Also, we are NOT asking people to submit their code - we are coding our streaming engine ourselves https://github.com/nanosai/stream-ops-java! Instead we are asking people to share their benchmarks with us and use case etc.
5 karma · joined January 29, 2019
Also, yes the streaming engine will be useful to some use cases for Kahler. However, it is open source and so anyone will be able to access the same underlying engine.
By the way, we already have several people signed to the challenge including people working for notable tech companies!:)
WE (at Nanosai.com) will attempt to build a data streaming engine that can process 1 billion records per second, and release it as open source. If YOU want to try the challenge too, that's fine (e.g. someone already working on data streaming engine tech). We did not mean for YOU to solve this problem for US.
Our initial measurements and calculations show that it should be possible to reach 1BRS, although the records would have to be small. Still, a data streaming engine will always have some record iteration overhead, so it would take some tuning to get that overhead small enough to reach 1BRS even with 1 byte records.