Hi folks. As you probably know in 4.0.2 we released pseudo-sharding which lets you parallelize your non-full-text query instantly to all your CPU cores. It’s good for purely analytical queries (w/o full-text), but doesn’t take any effect for most queries the people send to Manticore.

🔥 BUT recently we implemented pseudo-sharding for full-text which makes the new functionality beneficial for most use cases as long as you use plain indexes. What we are looking for now is people willing to try it out and provide us feedback on the new functionality.

🔥 What you’d have to do:

• measure the response time (and throughput if possible) you have now
• install a dev version (4.0.3). Instruction for ubuntu/debian, centos. If you run Manticore < 4 make sure to read about the breaking changes/migration.
• enable pseudo_sharding=1 in config
• measure the response time again
• provide us feedback
• let us know about any issues

What you may get:

• ✅ up to Nx lower response time where N is the number of physical CPU cores

Please let us know if you’d like to try the new feature and we’ll help you out.