eHarmony Increases The Speed Of Matchmaking With MongoDB

Dating website says the time it takes to analyse its user pool has gone down from 15 days to just 12 hours

eHarmony, a dating website that claims to process a billion potential matches daily, said that adopting open source NoSQL database MongoDB has helped it find matching couples around 95 percent quicker.

eHarmony was among the first dating services to apply algorithms and analytics to the art of matchmaking. The website has previously used a decentralised SQL-based database for its Compatibility Matching System, which connects members who are likely to enjoy a long-term relationship.

It says that the open source alternative has proven to be faster, more scalable, and comes with a long list of useful features.

Find me a find, catch me a catch

MongoDB, developed by 10gen, is one of the leaders in the emerging NoSQL movement, which rejects the orthodox approach of relational databases. NoSQL databases are not built on tables, and tend not to use structured query language to manipulate data, which allows for more flexibility.

fotoscooleHarmony says that running MongoDB instead of an unnamed SQL database has allowed it to cut the time it takes to go through the whole user pool in search of perfect couples from 15 days to just 12 hours.

“With our previous, decentralised SQL-based system, the entire user profile set was stored on each server, which impacted performance and impeded our ability to scale horizontally,” explained Thod Nguyen, CTO at eHarmony. “MongoDB supports the scale that our business demands and allows us to generate matches in real-time that appeal to new users when they first log in. Fast, accurate matching is at the heart of our business, and MongoDB is vital to that success.”

eHarmony says that it chose MongoDB for its rich query capabilities, flexible data model, built-in sharding and replication, and the ability to deploy new replica sets on demand.

“We’re seeing an increasing trend across verticals for organisations running real-time applications like the Compatibility Matching System that are benefiting from NoSQL technologies like MongoDB,” said Max Schireson, CEO at 10gen.

Do you know how many raspberries are used in the making of Raspberry Pi? Take our quiz!