[CASSANDRA-16340] STCS steady state load of table with vs. w/o GC performance test

• Type: Sub-task
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 4.0
• Component/s: Test/benchmark
• Change Category: Quality Assurance
• Complexity: Normal
• Platform: All

Source Control Link

none

Test and Documentation Plan

Perf. Test Plan

1. Run data prepopulation
2. Run steady state load for X seconds as Phase 1.
3. Make one test case dependent change, e.g. triggering garbagecollect, altering table, etc., and continue the steady state load for Y seconds as Phase 2.
4. Compare the performance metrics beween Phase 1 and Phase 2.

The workload used in the test was generated from tlp-stress.

Steady state load Read : Write : Delete == 5 : 4 : 1, and the QPS was kept at 3K/s.

• Assignee: Yifan Cai Assign to me
• Reporter: Yifan Cai

Authors

Yifan Cai

Reviewers

Marcus Eriksson

• Created: 12/Dec/20 00:32
• Updated: 10/Feb/21 09:15
• Resolved: 10/Feb/21 09:15

Description

The testing cluster should be pre-populated with ~200GB data in each node. The baseline cluster has the table created with provide_overlapping_tombstones disabled. The other cluster has the table with provide_overlapping_tombstones == row. Compare the read, write and compaction performance between those 2 clusters.