
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Chaos and Order</title>
      <link>https://www.youngju.dev/blog</link>
      <description>천천히 올바르게. AI Researcher &amp; DevOps Engineer Youngju&#39;s blog. GPU/CUDA, LLM, MLOps, Kubernetes AI workloads, and data engineering — plus mindset essays on confidence, routines, health, and sport psychology.</description>
      <language>ko</language>
      <managingEditor>fjvbn2003@gmail.com (Youngju Kim)</managingEditor>
      <webMaster>fjvbn2003@gmail.com (Youngju Kim)</webMaster>
      <lastBuildDate>Fri, 14 Aug 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://www.youngju.dev/tags/data-retention/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://www.youngju.dev/blog/observability/2026-08-14-langfuse-cost-retention-sampling.en</guid>
    <title>When Traces Become Cost — Retention, Sampling, and Masking in Langfuse</title>
    <link>https://www.youngju.dev/blog/observability/2026-08-14-langfuse-cost-retention-sampling.en</link>
    <description>Tracing feels free at low traffic and then comes back as storage cost and a privacy problem. This post splits the points where volume turns into cost into four, confirms from the official documentation that sampling decisions are made at the trace level and what that means in practice, and shows when the Python and JS masking paths actually redact data. It covers the fact that the retention default is to not delete anything, the minimum retention window and the timestamp fields deletion keys off, the extra bucket permission self-hosting requires, the lifecycle policy for the raw event bucket, and the settings that govern large payloads. Fifth post in the Langfuse and ClickHouse LLM tracing series.</description>
    <pubDate>Fri, 14 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>observability</category><category>langfuse</category><category>cost</category><category>sampling</category><category>data-retention</category><category>privacy</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/observability/2026-08-14-langfuse-cost-retention-sampling.ja</guid>
    <title>トレースが費用になるとき — Langfuseの保持、サンプリング、マスキング</title>
    <link>https://www.youngju.dev/blog/observability/2026-08-14-langfuse-cost-retention-sampling.ja</link>
    <description>トレーシングはトラフィックが少ないうちは無料のように見え、ある時点から保存費用と個人情報の問題として戻ってきます。本記事はLangfuseで量が費用に変わる地点を四つに分けて整理し、公式ドキュメントを基準にサンプリングがtrace単位で決まるという事実とそれが実務で何を意味するか、PythonとJSそれぞれのマスキング方式がデータをどの時点で隠すのかを確認します。保持ポリシーの既定値が削除しないであること、最小の保持期間と削除の基準になる時刻フィールド、セルフホストで追加で必要なバケット権限、原本イベントバケットのライフサイクルポリシー、そして大きなペイロードを扱う設定まで扱います。Langfuse + ClickHouse LLMトレーシングシリーズの第五回です。</description>
    <pubDate>Fri, 14 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>observability</category><category>langfuse</category><category>cost</category><category>sampling</category><category>data-retention</category><category>privacy</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/observability/2026-08-14-langfuse-cost-retention-sampling</guid>
    <title>트레이스가 비용이 될 때 — Langfuse의 보존, 샘플링, 마스킹</title>
    <link>https://www.youngju.dev/blog/observability/2026-08-14-langfuse-cost-retention-sampling</link>
    <description>트레이싱은 트래픽이 적을 때는 공짜처럼 보이다가 어느 순간 저장 비용과 개인정보 문제로 되돌아옵니다. 이 글은 Langfuse에서 양이 비용으로 바뀌는 지점을 네 곳으로 나눠 정리하고, 공식 문서 기준으로 샘플링이 trace 단위로 결정된다는 사실과 그것이 실무에서 무엇을 뜻하는지, 파이썬과 JS 각각의 마스킹 방식이 데이터를 어느 시점에 가리는지를 확인합니다. 보존 정책의 기본값이 삭제하지 않음이라는 점, 최소 보존 기간과 삭제 기준이 되는 시각 필드, 자체 호스팅에서 추가로 필요한 버킷 권한, 원본 이벤트 버킷의 수명 주기 정책, 그리고 큰 payload를 다루는 설정까지 다룹니다. Langfuse + ClickHouse LLM 트레이싱 시리즈의 다섯 번째 글입니다.</description>
    <pubDate>Fri, 14 Aug 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>observability</category><category>langfuse</category><category>cost</category><category>sampling</category><category>data-retention</category><category>privacy</category>
  </item>

    </channel>
  </rss>
