Starting my Journey with Yellowbrick Data

February 27, 2019company

I’m excited to let you know that I have embarked on a new phase of my technical career as chief technology officer at Yellowbrick Data. I’m leaving my post at Aerospike and joining the team at Yellowbrick.

Why the move, you ask? What’s so awesome about Yellowbrick Data?

I’m a fan of building hyper-efficient systems that allow businesses and developers to do more, and do it quickly. At Aerospike, I worked to build the best-in-class NoSQL database with Strong Consistency and NAND Flash and clustering. The Aerospike database is still the best available at any price (not that I’m biased…) and I look forward to its continued success.

While I’ve truly enjoyed building Aerospike and watching it grow, I’m thrilled to join Yellowbrick Data and start a new adventure. Companies that ingest and process more data than the competition have always profited, and the SQL language presents a powerful form of primitives for in-database processing which is well understood by millions of developers -- and providing the next technological leap to those architects is where Yellowbrick Data comes in.

Yellowbrick Data has attacked and solved a very hard database problem: the large-scale SQL data warehouse and analytics quagmire. Yellowbrick has worked out the kinks in making a reference-based database fast. A custom data layout, optimized for flash and focused on column storage, is the primary data store, but there is also a row-based store for recent changes. An appliance model allows hardware optimizations, such as a patented technique of Native Flash Queries avoiding several round-trips to the DRAM subsystem and enabling the performance of general purpose CPU’s vector instructions. Use of internal InfiniBand and RDMA allow cross-node references to be resolved with extraordinary low latency.

The focus on hardware optimizations that Yellowbrick Data provides is purpose-built for the largest scale private cloud and internal data center environments, and the only viable path to continued database innovation. This approach will eventually work well in public cloud environments and Database-as-a-Service offerings. Instead of being locked into a vendor’s database or having to utilize virtual machines, Yellowbrick will be able to locate innovative hardware only milliseconds away from major public clouds using dedicated network connections, and provide managed services that outperform any other offering on the planet.

The Yellowbrick Data Warehouse provides technologists and business users with the next level of power, while also working with existing datasets and queries… and working with existing business structures and analytic systems is a great benefit.

As you can see, I’m very happy to be on this journey with my fellow colleagues here at Yellowbrick. If you want to learn more about Yellowbrick, visit our website, and if you want to learn more about why I joined, read our news release.