Zync Up Engage Users with Real-time Location Services at Scale with DataStax EnterpriseJanuary 13, 2016
This post is one in a series of quick-hit interviews with companies using Apache Cassandra™ and/or DataStax Enterprise (DSE) for key parts of their business. For this interview, we talked with Aaron Benz, Founder and CEO of Zync Up.
DataStax: Tell us about Zync Up and your role there?
Zync Up: A simple night out currently involves messaging friends/family, looking up places to go (Yelp, Foursquare, etc…), figuring out a time that works, taking the action of going there (Uber, Lyft, Google Maps, etc…), finding the people you are planning on meeting, and finally keeping up with the group (for the common bar hoppers). We believe you shouldn’t have to use 3-4 apps to just go out with friends. Zync Up provides a one-stop-shop that enables you to do all these things from 1 app (well, shortly you will be able to do them all – some features in private beta). Overall, it is a location-centric mobile solution that streamlines the people, places and services in your world.
I am the CEO, Founder, and Janitor of Zync Up. One of my responsibilities is creating and managing our backend solution, which includes everything from the API to the database.
DataStax: Did you use a different technology before Cassandra?
Zync Up: We started off with a basic MySQL DB, but the plan was always to move to Cassandra – from my previous job I had some extensive experience with Cassandra & DSE and knew it was the right solution from the get go.
DataStax: Why did you pick Cassandra and DataStax Enterprise? What kind of data is stored there?
Zync Up: In addition to messaging and maintaining events, one of the biggest pieces of Zync Up is a real-time location streaming service. From a user’s perspective, they can share their location with specific people for a limited period of time -like until 3 a.m. (for all those situations when everyone is texting but none is making sense, knowing who is running late, and simply being able to find your friends/family when that “by the flagpole” description just doesn’t suffice).
From a technology perspective, that means that every user could be uploading their location every second. Well, if you want to do that at lightening speed planning for millions of users to be sharing their location with other users, places, or services, SQL is not going to get you very far. Through Cassandra, we are able to completely denormalize a user’s location for every “session” they are in – meaning every time the user wants an updated view of everything they are connected with, with Cassandra, it’s a Get, not a Query.
DataStax: You currently use the DataStax In memory option, what business use case does it fulfill?
Zync Up: A common use case is keeping the user table in memory (“traditionally” in something like Redis) because it gets called so often. For us, DSE in memory tables was an obvious solution purely so that we don’t have to maintain another database. That means we only have to speak one language, manage one cluster, and code with one package.
DataStax: What advice would you give to other startups that are thinking about using Cassandra for the first time in their solutions?
Zync Up: Really take some time and look over some articles/training specifically explaining why Cassandra and why it is not SQL. When you know why “denormalization is the new normal” you’ll know why or why not you would want to use Cassandra.
SHARE THIS PAGE