Currently I have a 4mb (compressed, according to /_cluster/state) cluster state and it's taking about 3 seconds to update every time it needs to be updated.
We have thousands of stored search templates in our cluster state and every now and then we need to update all of them, but it's taking hours to finish it.
- Is there any way exec multiple updates to scripts without updating cluster state each time?
- Will be a problem if I send batches of update requests? I mean using multiple queue workers to do the job.
- Any other ideas apart from that?
Thanks in advance,