Skip to content

Benchmark deserialization

Add 4 new benchmarks:

  • concurrent deserialization

This benchmark uses executeConcurrent endpoint to insert n rows containing uuid, int, timeuuid, inet, date, time into the database. Afterwards it uses executeConcurrent endpoint to select all (n) of the inserted rows from the database n times.

JS:

node concurrent_deser.js <driver> <Number of queries>

Rust:

CNT=<Number of queries> cargo run --bin concurrent_deser_benchmark -r
  • deserialization

This benchmark executes n client.execute queries, that insert a single row containing uuid, int, timeuuid, inet, date, time waiting for the result of the previous query before executing the next one. Afterwards it executes n client.execute queries, that select all (n) of the inserted rows, waiting for the result of the previous query before executing the next one.

JS:

node deser.js <driver> <Number of queries>

Rust:

CNT=<Number of queries> cargo run --bin deser_benchmark -r
  • concurrent serialization

This benchmark uses executeConcurrent endpoint to insert n*n rows containing uuid, int, timeuuid, inet, date, time into the database.

JS:

node concurrent_ser.js <driver> <Number of queries>

Rust:

CNT=<Number of queries> cargo run --bin concurrent_ser_benchmark -r
  • serialization

This benchmark executes n*n client.execute queries, that insert a single row containing uuid, int, timeuuid, inet, date, time waiting for the result of the previous query before executing the next one.

JS:

node ser.js <driver> <Number of queries>

Rust:

CNT=<Number of queries> cargo run --bin ser_benchmark -r

Merge request reports

Loading