This repo contains code to mimic the out ot order / backfilled workloads on RedisTimeSeries >= v1.4.
Several aspects can dictate the overall system performance, like the:
- Pipeline size
- Number of distinct clients ( each client has a dedicated time-serie )
- Compressed / Uncompressed series
- Out of order ratio
This benchmark go program is know to be supported for go >= 1.13.
The easiest way to get and install the Subscriber Go program is to use
go get
and then go install
:
# Fetch this repo
go get github.com/RedisTimeSeries/redistimeseries-ooo-benchmark
cd $GOPATH/src/github.com/RedisTimeSeries/redistimeseries-ooo-benchmark
make
Usage of redistimeseries-ooo-benchmark:
-chunk-size int
chunk size. (default 4096)
-client-update-tick int
client update tick. (default 1)
-compressed
test for compressed TS
-debug-level int
debug level.
-host string
redis host. (default "127.0.0.1:6379")
-json-out-file string
Name of json output file, if not set, will not print to json.
-ooo-percentage float
out of order percentage [0.0,100.0]
-pipeline int
pipeline. (default 1)
-random-seed int
random seed to be used. (default 12345)
-samples-per-ts uint
Number of total samples per timeseries. (default 100000)
-ts-maximum uint
channel ID maximum value ( each channel has a dedicated thread ). (default 100)
-ts-minimum uint
channel ID minimum value ( each channel has a dedicated thread ). (default 1)