{ "componentChunkName": "component---src-templates-post-tsx", "path": "/serving_term/", "result": {"data":{"logo":null,"markdownRemark":{"html":"

Sooftware Serving - Terminology

\n

NLP 모델 서빙 관련해서 계속 마주치는 용어들을 정리하는 페이

\n\n\"image\" \n\n

\n\n

서버의 상태가 통신이 불가능할 경우 서버를 서비스에서 제외하여 서비스를 원활하게 제공하기 위해 이용되는 방법

\n\n

한도를 넘어서는 부하가 걸렸을 때의 버그 등을 찾기 위함

\n\n\"image\"\n

\n\n\"image\" \n","htmlAst":{"type":"root","children":[{"type":"element","tagName":"h1","properties":{},"children":[{"type":"text","value":"Sooftware Serving - Terminology"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"p","properties":{},"children":[{"type":"text","value":"NLP 모델 서빙 관련해서 계속 마주치는 용어들을 정리하는 페이"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Balancer"}]},{"type":"text","value":": 서버에 가해지는 부하(로드)를 분산시켜주는 장치 또는 기술."}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"img","properties":{"width":400,"alt":"image","src":"https://user-images.githubusercontent.com/42150335/193868360-da7a4c4c-2a3b-4cd6-ac1b-ee9172451e5b.png"},"children":[]},{"type":"text","value":" \n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Balancing - Round Robin Method"}]},{"type":"text","value":": 서버에 들어온 요청을 순서대로 돌아가며 배정하는 방식"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Balancing - Weighted Round Robin Method"}]},{"type":"text","value":": 각 서버마다 가중치를 매기고 가중치가 높은 서버에 클라이언트 요청을 우선적으로 배분하는 방식"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Balancing - IP Hash Method"}]},{"type":"text","value":": 클라이언트의 IP 주소를 특정 서버로 매핑하여 요청을 처리하는 방식"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Balancing - Least Connection Method"}]},{"type":"text","value":": 요청이 들어온 시점에 가정 적은 연결상태를 보이는 서버에 우선적으로 트래픽을 배분하는 방식"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Balancing - Least Response Time Method"}]},{"type":"text","value":": 서버의 현재 연결 상태와 응답시간을 모두 고려하는 방식. 가장 적은 연결 상태와 가장 짧은 응답시간을 보이는 서버에 우선적으로 배분"}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"h3","properties":{},"children":[]},{"type":"text","value":"\n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Health Check"}]},{"type":"text","value":": 로드밸런싱 서비스를 제공하는 다수의 서버의 상태를 점검하기 위해 사용되는 기술로, 서버의 상태를 주기적으로 체크하여"}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"p","properties":{},"children":[{"type":"text","value":"서버의 상태가 통신이 불가능할 경우 서버를 서비스에서 제외하여 서비스를 원활하게 제공하기 위해 이용되는 방법"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Stress Test"}]},{"type":"text","value":": 어플리케이션에 각종 리소스(CPU, RAM, Disk 등)의 허용하는 한도를 넘어서서 비정상적인 높은 부하를 발생시켜보는 테스트."}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"p","properties":{},"children":[{"type":"text","value":"한도를 넘어서는 부하가 걸렸을 때의 버그 등을 찾기 위함"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Load Test"}]},{"type":"text","value":": 적절한 부하를 발생시켜 통계적으로 의미있는 수치를 측정하는 테스트. 1. 장시간 서비스 가능 여부를 확인하는 신뢰성 테스트, 2. 성능 테스트 (Performance Test)."}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"TPS (Transaction Per Second)"}]},{"type":"text","value":": 초당 트랜잭션의 갯수. 서비스 성능의 기준."}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Saturation Point"}]},{"type":"text","value":": TPS가 더 이상 증가하지 않는 시점."