For now, I have been using this to test a toy/custom 2d client-server framework. All client events are piped to the server and processed in 1 gigantic ring buffer. This allows for ridiculous amounts of throughput due to batching effect. I am also using recurring, high-precision timers scheduled per client for purposes of triggering redraws and other important events as appropriate. This allows for complete decoupling of the event handling and client rendering pipelines. The breakdown is something like:
1 thread for ultra-low-latency timer execution
1 thread for processing the actual client event ring buffer, producing a consistent snapshot after each microbatch execution.
14+ threads for servicing HTTP requests (i.e. enqueuing client events), timers and redrawing client views using the near-real-time snapshots of business state.
My thinking is that if I can build something in this domain that is satisfactory, I could consider bringing it to an HFT firm as well.
1 thread for ultra-low-latency timer execution
1 thread for processing the actual client event ring buffer, producing a consistent snapshot after each microbatch execution.
14+ threads for servicing HTTP requests (i.e. enqueuing client events), timers and redrawing client views using the near-real-time snapshots of business state.
My thinking is that if I can build something in this domain that is satisfactory, I could consider bringing it to an HFT firm as well.