Ticketmaster: Best Practices for Optimizing Public Cloud Usage

Brent Eubanks, VP Technology Optimization at Ticketmaster, explains several best practices for leveraging public cloud effectively to take advantage of cloud’s continuous evolution, while avoiding the potential pitfalls that everchanging complexity puts in place.

Brent details why cloud pricing models from vendors like Amazon Web Services seem so complex, and how Ticketmaster applies the Pareto optimality to optimizing compute and RAM during public cloud modernization to approach Pareto optimal state when managing AWS.

He also explains that one advantage of cloud providers like AWS is that they iterate hardware much more quickly than traditional on-premises IT approaches, allowing enterprises to get more benefit for each dollar spent on RAM and compute. The same amount of work may be possible on R3 or R4 instances, but if his workloads can complete more work on a current gen chassis, his team can exploit more modern technology and get more value for their dollar. Brent also describes how the Pareto optimality provides a framework for avoiding gotchas where it may make sense to fall back to smaller AWS R instances.

Finally, Brent describes the overall optimization process at Ticketmaster and how Densify’s machine-learning-driven service has enabled progressive, cumulative wins, and helped Ticketmaster respond predictively to the seasonality of their business.