}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"img","properties":{"width":400,"alt":"image","src":"https://user-images.githubusercontent.com/42150335/193871571-f78af74b-eac1-4381-8dda-e55069a34054.png"},"children":[]},{"type":"text","value":"\n"},{"type":"element","tagName":"h3","properties":{},"children":[]},{"type":"text","value":"\n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"WSGI (Web Server Gateway Interface)"}]},{"type":"text","value":": Python 어플리케이션이 웹 서버와 통신하기 위한 인터페이스. 웹 서버에서의 요청을 파싱하여 Python 어플리케이션 쪽으로 넘기는 역할을 수행"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Uvicorn"}]},{"type":"text","value":": uvloop 및 httptools를 사용하는 *ASGI Web Server\n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"ASGI (Asynchronous Server Gateway Interface)"}]},{"type":"text","value":": 비동기 Python 웹 서버"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Asyncio"}]},{"type":"text","value":": Asyncio는 async / await 구문을 사용하여 비동기(Asynchronous) 코드를 작성하는 라이브러리. "},{"type":"element","tagName":"strong","properties":{},"children":[{"type":"text","value":"non-blocking"}]},{"type":"text","value":"하게 처리함."}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"uvloop"}]},{"type":"text","value":": asyncio를 대체하기 위해 만들어진 라이브러리. Cython으로 작성되었음. 기타 Python 비동기 프레임워크보다 2배 이상 빠름."}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"Gunicorn"}]},{"type":"text","value":": Python WSGI로 Web Server(Nginx)로부터 요청을 받으면 서버 어플리케이션으로 전달해주는 역할을 수행."}]},{"type":"text","value":"\n"}]},{"type":"text","value":"\n"},{"type":"element","tagName":"img","properties":{"width":500,"alt":"image","src":"https://user-images.githubusercontent.com/42150335/193873850-f25635a4-6863-4da9-ad97-7fbcb67c91e6.png"},"children":[]},{"type":"text","value":" \n"},{"type":"element","tagName":"ul","properties":{},"children":[{"type":"text","value":"\n"},{"type":"element","tagName":"li","properties":{},"children":[{"type":"element","tagName":"code","properties":{"className":["language-text"]},"children":[{"type":"text","value":"NGINX"}]},{"type":"text","value":": 가벼우면서도 강력한 프로그램을 목표로 개발된 오픈소스 웹 서버 프로그램. Apache보다 동작이 단순하고, 전달자 역할만 하기 때문에 동시접속 처리에 특화되어 있다."}]},{"type":"text","value":"\n"}]}],"data":{"quirksMode":false}},"excerpt":"Sooftware Serving - Terminology NLP…","fields":{"readingTime":{"text":"5 min read"}},"frontmatter":{"title":"Sooftware Serving - Terminology","userDate":"4 October 2022","date":"2022-10-04T10:00:00.000Z","tags":["nlp","serving"],"excerpt":null,"image":{"childImageSharp":{"gatsbyImageData":{"layout":"fullWidth","backgroundColor":"#f8f8f8","images":{"fallback":{"src":"/static/84b2e141b3f9a93f47a0cfa786f28c22/3ad85/term.png","srcSet":"/static/84b2e141b3f9a93f47a0cfa786f28c22/912f4/term.png 750w,\n/static/84b2e141b3f9a93f47a0cfa786f28c22/3ad85/term.png 900w","sizes":"100vw"},"sources":[{"srcSet":"/static/84b2e141b3f9a93f47a0cfa786f28c22/c23ad/term.webp 750w,\n/static/84b2e141b3f9a93f47a0cfa786f28c22/fc8e8/term.webp 900w","type":"image/webp","sizes":"100vw"}]},"width":1,"height":0.7744444444444445}}},"author":[{"id":"Soohwan Kim","bio":"Co-founder/A.I. engineer at