I have a KVM single core VPS optimized for hosting static content running only Nginx 1.13.8 with SSD and 1GB RAM with a hosting provider on a package of 100 Mbit/s unmetered bandwidth.
OS in use is 3.10.0-693.17.1.el7.centos.plus.x86_64
The server is constantly at high load of
Load average: 2.59 2.16 1.90
There's no PHP or MySQL services installed or running on this server.
Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 1 On-line CPU(s) list: 0 Thread(s) per core: 1 Core(s) per socket: 1 Socket(s): 1 NUMA node(s): 1 Vendor ID: GenuineIntel CPU family: 6 Model: 13 Model name: QEMU Virtual CPU version (cpu64-rhel6) Stepping: 3 CPU MHz: 2266.746 BogoMIPS: 4533.49 Hypervisor vendor: KVM Virtualization type: full L1d cache: 32K L1i cache: 32K L2 cache: 4096K NUMA node0 CPU(s): 0 Flags: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pse36 clflush mmx fxsr sse sse2 syscall nx lm nopl pni cx16 hypervisor lahf_lm
The average daily bandwidth usage is 40Mbit/s
The maximum memory usage at any given time is between 300MB ~ 350MB
The nginx error log is only less than 8KB per day.
Looking at the
nginx service constantly shows 20~30% at the CPU column, which I believe isn't too bad.
What could be the possible causes of this high server load?
nginx is pretty resource efficient for serving static content, I can't think of a reason that's causing the bottleneck.
I'm suspecting the host maching hosting the VPS is overloaded, will this reflect the high figures in load average?
Given this high load average stats, nginx is still serving the static content and pushing the bandwidth smoothly.
But it's making me worry about the stability.
I've tried the exact same server deployment with another hosting company on a dual-core setup and distribute the traffic using RRDNS for testing.
The resource usage is only at 0.05 ~ 0.30 at all time with same amount of memory / bandwidth consumption.