Skip to content

Using of parameter -validate-data decrease throughput significant  #23

@juliayakovlev

Description

@juliayakovlev

When I run writes (sequential workload) with -validate-data parameter, the throughput became very low.
For example:
throughput without parameter: ~533 ops
throughput with parameter: ~50 ops

Example of command:
scylla-bench -workload=sequential -mode=write -replication-factor=3 -partition-count=3000 -clustering-row-count=1000 -clustering-row-size=51200 -partition-offset=15000 -concurrency=400 -rows-per-request=10 -max-rate=32000

Migrated to Jira: QATOOLS-48

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions