Zabbix community Agent NVPS (new values per second) challenge - who will achieve the highest Zabbix agent NVPS score.
TL;DR:
Use your server with preinstalled Docker + with the most powerful CPU(s) and execute this command:
export USER_ID="<your public user identification>"
sudo curl -s \
https://raw.githubusercontent.com/monitoringartist/zabbix-agent-nvps-challenge/master/script.sh \
| sudo -E bash
Your results are submitted automatically and are available in the Public Google Sheet.
Current record holder is Monitoring Artist (66217.308253 NVPS):
$ docker exec -t dockbix-agent-xxl-nvps-challenge zabbix_agent_bench -host 127.0.0.1 -key stress.ping[] -timelimit 10 -threads 9
Testing 1 keys with 9 threads (press Ctrl-C to cancel)...
stress.ping[] : 662181 0 0
=== Totals ===
Total values processed: 662181
Total unsupported values: 0
Total transport errors: 0
Total key list iterations: 662181
Finished! Processed 662181 values across 9 threads in 10.000119568s (66217.308253 NVPS)
- use original Zabbix agent source codes/packages
- use
stress.ping[]
item provided by a zabbix-agent-stress-test module - use zabbix_agent_bench for benchmarking
- use
localhost/127.0.0.1
network interface - use 10 seconds timelimit
- try to experiment with
StartAgents
and-threads
settings to achieve the highest NVPS score
The easy solution is to use Docker and dockbix-agent-xxl container:
docker run \
--name=dockbix-agent-xxl-nvps-challenge \
--privileged \
-v /:/rootfs \
-v /var/run:/var/run \
-e "ZA_Server=0.0.0.0/0" \
-e "ZA_StartAgents=100" \
-d monitoringartist/dockbix-agent-xxl-limited:latest
cat /proc/cpuinfo | grep "model name" | uniq -c | sed 's/model name.*:/x/g'
date
docker exec -t dockbix-agent-xxl-nvps-challenge zabbix_agent_bench -host 127.0.0.1 \
-key stress.ping[] -timelimit 10 -threads 128
docker rm -f dockbix-agent-xxl-nvps-challenge
Public Google Form to submit your results manually.
Selected tested servers and achieved Zabbix agent NVPS score:
NVPS | Server description |
---|---|
~66k | 72 x Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz; AWS EC2 c5.18xlarge |
~56k | 36 x Intel(R) Xeon(R) CPU E5-2666 v3 @ 2.90GHz; AWS EC2 c4.8xlarge |
~44k | 64 x Intel(R) Xeon(R) CPU E5-2686 v4 @ 2.30GHz; AWS EC2 m4.16xlarge |
~8k | 4 x Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz |
Public Google Sheet with all results.
Devops Monitoring Expert, who loves monitoring systems and cutting/bleeding edge technologies: Docker, Kubernetes, ECS, AWS, Google GCP, Terraform, Lambda, Zabbix, Grafana, Elasticsearch, Kibana, Prometheus, Sysdig,...
Summary:
- 2000+ GitHub stars
- 10 000+ Grafana dashboard downloads
- 1 000 000+ Docker image pulls
Professional devops / monitoring / consulting services: