Redis as atomic distributed cache system

In a high available high throughput distributed system, atomicity of any of the CRUD operation is extremely critical. Any inconsistency could easily propagate and extremely difficult to trace when we are talking about 1m transactions per second.

While it is hard to imagine a single server can handle 1m transactions per second, it is not so hard when it comes to highly clustered microservices. Then the question arise how do we maintain awareness between each microservices. Ideally a microservice should be independent to another instance of itself, but there are edge cases whereby there is a need for all instances to be aware or share information to one another.

This sharing can be achieved using database, or MQ, or Redis. In this post I will be talking about Redis and how the Transaction in Redis works, and provide example of high throughput concurrent test that Redis manage to handle easily

Redis Transction

Transaction is rather unique in Redis and takes time to understand, Redis guarantees atomicity for all comamnds running in between MULTI and EXEC, but it doesn't give any output. Output is returned only after EXEC is run. Therefore when handling transaction in Redis, we will need to think about how to and when to execute commands differently, for example below is a POP function for HASH, since redis does not return values while in transaction, we need to GET and then DELETE to emulate a POP.


The Concurrency Test

The concurrency test is rather simple, 2 Spring Boot application looping 10,000 times sending a request of fakeId between fakeId0 to fakeId10. Once received in the server, the server will store the fakeId and its count of occurences in redis HASH.


After the loop, the test will sleep for 30s to give time for each other (2 Spring Boot apps) to complete their iterations before calling processRequest() which will POP the HASH set.


The Result


As we can see below, the result from both apps, one of them will print exactly 20000 entries while the other will have nothing. The most important take away here is that since both apps are executing the for loop at roughly the same time and as fast as the CPU can go, it creates a heaven like environment for concurrency problem. The fact that the final output is still 20000 instead of some other number suggested that Redis is extremely resillient when it comes to concurrency handling






Source code: https://github.com/overtakerx/redistransactiondemo

Comments

Post a Comment

Popular posts from this blog

Image
Implementing gRPC Using Spring Boot and Gradle Background GRPC (gRPC) is a RPC framework pioneered by google later on made public and open source. GRPC has not been officially supported by Spring Framework, so far there are few community libraries that can work seamlessly with Spring Boot. In this article, I will use the lognet's grpc-spring-boot-starter. Project Structure Ideally, the proto files for gRPC is located in another directory, which will then be accessed by the gradle plugin to auto generate the required files.   Gradle Script The heart of gRPC enablement lies on the grpc boot starter, which includes few other tasks outlined as below. https://github.com/overtakerx/grpcbootdemo/blob/main/build.gradle Few important note: first you have to run gradlew compileJava for the auto generation to generate the files from the proto file, then you will need to rebuild the project for Intellij to recognise the auto-generated files. Full demo: https://github.com/overtakerx/grpcbootdem
Read more

Spring Boot 2: Parallelism with Spring WebFlux

Image
Spring Boot 2: Parallelism with Spring WebFlux Parallelism is rather unusual in Spring application, however when we encounter the need of HITLOL (HIgh Throughput LOw Latency), we need ways to make async calls. With Spring Boot 1, in high level definition, we can declare async on all the functions that are meant to be parallel and return Futures from there. Here is more detailed info:  https://medium.com/sipios/how-to-make-parallel-calls-in-java-springboot-application-and-how-to-test-them-dcc27318a0cf For Spring Boot 2, we shall employ Spring WebFlux libraries to accomplish parallelism. Mainly focused on functions: Mono.zipDelayError(), Mono.subscribeOn() and Mono.onErrorResume() The main function that drives the parallelism is zipDelayError(), the purpose of zipDelayError() as supposed to using zip() is to enable the other Monos to run to completion even if one Mono failed. AllDtos allDtos = Mono. zipDelayError (Arrays. asList (firstDto , secondDto , (Mono&l
Read more

Spring Boot Reactive API Part 2

Image
 Spring Boot Reactive: Improving CPU bound performance using Scheduler In part 1 , we have seen that Spring Boot Reactive doesn't really improve performance if the WAITING time is CPU bound / CPU intensive tasks. This article show a quick way on how we can improve the speed through specific configuration Setup Spring Boot's embedded Tomcat is reconfigured to only use 10 worker threads, this way we can clearly see the performance limitation and tweaks. Gatling is configured to simulate 30 concurrent requests with 30000 milliseconds request timeout. APIs internal process is calling java native code that calculate the value of PI over 1 million iterations Default / No scheduler The main reason why CPU bound tasks do not improve (or even degrade) with high parallelism is because of the costly overhead for context switching in the low level. In the scenario below, 10 main threads are all being used to calculate PI values and because there are less CPU than threads, a lot of context
Read more