TUNiB.","avatar":{"children":[{"gatsbyImageData":{"layout":"fullWidth","backgroundColor":"#282838","images":{"fallback":{"src":"/static/a9e6b445142b247ee4cfa66155398bb2/0d6f4/soohwan.png","srcSet":"/static/a9e6b445142b247ee4cfa66155398bb2/248f9/soohwan.png 40w,\n/static/a9e6b445142b247ee4cfa66155398bb2/fd435/soohwan.png 80w,\n/static/a9e6b445142b247ee4cfa66155398bb2/0d6f4/soohwan.png 120w","sizes":"100vw"},"sources":[{"srcSet":"/static/a9e6b445142b247ee4cfa66155398bb2/e7f45/soohwan.webp 40w,\n/static/a9e6b445142b247ee4cfa66155398bb2/589ec/soohwan.webp 80w,\n/static/a9e6b445142b247ee4cfa66155398bb2/71a38/soohwan.webp 120w","type":"image/webp","sizes":"100vw"}]},"width":1,"height":0.6833333333333333}}]}}]}},"relatedPosts":{"totalCount":50,"edges":[{"node":{"id":"6529d72e-80e7-5d61-a672-d66276a3f641","excerpt":"MoE(Mixture of Experts) 기초부터 DeepSeek 혁신까지 작년 이맘때쯤 DeepSeek-V3가 저비용으로 엄청난 성능을 보이면서 화제가 되었습니다. 그 핵심 기술인 MoE(Mixture of Experts…","frontmatter":{"title":"MoE(Mixture of Experts) 기초부터 DeepSeek 혁신까지","date":"2026-01-15T01:11:55.000Z"},"fields":{"readingTime":{"text":"8 min read"},"slug":"/developed-moe/"}}},{"node":{"id":"8b49d3ef-4ce3-568c-9d71-240ff17fc3e0","excerpt":"BERT는 사실 Diffusion 모델이였다?! 최근 굉장히 흥미로운 글을 읽게되어 공유합니다. 원문 : link BERT와 Diffusion이 같은 방식이다?! NLP 연구자들에게 BERT는 너무 익숙한 모델입니다. 201…","frontmatter":{"title":"BERT는 사실 Diffusion 모델이였다?!","date":"2025-10-21T12:00:00.000Z"},"fields":{"readingTime":{"text":"8 min read"},"slug":"/bert_diffusion/"}}},{"node":{"id":"7c7e1676-ea84-58de-970c-ddec9aa10660","excerpt":"RLHF는 수다쟁이를 만든다?! (Does RLHF Breed Verbose Chatterboxes?!) RLHF(Reinforcement Learning from Human Feedback)는 OpenAI의 ChatGPT…","frontmatter":{"title":"RLHF는 수다쟁이를 만든다?! (Does RLHF Breed Verbose Chatterboxes?!)","date":"2024-03-13T01:11:55.000Z"},"fields":{"readingTime":{"text":"7 min read"},"slug":"/rlhf-vervosity/"}}},{"node":{"id":"2b9d3e22-1796-5fab-878d-5941d2e76e9d","excerpt":"LLM Paper Abstract - 2023.12 LLM…","frontmatter":{"title":"LLM Paper Abstract - 2023.12","date":"2024-01-05T10:00:00.000Z"},"fields":{"readingTime":{"text":"5 min read"},"slug":"/llm-abs-202312/"}}},{"node":{"id":"f43fa33b-917b-5c00-8462-937d99592ad7","excerpt":"What it MoE? (Mixture of Experts) 현존 최강 LLM인 GPT-4에서 “MoE (Mixture of Experts)” 방식을 채택하여 사용하고 있다고 알려졌는데요, 최근 AI계의 뜨거운 감자 Mistral AI…","frontmatter":{"title":"What is MoE? (Mixture of Experts)","date":"2023-12-22T01:11:55.000Z"},"fields":{"readingTime":{"text":"9 min read"},"slug":"/moe/"}}}]}},"pageContext":{"slug":"/serving_term/","prev":{"excerpt":"Sooftware Serving - Triton Inference Server Triton Inference Server는 인공지능 모델의 인퍼런스를 도와주는 오픈소스 소프트웨어다. 다양한 프레임워크(TensorRT, TensorFlow…","frontmatter":{"title":"Sooftware Serving - Triton Inference Server","tags":["nlp","serving"],"date":"2022-09-20T10:00:00.000Z","draft":false,"excerpt":null,"image":{"childImageSharp":{"gatsbyImageData":{"layout":"fullWidth","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/2a825b8546ee78dfd77964613068ac58/60936/triton.png","srcSet":"/static/2a825b8546ee78dfd77964613068ac58/60936/triton.png 540w","sizes":"100vw"},"sources":[{"srcSet":"/static/2a825b8546ee78dfd77964613068ac58/1c40e/triton.webp 