Rest API architecture for handling thousands of requests per second?
Hi Guys,
Could you please help me with below scenario?
I have created rest api for receiving data based upon query parameters.
Right now I tested it with making multiple API calls.
Lets we get thousands of requests per second made to this api, how are we going to make changes so that it can handle those requests properly and respond back to each request?
How API services handle them?
Looking forward to your valuable insights! :)
Its a little hard to give a specific answer as it will depend on the computation your server requires per API request.
But assuming a basic CRUD type system where you're query a database, you might want to set up a caching system such as Redis which will save frequently accessed key-value pairings in memory (which is much quicker to access than reading from a database)
Beyond that you might consider a setup which will spin up multiple copies of your database that are then accessed via load balancing servers which handle and direct incoming requests.
HTH
Hi @srbushy,
Thank you for taking time and proving this information.
Yes I will definitely set up Redis.
Which load balancing service will you recommend to use? How can I implement scalable API? Lets initially we will get thousands of requests per seconds and then it may go up to millions per second.
I am new to API based designed. I would love to hear more from you. Thanks