In a cloud operating environment, it is of utmost importance to understand performance, capacity and scale needs. It is also equally critical to identify areas of optimization and improvement before performance, capacity or customers get impacted negatively. This session talks about how to compute transaction efficiency and thereby find optimization opportunities proactively.
It’s 2:00am Saturday morning and you join a severity call because one of your biggest customers is facing a perf degradation and is unhappy! It may take you a few hours or even a few days because of the way data is dispersed but being the customer champion you are, you identify the bottleneck and get the problem resolved, putting a smile back on the customer’s face. Now imagine if you were aware of this impending perf issue a week, 2 weeks or even a month in advance? Wouldn’t that be great? Wouldn’t that be useful information to have to proactively tune and/or prepare for an upcoming surge in traffic? Engineering pain is just as real as customer pain! This session talks about how to identify optimization opportunities in code by computing transaction efficiency and keeping track of cost-to-serve (CTS). Any massive change in CTS on a weekly basis triggers the computation of CTS/efficiency (measured in terms of performance metrics) at a transaction level, which in turn provides data on opportunities to optimize and improve performance. This information is useful not only for performance engineers but also capacity planners who can then be made aware of situations where more servers or traffic migrations may be needed, in order to absorb the increase in CTS.
Principal Performance Engineer
Performance Engineering and DevOps, Data Centers and Cloud Infrastructure