Skip to main content

What Synapsys is

Most companies do not know what is running where. A cron job on a server someone left. A Python script on a laptop. A Spring service nobody restarts because they are not sure what depends on it.

Synapsys gives all of them one place to be registered, started, stopped and accounted for — whatever language they are written in, whatever machine they sit on.

The model

  • A Core is the central node. One binary, one database, one console.
  • A worker (a neuron) is one of your applications. It connects out to the Core and never accepts inbound connections.
  • A process is a unit of work inside a worker — a job, a loop, a task.
  • An execution is one run of a process, with a start, an end and an outcome.

What it does not do

  • It does not run your code. Your process stays in your application, on your machine, with your dependencies. Synapsys starts and stops it; it does not host it.
  • It is not a scheduler. There is no cron syntax. If you need a schedule, your process owns it.
  • It is not a queue. Nothing is distributed to workers. A process belongs to exactly one worker.

Where to go next

  1. Run the Core — one command
  2. Your first worker — add the starter, watch it appear
  3. Start and stop a process — drive it from the console