| 1 | htop | reads /proc every 1.5 s and draws the differences | user space | the whole screen | short spikes averaged away; a rough first frame |
|---|
| 2 | /proc | the kernel’s counters as files, written when read | the kernel | every number | nothing: it reports and doesn’t fail |
|---|
| 3 | processes and threads | the programs; each thread is a task | user space | the rows, PID, USER, TIME+, the tree, Tasks | Z rows; fork failures when the PID limit is reached |
|---|
| 4 | logical CPUs | run instructions; every tick is charged to a mode | hardware | one bar each; CPU% | full bars and more runnable tasks than CPUs |
|---|
| 5 | system call interface | the door from a program into kernel mode | between user space and the kernel | the red share of the bars; system time in TIME+ | red growing while traffic doesn’t |
|---|
| 6 | interrupts and softirqs | handle hardware events, such as packets | the kernel | irq and softirq (detailed CPU time); ksoftirqd | one CPU busy with softirq; a busy ksoftirqd |
|---|
| 7 | network card (NIC) | receives packets and raises interrupts | hardware | nothing directly | interrupts piled on one CPU |
|---|
| 8 | scheduler | chooses who runs on each CPU, by weight | the kernel | S column, Tasks running, load average, PRI and NI, blue time | load above the CPU count with full bars |
|---|
| 9 | page tables | map each process’s pages to RAM | the kernel | VIRT, RES, SHR, MEM% | RES climbing and never falling |
|---|
| 10 | MMU | translates every address a program uses | inside each CPU | minor and major page faults | major faults every second |
|---|
| 11 | RAM | holds every page in use | hardware | the Memory bar’s total | MemAvailable near zero |
|---|
| 12 | page cache | keeps file data in spare RAM | the kernel, in RAM | the Memory bar’s cache and buffers | a cold cache: slow first reads and high DISK READ |
|---|
| 13 | block layer | queues reads and writes for the disk | the kernel | D state, iowait, the I/O tab | high load with idle CPUs; tasks in D |
|---|
| 14 | disk | stores the database files and the swap file | hardware | DISK READ and DISK WRITE | reads stuck at the volume’s limit |
|---|
| 15 | swap file | holds program memory moved out of RAM | the disk | the Swap bar | swap filling while major faults climb |
|---|
| 16 | kswapd and the OOM killer | free memory; kill a task when that fails | the kernel | kswapd0’s CPU%, the OOM column | a row that vanishes and a kernel log line |
|---|
| 17 | cgroups | cap a group’s CPU, memory and I/O | the kernel | only the CGROUP column | idle CPUs and a slow job; a kill with memory free |
|---|
| 18 | hypervisor | runs the VM’s vCPUs on real cores | under the VM | steal time, in the virt segment | a slice of every bar no process explains |
|---|