Matchlight is powered by the latest in large-scale computational technologies.

Matchlight bridges the gap between big data and information security. Built on technologies including the MapR Distribution of Apache Hadoop, Apache Spark, and Python, Matchlight constantly scours the web for matches against the fingerprints of your most sensitive data.

When you submit an asset to Matchlight, you compute a fingerprint of your data. That fingerprint is a one-way cryptographic representation of your data. No other party, not even Terbium Labs, can reverse the fingerprint into the original document. Our patented fingerprinting technology is based on standard cryptographic primitives and is fully auditable by our customers.

Your document’s fingerprint is loaded into our massive compute cluster, which immediately and automatically begins scouring the web and the dark web for matches. When we find a hit, we calculate a Matchlight Score and notify you immediately.

A system like Matchlight must be designed from the start for scale. And not just future scale — we already hold an immense number of fingerprints and grow by billions every day. Apache Hadoop, and especially the MapR distribution of Apache Hadoop with MapR-DB, is built for this scale. Apache Spark is a game-changer for batch and streaming processing across massive data stores, enabling us to match find hits on your most sensitive assets immediately upon seeing them in the wild.

Just a few years ago, monitoring assets at this scale with this level of privacy would have been impossible. With today’s cutting edge technology, coupled with Terbium’s expertise, we can provide this service and enable a customer to know within minutes when their most sensitive assets appear in a place they should not be.