# Direct I/O 4K Random Write & Read Stress (libaio engine, iodepth=64)
$ fio --name=randwrite --ioengine=libaio --iodepth=64 --rw=randwrite --bs=4k --direct=1 --size=10G
Run status group 0 (all jobs):
WRITE: bw=4,120 MiB/s (4,320 MB/s), 1,030k IOPS, run=2487-2487msec
READ: bw=4,850 MiB/s (5,085 MB/s), 1,212k IOPS, run=2062-2062msec
clat (usec): min=18, max=312, avg=42.18, stdev=12.44
Disk Status: [RockHoster NVMe Gen4 PCIe] Zero I/O throttling detected over 30 days.
# Multi-Threaded CPU Event Rate (Prime verification up to 20,000)
$ sysbench cpu --cpu-max-prime=20000 --threads=4 run
Running the test with 4 threads...
CPU speed: events per second: 12,480.12
General statistics: total time: 10.0008s | total events: 124820
Latency (ms): min=0.28, avg=0.32, max=1.45, 95th percentile=0.35
Tested Hardware: AMD EPYC 9004 Series β’ CPU stolen time (%st): 0.00%
# MTR 100-Packet Route Quality & Latency from Frankfurt Test Rig
$ mtr --report --report-cycles=100 target-node.globalvps
HOST: fra-test-01.globalvps.site Loss% Snt Last Avg Best Wrst
1.|-- 10.0.0.1 (Gateway) 0.0% 100 0.3 0.3 0.2 0.8
2.|-- fra-ix.de-cix.net 0.0% 100 1.2 1.4 1.1 2.8
3.|-- Target VPS Instance (RockHoster)0.0% 100 3.8 4.1 3.7 6.2
Network Summary: 0.0% packet loss across 100 continuous cycles.