My inspiration is my brother and he introduce me with these kind of tech terms and motivate me every time do all this stuff
What it does
A bottleneck Localizer finds the bottleneck in a huge microservice environment.
How we built it
First I tried to Understand What is the main Difference between Microservice and Monolithic architecture and How use them to design a service , after that I designed my own microservice it is present in my GitHub repo, after that I started exploring distributed tracing concept and how can I Introduce tracing in my code and then I run my microservice in Kubernetes cluster and it is working fine after that I design the algorithm and started writing code and after that we got the good result and our code is working fine in our microservice application but our project is not vendor specific so I gone through benchmark in which I can test my code so check our on opensource GoogleCloudPlatform microservice website and it is working fine and giving us correct result.
Challenges we ran into
The first challenge we faced is how we can introduce sampling in our code because it is impossible to take all the traces from jaeger and do operation on them. So we went through some paper like dapper etc.. and then we know that we can't store every trace we introduce sample in our code which help us to improve the performance of our code and make our import from jaeger easy. We introduce 2 level of sampling in our code which can we modified according to needs.
Accomplishments that we're proud of
We can say proudly we tested our code on big microservice architecture website ex - GoogleCloudPlatform website and Heapster Service. The Code giving us appropriate result.
What I learned
I learned a lot from this project I get to know about some popular tools and how they are helping to solve problems , I learned about Container, Distributed Tracing, Jaeger, Kubernetes, Microservice and Monolithic Architecture, Azure, Go, Linux and other cool stuff.
What's next for Bottleneck Localizer
We are going to opensource the code base and try to show how this is useful and we will also publish a paper on this problem statement.
Log in or sign up for Devpost to join the conversation.