One of the benefits we have working at OpenCredo (OC) is the opportunity to both attend and speak (although not on this occasion) at conferences. For some of you, this may be pretty common, but OC was actually the first to offer me this as part of a broader learning and development plan.
Cloud-native development and delivery is a core area of expertise for OC and we are always looking for what’s new and interesting in this space. So when I was offered the chance to go to CloudNative London it seemed like a good place to start. With its diversity in topics and technologies, the conference provided a perfect opportunity to collaborate and hear from others in the industry and what they are doing in this space.
There is much to be gained by simply spending time chatting with people and I would highly recommend this. But for this blog, I thought it would be helpful to highlight and give a shout out to some of the talks which I found interesting. Here are three talks from Day 1 at CloudNative London 2020 that I would recommend watching:
Sam Newman opened up CloudNative London 2019 with a look at the history of public cloud adoption. Highlighting that traditional data centres and private clouds still account for the majority of spending, the keynote questions our need for control and explores the idea of vendor lock-in.
I’ve spent a lot of time working in the Financial Services and FinTech spaces and there is often a strong feeling that the challenges in this sector are unique and we are somehow special. That specialised security, privacy and large scale data processing requirements automatically preclude being able to use more commoditised offerings.
This talk very much encourages us to question this reality. Are we really so special that we need to miss out on the benefits of the public cloud and struggle to build out our own private versions? It offers some good insight into the considerations and tradeoffs around this topic, specifically analysing the concerns about vendor lock-in as it relates to the adoption of the public cloud.
Experience has shown me that developing within a distributed architecture is inherently unreliable and as Josh highlights in his talk, a narrow focus on log and metric collection can only provide a limited view of individual services. Instead, he focuses on how good distributed tracing techniques and approaches can be used to unlock a more holistic view of your world. I recommend watching this talk to get an overview of tracing and an update on the CNCF OpenTelemetry project that will land up merging the CNCF OpenTracing and Google originating OpenCensus tracing specifications. The talk is nicely balanced by a demonstration of the Jaeger tracing platform that aims to complement the OpenTelemetry specification.
For me, this was a timely talk as I am in the process of helping a client implement various observability tools. It also got me thinking about whether or not distributed tracing would’ve helped me in past projects; particularly in large-scale distributed trading systems.
The last talk that I recommend watching was delivered by the CTO of Nationwide Aubrey Steam. The talk covers a lot of ground from technology choices and delivery practices (testable Infrastructure-As-Code) to culture and communication. Aubrey looks at every stage of delivery and focuses on establishing a culture that fosters ownership, team identity that is supported by the right level of governance.
This was an entertaining talk and I found it interesting to draw parallels with what we are doing with some of our clients in regulated markets, including trying to take some of the highly agile start-up techniques and applying it to them.