Closed 107dipan closed 2 years ago
You are asking because you consider using this directly towards users interactively? There could be some things that could be sped up somewhat, but it's never going to be a good fit for that. Schema changes should be viewed more like a code change both in terms of resources spent and risk profile.
Hey,
We observed that response time was slower when we use map fields instead of normal fields. We thought instead of using map fields we can update the schema every time a new field is introduced if the schema update process is fast.
Thanks , Dipan
We observed that response time was slower when we use map fields instead of normal fields.
No, that isn't true. In any case the point of using a map here is that you don't need to update the schema and therefore there is no delay.
Just wanted to clarify I mean search response time in the previous post and not the time taken to deploy application ackage. Reference - https://github.com/vespa-engine/vespa/issues/19441
Thanks!
Yes, there it could be a difference, depending on what you do precisely, whether the fields have fast-search etc.
Note that in https://github.com/vespa-engine/vespa/issues/19441, for the second query, you use regular expression in the query (using matches instead of contains).
Got it... That cleared up my issue.
Thanks a lot!
Describe the bug We have a vespa setup with one config nodes, 8 container nodes and 18 content nodes using K8s. We are trying re-deploying the application package after updating the schema with one field added to an existing schema. But this process is taking around 4-5 mins. Is there any way we can reduce the time.
To Reproduce Steps to reproduce the behavior:
Expected behavior Is there any way to reduce the time taken to add a field to an existing schema.
Screenshots If applicable, add screenshots to help explain your problem.
Environment (please complete the following information):
Vespa version We used vespa:latest tag in the docker image
Additional context Add any other context about the problem here.