I am working on modeling a certain system's cache layer and comparing two algorithms.

One outperforms the other one; it shows the higher cache hit rate (%).

But what is important is the impact of such an improvement at cache layer on the entire system performance and on the user experience.

The system performs message deliveries from producers to consumers. So the latency would be a good measure of the user experience.

However, because I am new to this area so I am a little lack of the background knowledge on it.

Is there any one who has worked on similar topics? Could you give me reference papers to get some understandings on what kind of measure can be used, how those can be represented, ... etc.

(For example, a paper with title "Improving Web Caching System by Adaptive ABC algorithm" can have such information.)

  • $\begingroup$ It sounds to me like you've shown that the cache is better (because of the improved hit rate) in theory and now is the time to implement a prototype and do some benchmarking. $\endgroup$ – adrianN Jul 21 '17 at 7:59

I think that a good starting point to understand how the memory (including caching) works is What Every Programmer Should Know About Memory from U. Drepper. In fact, this is one of the best papers I have ever read about computers.

More recently, I found How misaligning data can increase performance quite entertaining. It is a direct application of the things you can learn from the first one.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.