- 필수 기능
- 시작하기
- Glossary
- 표준 속성
- Guides
- Agent
- 통합
- 개방형텔레메트리
- 개발자
- API
- Datadog Mobile App
- CoScreen
- Cloudcraft
- 앱 내
- 서비스 관리
- 인프라스트럭처
- 애플리케이션 성능
- APM
- Continuous Profiler
- 스팬 시각화
- 데이터 스트림 모니터링
- 데이터 작업 모니터링
- 디지털 경험
- 소프트웨어 제공
- 보안
- AI Observability
- 로그 관리
- 관리
The manner in which Datadog calculates the system.mem.used
metric produces a value that may sometimes be different from what might be displayed by common system resource reporting tools.
For example, running ‘free -m’ on an Ubuntu machine may produce the following memory breakdown (values represent megabytes):
total | used | free | shared | cached | available |
128831 | 1203 | 71975 | 4089 | 55653 | 122380 |
A Datadog Agent running on this same machine reports a system.mem.used
metric with a value of 56856 MB—clearly different from the ‘free -m’ used memory value of 1203 MB.
The reason for this discrepancy is that Datadog includes cached memory in its formula for used memory, where ‘free -m’ does not.
Datadog calculates used memory as follows:
Again, Datadog’s system.mem.used
metric includes cached memory, so subtracting this cached memory from used memory results in the following value:
1203 MB—identical to the used memory value reported by ‘free -m’ in the example above.
The system.mem.usable
metric represents free memory plus cached memory plus buffers (on Linux, it reflects “MemAvailable” attribute from /proc/meminfo whenever possible).
추가 유용한 문서, 링크 및 기사: