Deleng panganggunaan sumber daya lan kejenuhan, banjur pikirake kanthi pangukuran. Cara sistematis — metode USE (Utilization, Saturation, Errors) — mriksa saben sumber daya supaya sampeyan ora salah nyalahake.
Kareben saben sumber daya (CPU, memori, disk, jaringan), takon telu bab:
UTILIZATION → how busy is it (% time in use)?
SATURATION → how much queued/waiting work (load, run queue, swap)?
ERRORS → any errors (dropped packets, disk errors, OOM kills)?
→ a resource that is saturated, not just busy, is usually the bottleneck.
Saben sumber daya nduweni tanda khas sing bisa sampeyan wacan karo alat standar:
CPU → top/htop: high %us or load > #cores, run queue backed up
MEMORY → free/vmstat: low free RAM + high SWAP in/out (si/so) → swapping = thrashing
DISK I/O→ iostat/vmstat: high %iowait, %util ~100%, high await → I/O-bound
NETWORK → iftop/sar/ss: bandwidth saturated, retransmits, high latency
Bedane utama: beban dhuwur karo %us dhuwur tegese CPU-bound; swapping (si/so) tegese tekanan memori; %iowait dhuwur tegese CPU iku wait nunggu disk; latency/retransmit dhuwur nunjuk jaringan.
$ vmstat 1
r b swpd free si so bi bo us sy id wa
8 0 0 120m 0 0 12 20 92 6 2 0 ← CPU-bound (id=0, wa=0)
1 3 800m 30m 40 120 900 8 5 3 2 90 ← I/O + swap (wa=90, so>0)
Jangan mandheg ing siji nomer. Profile proseso (contone perf, CPU/heap profiler) kanggo nemokake fungsi panas utawa alokasi, rubah siji bab, lan pangukuran maneh. Gejala bisa ngapusi — CPU dhuwur bisa sejatine busy-wait sing disebabake lock contention, dudu kerja nyata.
Nangani sumber daya sing salah (nambah RAM ing box CPU-bound) bubar duwit lan ora mutunga. Metode USE jeneng alat sing bener mengisolasi kendala asli kanthi cepet, lan pangukuran maneh pikirake yen perbaikan pindah jarum.