We could talk about the fact that we are making display perform better. And this is what we do right here. But the exciting part for us engineers is how we are doing this, really. Our technology takes an algorithmic approach to determining what user we show an ad to, when, and for what products. For billions of unique adverts that are created and displayed at lightning fast speeds every day.
Here are a few figures, because we like data, yes indeed we do:
- 15 datacenters (8 with computing capacity + 7 dedicated to network connectivity) across US, EU, APAC
- More than 15000 servers and a few tens of Petabytes of storage on our HPC cluster
- 37B HTTP requests and 3 B unique banners displayed per day
- Respond to bids in 80ms or less, 24/7
- Peak 2M http requests per second
- 3B ads displayed every day
- 6 ms on average to select optimal campaign
- 10 ms to find best product in catalogue of 100M
- 22B ad calls per day
- 20TB of new data stored daily
- 37PB of raw storage
- Largest public Machine Learning Dataset in the world with over 4 billion lines and over 1TB in size
- Technologies: Hadoop, Couchbase, Redis, Mesos, Kafka, Storm, Cassandra
The R&D department develops the infrastructure and tools that allow us to manage these ever increasing volumes of data. More advertisers supplying more catalogues of products, more publishers and more user variables mean that we are working in a very fast-paced release cycle and are adding new capabilities weekly and even daily.
The 300+ engineers@criteo are building the next generation of digital advertising technologies that allow us to manage billions of ad impressions, each one personalised automatically for its audience. This is truly big data and machine learning without the buzzwords! We have one of the best engineering teams in Europe, and our values are getting stuff done and having fun.