Cloud Chaos & Microservices Mayhem • Holly Cummins • GOTO 2022

Cloud chaos and microservices mayhem require additional ops and management responsibilities, but understanding cost and optimization is key to success in a rapidly changing landscape.

Key takeaways
  • Chaos and complexity multiply with cloud and microservices - require additional ops and management responsibilities
  • Uncertainty and manual effort are increased, actual work load is underestimated (e.g. garbage collection), which leads to increased risk
  • Need to differentiate between “digital native” companies and others that lack expertise
  • Cloud usage has exploded, but FinOps is not a widely adopted practice yet - understanding cost and optimization is becoming crucial
  • Microservices goals may not be aligned with actual operational reality (e.g. legacy infrastructure)
  • Need for modularity is not necessarily solved with distribution and decentralization
  • Decentralized monitoring and observability of systems becomes increasingly important and poses a significant challenge
  • “Inadequate” testing and limited expertise lead to issues in legacy and microservices environments
  • Legacy processes, such as monoliths and waterfalls, can resist changes, especially in the face of disruption from newer technologies
  • Dev-Ops culture is not always properly understood and executed, leading to integration friction
  • As the pace of progress increases, testing and automation continue to evolve, and contract-driven testing may become more prevalent, potentially as a way to automate certain parts of the process