ARTICLE.md

Queuing Theory for Software Engineers

• Capacity planning -> from DAU to throughput
• Predicting software performance with Little's Law
• Monitoring Queues
• Traffic shaping with Queues

Exec Summary

Interesting Use-Cases Why????

Queues Are Everywhere!

• Fork-join pool with work-stealing mechanism
• Java thread pools & unbounded queues
• Scylla DB?

The Basics

Let's cover the basic concepts first before we discuss practical applications. What are the key metrics that are in play, when we're investigating a queue? We have the following relevant metrics when talking about queues in general.

• Arrival rate: The rate at which new work items arrive in the queue.
• Latency:

Capacity Planning: From DAU to Throughput

References

• https://en.wikipedia.org/wiki/Kendall%27s_notation
• http://elearning.kocw.net/document/lec/2012/JoongAng/ChaHoonSang/7.pdf
• http://elearning.kocw.net/document/lec/2012/JoongAng/ChaHoonSang/8.pdf
• https://qmodels.readthedocs.io/en/latest/mm1.html
• https://www.youtube.com/watch?v=12XbrjiZ1FA
• https://github.com/miguelrizzog96/Queue_Simulation_Python/blob/master/server.ipynb
• https://notebook.community/xunilrj/sandbox/courses/IMTx-Queue-Theory/Week2_Lab_MM1
• https://github.com/eveneveno/MMC_queue
• https://github.com/miguelrizzog96/Queue_Simulation_Python

Distributions

• https://numpy.org/doc/stable/reference/random/generated/numpy.random.exponential.html

Numpy

• https://numpy.org/doc/stable/user/basics.creation.html