Thoughts on Tenant Isolation and Load Shedding

The Noisy Neighbor Problem

Our example service architecture.
A noisy-neighbor fills our server queue.
From top to bottom for each tenant: request latencies, rate of requests sent, rate of requests throttled, and request success %.

Don’t Use Fixed Resource Limits

Tenant Isolation in the Queue

A closer look at a fair queue.
A noisy-neighbor test with isolated server queues.
The WFQ adapts to the request rates of the different tenants.
In the absence of a WFQ, all tenants are throttled.

Final Thoughts

--

--

--

Programmer and aspiring lich in Seattle.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Guide to Making a Beautiful Game: Skyboxes in URP

What’s This and Can I Delete It? Examining A Default ASP.NET MVC Project

Grand Central Dispatch (GCD) part I

Task 19 | Configure K8S Multi Node Cluster Using Ansible | ARTH

5 Basic Problem Solving Techniques from “Think Like a Programmer”

How To Publish Your Own Python Package

Wordpress load times below 100ms

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Tony Allen

Tony Allen

Programmer and aspiring lich in Seattle.

More from Medium

Do you Waterproof your Software?

Importance of Logging + Best Practices

On Service Health Checks

Siren Song