Real-time scheduling and enqueueing

Do your jobs need to run some jobs on an exact moment? JobRunr Pro will do realtime enqueueing out-of-the-box!

Some history

For performance reasons, JobRunr fetches all the jobs that need to be executed during the whole next pollIntervalInSeconds. This allows JobRunr to query, update and save jobs in bulk and that’s the reason why scheduled and recurring jobs are sometimes performed a couple of seconds too early.

JobRunr Pro to the rescue!

JobRunr Pro improves on this and:

• if the amount of jobs to schedule during the next pollIntervalInSeconds is less than 1000, it will automatically switch to real-time enqueueing with a one-second precision. This means if you have scheduled a job at 8pm, it will typically start processing within 50 milliseconds after 8pm.
• if the amount of jobs to schedule during the next pollIntervalInSeconds is more than 1000, JobRunr Pro will automatically fall back to near real-time enqueueing (so, a couple of seconds difference) again for the same performance reasons mentioned above.

One-second precision means JobRunr Pro will group all the jobs that need to be scheduled in buckets per second and enqueue these jobs at that second. So, if there are 2 jobs that are scheduled at 2023-01-30T20:00:00.350 and 2023-01-30T20:00:00.550, they will both be enqueued at 2023-01-30T20:00:00.000 (give or take a couple of milliseconds)