Overview
Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore performance monitoring techniques for distributed systems through the lens of theme park management in this 44-minute conference talk. Learn how to identify inefficiencies, optimize bottlenecks, and measure system performance using real-world examples from popular attractions. Discover key metrics such as throughput, concurrency, saturation, and queue length, and understand their significance in both theme parks and complex software systems. Gain insights from speaker Mike Minutillo's 18 years of experience developing and monitoring intricate systems, as well as his visits to world-renowned theme parks. Apply these concepts to improve the efficiency of your own distributed systems and, as a bonus, learn valuable lessons for running a successful theme park.
Syllabus
Intro
Who is Mike
Mikes first system
Disk space monitoring
Microservices
Disneyland
What is throughput
Flatlines
Roller coaster example
Maximum throughput formula
Experience concurrency
Experience throughput
Saturation
Saturation Graph
Full Saturation
Queue Length
Key Wait Time
Queue Wait Time
Back to work
Increasing concurrency
Wrapup
Taught by
NDC Conferences