SREcon16 Europe

The concept of the error budget is a great way to hack SLAs and make them into a positive tool for system engineers. But how can you take the same idea from a world that handles millions of transactions in a day to one that handles hundreds, but on the same hardware scale? High Performance Computing jobs run for hours, days, or weeks at a time, resulting in unique challenges related to system availability, maintenance, and experimentation. In this talk I will explore how we plan to modify the error budget concept to fit in an HPC environment by applying the same idea to cluster outages. With that specific example as the foundation, I will conclude with some thoughts on how the ideas generated in large scale web environments can be used in similarly sized environments running very different workloads.