Dynamic system feedback
We work with your teams to define and monitor the metrics and patterns that matter, providing real-time feedback on system changes in any environment.
Kubernetes infrastructure validation
We establish the tools and processes you need to validate your architecture, and effectively discover and troubleshoot issues that might arise.
Operational and DR practices
We equip your teams with the tools and techniques they need to establish and maintain a shared understanding of the operational nature of your system.
Red Team / Blue Team war-gaming
We boost the capability and confidence of your teams by facilitating active “war-gaming” to give your teams hand-on experience and real feedback.
The dynamic nature of a Kubernetes infrastructure means that the system is constantly evolving. Services can be added, removed and updated at any point – possibly many times a day. LiveWyer works with development and operations teams to hone their ability to support constant innovation.
Have confidence in your infrastructure.
Kubernetes allows for a significant reduction in spare capacity. Rather than running at 20% and keeping 80% of excess overheads to cope with spikes, when running on a dynamic plaform, you want your infrastructure running at 80% all the time.
There is so much data generated in a dynamic platform that the numbers are essentially meaningless. What matters is how the numbers change, and what causes them to change. Build your monitoring around these patterns.
Ensure that your platform defines and enforces standard ways to gather health and performance data from any application deployed to it. The system can only adapt based on what it can see. When the internals of an application are visible, responses can be far more nuanced.
Successfully gaining platform adoption brings many challenges with it. Understanding how your whole system responds, discovering how different system components impact each other and refining your core service configuration should not be happening in a crisis.
There is very little wasted resource on a well tuned Kubernetes infrastructure. At any point, the state of the system is well known and so it's able to stay balanced, moving processes across its underlying hardware as needed. There is no need to keep all sorts of reserve capability ready, just in case, as the various possible states have been well tested and the resources tuned to cope. If the data shows that new resources need to be added, it's a very easy, well rehearsed process.
Understand how your platform responds to changes.
In the old (static) paradigm, change was something rare, something to allocate to special disaster recovery scenarios. In a dynamic system, change is a day-to-day occurrence. In this environment, all aspects of your system, human and technical, need to be capable of managing change.
Kubernetes is designed for change, it has many built-in behaviours to manage situations. Your operations teams need to experience the system in real life to know how to stay in control in more difficult situations.
Use "war gaming" to train your teams and manage system handover. Simulate applications in a range of situations, get teams "hands-on" with problems before they happen in production.
Record actions and learnings from every scenario to keep your documentation aligned with the current system state.
The teams running a dynamic infrastructure are at home with its changing nature. They regularly simulate different possible system states to ensure that their whole team understands the way in which the platform and the hosted applications react to changes. Experiences from every simulation are recorded, feeding into system documentation as dynamic as the implementation.