540w","type":"image/webp","sizes":"100vw"}]},"width":1,"height":0.5629629629629629}}},"author":[{"id":"Soohwan Kim","bio":"Co-founder/A.I. engineer at TUNiB.","avatar":{"childImageSharp":{"gatsbyImageData":{"layout":"fullWidth","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/a9e6b445142b247ee4cfa66155398bb2/7cf1f/soohwan.png","srcSet":"/static/a9e6b445142b247ee4cfa66155398bb2/34f77/soohwan.png 750w,\n/static/a9e6b445142b247ee4cfa66155398bb2/a94f6/soohwan.png 1080w,\n/static/a9e6b445142b247ee4cfa66155398bb2/7cf1f/soohwan.png 1148w","sizes":"100vw"},"sources":[{"srcSet":"/static/a9e6b445142b247ee4cfa66155398bb2/38420/soohwan.webp 750w,\n/static/a9e6b445142b247ee4cfa66155398bb2/7470d/soohwan.webp 1080w,\n/static/a9e6b445142b247ee4cfa66155398bb2/b5ef6/soohwan.webp 1148w","type":"image/webp","sizes":"100vw"}]},"width":1,"height":0.6829268292682927}}}}]},"fields":{"readingTime":{"text":"3 min read"},"layout":"","slug":"/triton/"}},"next":{"excerpt":"Sooftware Serving - Huggingface Optimum 허깅페이스에서 나온 Transformers의 Extension 라이브러리이다. 목적은 모델 학습 및 인퍼런스를 더욱 빠르게 해주기 위한 라이브러리이다. Exporting…","frontmatter":{"title":"Sooftware Serving - Huggingface Optimum","tags":["huggingface","nlp","serving"],"date":"2022-10-05T15:11:55.000Z","draft":false,"excerpt":null,"image":{"childImageSharp":{"gatsbyImageData":{"layout":"fullWidth","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/d0b8ba0447d1f90942743bf896396659/90018/optimum.png","srcSet":"/static/d0b8ba0447d1f90942743bf896396659/f503b/optimum.png 750w,\n/static/d0b8ba0447d1f90942743bf896396659/e4953/optimum.png 1080w,\n/static/d0b8ba0447d1f90942743bf896396659/c2802/optimum.png 1366w,\n/static/d0b8ba0447d1f90942743bf896396659/90018/optimum.png 1720w","sizes":"100vw"},"sources":[{"srcSet":"/static/d0b8ba0447d1f90942743bf896396659/77272/optimum.webp 750w,\n/static/d0b8ba0447d1f90942743bf896396659/317ed/optimum.webp 1080w,\n/static/d0b8ba0447d1f90942743bf896396659/b5c49/optimum.webp 1366w,\n/static/d0b8ba0447d1f90942743bf896396659/e0b55/optimum.webp 1720w","type":"image/webp","sizes":"100vw"}]},"width":1,"height":0.5232558139534883}}},"author":[{"id":"Soohwan Kim","bio":"Co-founder/A.I. engineer at TUNiB.","avatar":{"childImageSharp":{"gatsbyImageData":{"layout":"fullWidth","placeholder":{"fallback":"data:image/png;base64,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"},"images":{"fallback":{"src":"/static/a9e6b445142b247ee4cfa66155398bb2/7cf1f/soohwan.png","srcSet":"/static/a9e6b445142b247ee4cfa66155398bb2/34f77/soohwan.png 750w,\n/static/a9e6b445142b247ee4cfa66155398bb2/a94f6/soohwan.png 1080w,\n/static/a9e6b445142b247ee4cfa66155398bb2/7cf1f/soohwan.png 1148w","sizes":"100vw"},"sources":[{"srcSet":"/static/a9e6b445142b247ee4cfa66155398bb2/38420/soohwan.webp 750w,\n/static/a9e6b445142b247ee4cfa66155398bb2/7470d/soohwan.webp 1080w,\n/static/a9e6b445142b247ee4cfa66155398bb2/b5ef6/soohwan.webp 1148w","type":"image/webp","sizes":"100vw"}]},"width":1,"height":0.6829268292682927}}}}]},"fields":{"readingTime":{"text":"3 min read"},"layout":"post","slug":"/huggingface_optimum/"}},"primaryTag":"nlp"}}, "staticQueryHashes": ["3170763342","3229353822"]}