Optimizing NoSQL Databases for SSDs to Achieve Real-time Performance

Unprecedented volumes of data are generated at high velocity via Web, social and mobile interactions and sensors, giving rise to new applications that respond immediately to what people, devices and systems are doing now. And it’s driving demand for NoSQL databases that can respond in milliseconds, scale to hundreds of thousands of TPS while managing billions of objects and terabytes of data, and maintain 100% uptime. This presentation:

Examines the economic and technical factors that allow NoSQL databases running on SSDs to maintain less than a 20-millisecond difference in speed at about one-eighth the cost of in-memory.
Discusses how the Aerospike database has been optimized for SSDs through a highly parallelized, distributed architecture.
Explores a range of use cases, including real-time consumer interactions on websites, mobile devices or games; hyper-fast fraud detection that occurs too quickly for a human to notice; and real-time analysis of network data to isolate and address issues before they significantly affect operations.

Brian Bulkowski, founder and CTO of Aerospike Inc. has 20-plus years experience designing, developing and tuning networking systems and high-performance Web-scale infrastructures. He founded Aerospike after learning first hand, the scaling limitations of sharded MySQL systems at Aggregate Knowledge as director of performance at this media intelligence SaaS company. Brian also has served as a founding member of the digital TV team at Navio Communications, chief architect of Cable Solutions at Liberate Technologies, and lead engineer at Novell.