test(performance): Add 90% storage utilization scaleout/scalein test

Perform scale-out around 90% disk utilization and cluster scale-in
by decommissioning the newly added nodes while measuring latency under
mixed-load stress.

#9156

Signed-off-by: Lakshmipathi.Ganapathi <[email protected]>

Author: Lakshmipathi

jenkins-pipelines/oss/features/elasticity/perf-regression-latency-i4i_2xlarge-elasticity-90-percent-scalein-scaleout.jenkinsfile

@@ -0,0 +1,12 @@
+#!groovy
+
+// trick from https://github.com/jenkinsci/workflow-cps-global-lib-plugin/pull/43
+def lib = library identifier: 'sct@snapshot', retriever: legacySCM(scm)
+
+perfRegressionParallelPipeline(
+    backend: "aws",
+    availability_zone: 'a',
+    test_name: "performance_regression_test.PerformanceRegressionTest",
+    test_config: """["test-cases/features/elasticity/scalein-scaleout-elasticity-90-percent.yaml", "configurations/disable_kms.yaml"]""",
+    sub_tests: ["test_latency_mixed_with_nemesis"],
+)

test-cases/features/elasticity/scalein-scaleout-elasticity-90-percent.yaml

@@ -0,0 +1,43 @@
+test_duration: 3000
+prepare_write_cmd: [
+  "cassandra-stress write no-warmup cl=ALL n=300000000 -schema 'replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate threads=200 -col 'size=FIXED(128) n=FIXED(8)' -pop seq=1..300000000",
+  "cassandra-stress write no-warmup cl=ALL n=300000000 -schema 'replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate threads=200 -col 'size=FIXED(128) n=FIXED(8)' -pop seq=300000001..600000000",
+  "cassandra-stress write no-warmup cl=ALL n=300000000 -schema 'replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate threads=200 -col 'size=FIXED(128) n=FIXED(8)' -pop seq=600000001..900000000",
+  "cassandra-stress write no-warmup cl=ALL n=250000000 -schema 'replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate threads=200 -col 'size=FIXED(128) n=FIXED(8)' -pop seq=900000001..1200000000",
+  "cassandra-stress write no-warmup cl=ALL n=300000000 -schema 'replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate threads=200 -col 'size=FIXED(128) n=FIXED(8)' -pop seq=1200000001..1500000000",
+]
+
+stress_cmd_w: "cassandra-stress write no-warmup cl=ALL n=335544320 -schema 'keyspace=refill_keyspace replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate threads=300 -col 'size=FIXED(128) n=FIXED(8)' -pop seq=1500000001..1835544321"
+stress_cmd_m: "cassandra-stress mixed no-warmup cl=QUORUM duration=800m -schema 'replication(strategy=NetworkTopologyStrategy,replication_factor=3)' -mode cql3 native -rate 'threads=300 fixed=16875/s' -col 'size=FIXED(128) n=FIXED(8)' -pop 'dist=gauss(1..600000000,300000000,6000000)'"
+
+simulated_racks: 3
+n_db_nodes: 3
+n_loaders: 4
+n_monitor_nodes: 1
+
+instance_type_loader: 'c6i.2xlarge'
+instance_type_monitor: 't3.large'
+instance_type_db: 'i4i.2xlarge'
+
+nemesis_class_name: 'GrowShrinkClusterNemesis'
+nemesis_interval: 30
+nemesis_sequence_sleep_between_ops: 10
+nemesis_add_node_cnt: 3
+nemesis_grow_shrink_instance_type: 'i4i.2xlarge'
+
+user_prefix: 'fullstorage-utilization-test'
+
+round_robin: true
+append_scylla_args: '--blocked-reactor-notify-ms 5 --abort-on-lsa-bad-alloc 1 --abort-on-seastar-bad-alloc --abort-on-internal-error 1 --abort-on-ebadf 1'
+backtrace_decoding: false
+print_kernel_callstack: true
+
+store_perf_results: true
+use_hdrhistogram: true
+email_recipients: ["[email protected]"]
+email_subject_postfix: 'fullstorage utilization test'
+parallel_node_operations: false
+
+append_scylla_yaml:
+  enable_tablets: true
+  auto_snapshot: false