Five nines uptime without developer burnout

podcast

September 22, 2022

The times of conventional utility monitoring are fading. Purposes as we speak are now not a single program, however a community of companies related by API and RPC endpoints throughout cloud containers which can be created and eliminated as wanted.

Avatar for Ben Popper

The times of conventional utility monitoring are fading. Purposes as we speak are now not a single program, however a community of companies related by API and RPC endpoints throughout cloud containers which can be created and eliminated as wanted. There’s extra knowledge coming from sources that won’t persist for lengthy. Diagnosing errors turns into more durable, particularly in a disaster. 

On this sponsored episode of the podcast, we spoke with Martin Mao, CEO and co-founder of Chronosphere, concerning the elevated issue in diagnosing errors in manufacturing, particularly throughout a disaster. Advanced methods create extra knowledge, and to paraphrase: extra knowledge, extra issues. Chronosphere needs to do extra than simply offer you observability into your issues; they need to assist automate the response. 

Present notes

Like other folks we’ve talked to on the podcast, Chronosphere was born out of labor pioneered at Uber. When you may’t discover options that can assist you scale, generally it’s a must to construct them. 

All the pieces in Chronosphere was constructed from scratch, from the ingestion tier to the question layer. If you happen to’re going to construct one thing cloud native from the bottom up, the clear alternative for the staff was Go

Cloud native observability modifications the best way builders work together with their code in manufacturing. Infrastructure is extra advanced, dev and check environments are gone, and knowledge will increase massively whereas knowledge sources are extra ephemeral. 

Congrats to david, who received a lifeboat badge for his or her reply to Can we convert a byte array into an InputStream in Java?

Tags: partnercontent, the stack overflow podcast

More Posts