barriers. It is extremely problematic to apply a new technology in a monolithic application because then the entire application has to be rewritten.

4. Microservices

5. Strengths

6. Distributed Transactions

7. Database per service vs shared Database

8. Thesis Goals

9. Building Applications for Research

10. Building SubApp Monolithic Application

11. Building SubApp Microservices

12. E xperiment 1   Monolith Microservice with gateway Microservice without gateway Sample size 447 241 123 644 406 470 Failed requests 0 % 1,24 % 0 % Minimum (ms) 28 103 43 Maximum successful (ms) 64 717 71 892 10 850 To compare the monolith and the microservice prototype on single service queries, three different GET requests were sent to the two prototypes. For each request, the response time, the status code, and the number of active threads when the request was initiated were collected

13. Experiment 2 The aim of the second experiment was to compare the monolith prototype to the microservice solution with the API Composer service. This means that queries in the monolith prototypes translate into multiservice queries in the microservice prototype. The API Composer service handled these multiservice queries.   Monolith Microservices Sample size 332 602 42 251 Failed requests 0 % 2.51 % Minimum (ms) 106 139 Maximum successful (ms) 127 474 80 143

14. Experiment 3 This third experiment is an extension of the second experiment. However, instead of increasing the concurrent request made to the SubApp prototypes, the number of documents returned by the prototypes was increased for each iteration

15. Questions?