In-Memory Performance at the Cost of Flash
Reducing the number of data stack layers and incorporating real-time data processing engines which utilize CPU parallelism enables bare-metal application performance, running millions of application ops/sec per node with unprecedented latencies. Using flash as in-memory can substantially reduce costs and be extremely fast, but we only see a fraction of that speed when we layer OS abstractions, middleware and apps on top, forcing us to use a lot more hardware resources and settling for high and unpredictable latencies. Yaron Haviv will introduce this new architecture approach and present a smart mobility case study which illustrates how a major ride-hailing company achieved in-memory performance at the cost of flash with a parallel processing engine. Participants will learn how to utilize flash beyond storage without compromising speed.