I’m the creator of thatDot Connect [1], a distributed asynchronous streaming graph interpreter, where we’ve been working on exactly this problem for about 7 yea

I’m the creator of thatDot Connect [1], a distributed asynchronous streaming graph interpreter, where we’ve been working on exactly this problem for about 7 years now. Scalable distributed graph processing has so many applications! We were fortunate enough to be funded by DARPA for a while working on automated APT detection for cyber-security. Some of our recent work has shown linear scalability of this approach on large clusters at high throughput. For example, we got to half a million complex graph events ingested per second on a combined read/write workload and could have just kept scaling. We’re in the process of writing up this work and sharing all the details. [1] https://www.thatdot.com/technology/ https://www.thatdot.com/technology/