20 - CHAOS ENGINEERING questions and answers with solutions 2025
0 view 0 purchase
Module
Chaos
Institution
Chaos
Distributed systems are challenging & pervasive - ANSWER - multiple participants
- unreliable communication channels
- may be allowed to crash
- may need to tolerate malicious participants
- must eventually agree an some set of decisions
In general
- failure is always an option
- ordering ...
20 - CHAOS ENGINEERING questions
and answers with solutions 2025
Distributed systems are challenging & pervasive - ANSWER - multiple participants
- unreliable communication channels
- may be allowed to crash
- may need to tolerate malicious participants
- must eventually agree an some set of decisions
In general
- failure is always an option
- ordering is hard
- agreement is hard
- the software and what you think it does may differ
chaos engineering - ANSWER experimenting on a system to build confidence in the system's capability to
withstand turbulent conditions in production
- about finding the latent chaos in the system
*goal: manage chaos
- can think about chaos engineering as A/B testing for distributed systems where tests focus on
pathologies of system reliability
instead of looking for improvements, you look for degradation - ANSWER
The 8 fallacies of distributed computing - ANSWER - network is reliable
- latency is 0
- bandwidth is infinity
, - network is secure
- topology doesn't change
- one administrator
- transport cost is 0
- network is homogeneous
originally, experiments targeted these, but others are inspired by fault injection, race conditions, ...
How failure is handled varies depending on a system (5) - ANSWER - logging + continuing
- rerouting
- approximation + quality of service degradation
- error reporting
- terminal failure
(poorly tested + understood in distributed systems)
What impact might fallback strategies have on business performance? - ANSWER - companies don't
make money that day
- customers dislike service --> stop using it
Four common steps for a chaos experiment - ANSWER - measure + define the baseline behaviour of the
system
- hypothesize that the baseline should continue under stress
- simulate pathological behaviours on the deployed system
- try to disprove the hypothesis (show that there is a difference)
* the harder it is to show a difference, the more confidence you have in the robustness of your system
Just as in sequential hypothesis testing, you might want an "early out"
The benefits of buying summaries with Stuvia:
Guaranteed quality through customer reviews
Stuvia customers have reviewed more than 700,000 summaries. This how you know that you are buying the best documents.
Quick and easy check-out
You can quickly pay through credit card for the summaries. There is no membership needed.
Focus on what matters
Your fellow students write the study notes themselves, which is why the documents are always reliable and up-to-date. This ensures you quickly get to the core!
Frequently asked questions
What do I get when I buy this document?
You get a PDF, available immediately after your purchase. The purchased document is accessible anytime, anywhere and indefinitely through your profile.
Satisfaction guarantee: how does it work?
Our satisfaction guarantee ensures that you always find a study document that suits you well. You fill out a form, and our customer service team takes care of the rest.
Who am I buying these notes from?
Stuvia is a marketplace, so you are not buying this document from us, but from seller Performance. Stuvia facilitates payment to the seller.
Will I be stuck with a subscription?
No, you only buy these notes for £10.85. You're not tied to anything after your purchase.
