CODEWITHSUNDEEP
timescaledb
user27962
user27962

Reputation: 5

continuous aggregate is too slower than hypertable

timescale-db continuous aggregate materialized only = true is about 100 times faster than materialized only = false.

i expect to continuous aggregate be faster than calculate from real time. but real time calculate is faster than continuous aggregate.

also in this link i shared more detailtimescaldb-forum

i think continuous aggregate must be faster but they are too slow. i think something is wrong there

i turn materialized_view_only to true it works well then. problem is when is set materialized_view_only=false

Upvotes: 0

Views: 102

Answers (2)

Arlo Guthrie
Arlo Guthrie

Reputation: 1234

timescale-db continuous aggregate materialized only = true is about 100 times faster than materialized only = false.

i expect to continuous aggregate be faster than calculate from real time. but real time calculate is faster than continuous aggregate.

You have an incorrect understanding of materialized_only. When materialized_only=false, timescale performs a union with the live table to generate query results. This is the "real-time" option.

Using materialized_only=true limits the query to only the data that has been materialized previously. This option will be much much faster than performing a union with the live hypertable.

Upvotes: 0

jonatasdp
jonatasdp

Reputation: 1412

i turn materialized_view_only to true it works well then. problem is when is set materialized_view_only=false

Thanks for sharing the solution!

Upvotes: 0

Related Questions