We have all heard about the major outages of Amazon and Azure in the past and the many online services that were impacted by those outages. So how can you protect yourself against being "offline" for hours or days and what tools you can use to protect yourself against it? Learn how we protect our customers with distributed systems (cloud and on-prem) to mitigate outages and stay online even when the lights go out.


Comments are closed.

Dan Fey at 12:11 on 15 Nov 2017

Interesting high level discussion about best practices and scaling. Would have loved to see more in depth example of architectures.

Brady Pacha at 12:52 on 15 Nov 2017

Great talk, I'll be saving these slides.

Philip Sharp at 13:32 on 15 Nov 2017

Good introduction, but didn't go into much depth, even though there was plenty of time.

Colin O'Dell at 15:22 on 15 Nov 2017

This talk was a good high-level overview. It covered "what" you should do, but lacked some detail on the "how". I would have loved if the full hour was used to show several specific examples of addressing these problems. For example, on the topic of connecting to databases, maybe show an example of implementing Secrets Management or talking about distributing a database across multiple locations. Nevertheless, I do think it provided attendees with a good framework of things to investigate further.

Ryan Howe at 16:22 on 15 Nov 2017

i liked the talk as a good point to start thinking about for disturbing your app. Only complaint is to please change the color scheme of your slides to a higher contrast set.

Good talk that introduced important concepts and tools available for development. I would have liked to see more technical details on implementation to fill in the rest of the hour. The direction given was good though to help attendees know what is available and good practices. The lighting made the grey on black text a little hard to read.

Anonymous at 11:56 on 16 Nov 2017

promising start, but ending the presentation after 30 minutes (of 60) without any explanation is a no-go.

Bobby Pearson at 13:51 on 16 Nov 2017

Pros: Michelangelo clearly knows his stuff inside and out. He projects confidence without arrogance. The talk gave me a list of tools and practices to at least research, if not actively implement. The Simian Army discussion made me think about how my organization could be more fault-tolerant.

Cons: Needed more implementation detail, perhaps in discussing *how* a multi-provider, multi-region application can be set up. He (and all speakers in the mail hall) was poorly served by the room's acoustics.

Joanne Garlow at 13:59 on 16 Nov 2017

I liked the way you built up the stack of things to pay attention to.

I like the subject and it was more oriented to the methodology or strategy than the technical part of it. Also, I think that it should have lasted to cover the whole hour.