Tag Archives: Systems

Defending Against Complexity With Exercise

How do you manage complexity? Something we talk about a lot in Cloud2030 is how challenging it is to understand complexity, measure it and cope with it.

Richard Cooke wrote a paper called “How Complex Systems Fail,” (how.complexsystems.fail) and in it he talks about complex systems having strong defense mechanisms against failure. That’s what we talked about today. How do we build defense mechanisms for complex systems, not by making them simpler, but by exercising them and testing them?

We discuss the importance of testing, validation, and layer of abstraction and testing the layers in this conversation. If you deal with complex systems, this discussion will be fascinating and actionable.

Transcript: otter.ai/u/SP-z7OAJWAmJlql8Dh62rNk2hlo
Image: www.pexels.com/photo/man-woman-m…ng-young-4058411/

Rob’s Hot Take:

In the May 24th DevOps lunch and learn, Rob Hirschfeld delves into the concept of making complex systems defensible by exercising and testing them thoroughly. Emphasizing the importance of shared automation and collaborative efforts within communities, he cites examples like Kubernetes and OpenStack as complex systems made more defensible through widespread testing and shared code. While complexity cannot be eliminated, actively exercising systems enhances their defensibility. Join the ongoing discussions and explore the intricacies of complexity management at the2030.cloud.

Can We Measure Complexity?

We seem to be very worried about complexity in technology, but how bad is it really? Do we have a way of measuring complexity? Figuring out how to actually quantify it could help eliminate and manage it.

We started by discussing mathematical concepts to capture the systemic nature of complexity. That turns out to be really hard, so we got into some really interesting thoughts about what it takes to manage and understand complexity. Is it even possible to measure complexity? The group is mixed.

Transcript: otter.ai/u/qWkkgyKCXX89jcirBdni9ExkOq8
Photo: www.pexels.com/photo/random-obje…-balance-9304725/

Can Machines Update Themselves?

We know that humans have trouble keeping systems updated, but… how can we address the challenge of knowing which updates are required and, critically, if the updates with break other systems? Even knowing if they worked is a really thorny problem!

In this episode, we focus on actions about what’s going on and why this problem has persisted in industry for so long. Starting from the news of the day about CentOS 8 mirrors being taken down. That’s exactly the type of challenge we are facing when we think about where updates and repos are coming from.

Transcript: otter.ai/u/rRMIT6kkTTtyWrzdBnuq63nvKuE
Photo: www.pexels.com/photo/a-man-using…quipment-5996696/

Rob’s Hot Take:

Rob Hirschfeld, CEO and co-founder of RackN, discusses the challenges of system maintenance and lifecycle in the Cloud 2030 podcast. He emphasizes the difficulty of keeping systems up to date and understanding dependencies, leading to a lack of confidence in system updates due to the fear of breaking or degrading them. Hirschfeld advocates for a change in the industry to prioritize test and verification practices, enabling more effective and confident system updates.

Evolution of Networking Systems

How do we evolve technology in the future?  We centered the answer on networking, but in a very general way.

The ability for a vendor to distribute technology and then connect things together and then build networks of that technology is a core component of how networking is evolving.   Ultimately, this is about building technology systems.

Sadly, that led us into a very dark place where we really thought through who’s going to own all of that infrastructure and their motivations.  How can we make sure that the people’s needs and the systems and the vendors’ needs are well aligned?

Transcript: https://otter.ai/u/H8ik1HcmSExVBRklYbSCsV4yu2k

Image: https://www.pexels.com/photo/shallow-focus-photography-of-keychains-1194036/

Rob’s Hot Take:

Rob Hirschfeld, CEO and co-founder of RackN and host of the Cloud 2030 Podcast, reflects on the November 2nd DevOps Lunch and Learn session, which explored the motivations and impacts of building highly networked systems. While acknowledging the significant societal benefits of interconnected technology, Hirschfeld emphasizes the importance of understanding and mitigating the potential risks, particularly regarding data control and accessibility. He advocates for building inclusive systems that prioritize adaptability and service to the greater good, inviting listeners to explore the comprehensive discussion on these critical topics at the2030.cloud.

Securing Software Supply Chains

Today we talked about supply chains, but mainly security and the security aspects of supply chains because we have a very serious challenges here.

We have made software and on boarding software for developers so easy, but haven’t put the same efforts in how to manage production systems! The team really talked about what it takes to build production systems that respect security, supply chains, dependency graphs, and inclusion in a way that cross teams.

It’s an incredibly important topic, and it is the foundation of any successful supply chain hardening effort.

Transcript: otter.ai/u/6zfld2gBpZMSGT8Vk_1Ka3pWtN0
Image: www.pexels.com/photo/light-city-…traffic-10390684/