Preparing to run the benchmark
If the cluster nodes have SSD/NVMe caches, make sure they have been flushed before running any benchmarks. You can check the caches as follows:
-
On a cluster node, run:
# vstorage -c <cluster_name> top
- While in the text-based dashboard, press c to expand the chunks tab, and then cycle columns by pressing i until you see the
JRN_FULL
andFLAGS
columns. -
Wait until
JRN_FULL
is 0% and each CS ID is marked by thec
flag, for example,JCc
. This may take some time when the cluster is not under I/O load. The journal cleaning can be forced by running:# vstorage -c <cluster_name> set-config cs.force_checkpoint=1 # vstorage -c <cluster_name> set-config cs.force_checkpoint=0