Open parc02 opened 7 months ago
https://github.com/miniproject4-team4/NGINX-PROXY-SERVER/issues/7#issuecomment-1970410758
-Vuser 80 캐시 key 만 남김 - 결과가 가장 좋음 https://github.com/miniproject4-team4/NGINX-PROXY-SERVER/issues/9#issuecomment-1970475629
캐싱을 적용하면 당연히 성능이 좋아질 줄 알았다. 하지만 캐싱도 여러가지 조건을 줘서 시스템에 최적화를 시켜줘야하고 테스트를 거듭할수록 캐싱이 성능개선에 많은 도움을 주는 지 의문점을 가지게 되었다.
시간적 여유가 부족해 다양한 조건들을 모두 적용해보며 최적의 캐싱을 찾지 못한게 아쉬웠다.
같은 조건에서 정상적으로 테스트 되던 케이스가 캐싱 이후 에러로 중단되어 어려움을 겪었고, 캐싱 설정의 중요성을 알게 되었다.
하나의 PC에서 진행한 테스트의 경우에는 CPU, 메모리가 100%를 넘어가 중단이 되었는데, 여러 PC에서 테스트를 진행한 경우 CPU, 메모리의 사용률이 높지 않았음에도 에러로 테스트가 중단되어 해결법을 찾는 데에 어려움을 겪었다.
PROXY CACHING 전
https://github.com/miniproject4-team4/NGINX-PROXY-SERVER/issues/7#issuecomment-1970410758
PROXY CACHING 후
-Vuser 80 캐시 key 만 남김 - 결과가 가장 좋음 https://github.com/miniproject4-team4/NGINX-PROXY-SERVER/issues/9#issuecomment-1970475629
결론
캐싱을 적용하면 당연히 성능이 좋아질 줄 알았다. 하지만 캐싱도 여러가지 조건을 줘서 시스템에 최적화를 시켜줘야하고 테스트를 거듭할수록 캐싱이 성능개선에 많은 도움을 주는 지 의문점을 가지게 되었다.
시간적 여유가 부족해 다양한 조건들을 모두 적용해보며 최적의 캐싱을 찾지 못한게 아쉬웠다.
같은 조건에서 정상적으로 테스트 되던 케이스가 캐싱 이후 에러로 중단되어 어려움을 겪었고, 캐싱 설정의 중요성을 알게 되었다.
하나의 PC에서 진행한 테스트의 경우에는 CPU, 메모리가 100%를 넘어가 중단이 되었는데, 여러 PC에서 테스트를 진행한 경우 CPU, 메모리의 사용률이 높지 않았음에도 에러로 테스트가 중단되어 해결법을 찾는 데에 어려움을 겪었다.