<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Milvus 对我们是不是太重了？]]></title><description><![CDATA[<p dir="auto">老板看到 Milvus 就说“专业”，但我们知识库只有 30 万 chunk。是不是上 Milvus 太重？</p>
]]></description><link>https://localaihub.com/topic/59/milvus-对我们是不是太重了</link><generator>RSS for Node</generator><lastBuildDate>Wed, 03 Jun 2026 18:50:37 GMT</lastBuildDate><atom:link href="https://localaihub.com/topic/59.rss" rel="self" type="application/rss+xml"/><pubDate>Sun, 03 May 2026 00:07:00 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 20:00:00 GMT]]></title><description><![CDATA[<p dir="auto">记得测更新。很多系统查很快，批量更新时把线上拖住。</p>
]]></description><link>https://localaihub.com/post/168</link><guid isPermaLink="true">https://localaihub.com/post/168</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Sun, 03 May 2026 20:00:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 19:23:00 GMT]]></title><description><![CDATA[<p dir="auto">也可以留一个 benchmark 脚本，别争论。用同一批数据和问题跑。</p>
]]></description><link>https://localaihub.com/post/167</link><guid isPermaLink="true">https://localaihub.com/post/167</guid><dc:creator><![CDATA[阿树]]></dc:creator><pubDate>Sun, 03 May 2026 19:23:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 18:13:00 GMT]]></title><description><![CDATA[<p dir="auto">这个判断靠谱。生产级不是选最重，是选你们能稳定承担的。</p>
]]></description><link>https://localaihub.com/post/166</link><guid isPermaLink="true">https://localaihub.com/post/166</guid><dc:creator><![CDATA[小乔同学]]></dc:creator><pubDate>Sun, 03 May 2026 18:13:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 15:20:00 GMT]]></title><description><![CDATA[<p dir="auto">我现在倾向 Qdrant 或 pgvector 起步，Milvus 等规模上来再看。</p>
]]></description><link>https://localaihub.com/post/165</link><guid isPermaLink="true">https://localaihub.com/post/165</guid><dc:creator><![CDATA[小马过河]]></dc:creator><pubDate>Sun, 03 May 2026 15:20:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 14:13:00 GMT]]></title><description><![CDATA[<p dir="auto">Milvus filtered search 文档要看，别把所有过滤都丢给后处理。</p>
]]></description><link>https://localaihub.com/post/164</link><guid isPermaLink="true">https://localaihub.com/post/164</guid><dc:creator><![CDATA[阿航]]></dc:creator><pubDate>Sun, 03 May 2026 14:13:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 13:31:00 GMT]]></title><description><![CDATA[<p dir="auto">还有备份恢复演练。向量库不是只要能查，误删 collection 时能不能回来才是生产问题。</p>
]]></description><link>https://localaihub.com/post/163</link><guid isPermaLink="true">https://localaihub.com/post/163</guid><dc:creator><![CDATA[小路灯]]></dc:creator><pubDate>Sun, 03 May 2026 13:31:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 11:21:00 GMT]]></title><description><![CDATA[<p dir="auto">抽象数据层。chunk_id、doc_id、metadata、vector、version 这些字段别乱，迁移就没那么痛。</p>
]]></description><link>https://localaihub.com/post/162</link><guid isPermaLink="true">https://localaihub.com/post/162</guid><dc:creator><![CDATA[rootless]]></dc:creator><pubDate>Sun, 03 May 2026 11:21:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 10:37:00 GMT]]></title><description><![CDATA[<p dir="auto">我担心以后迁移。现在用 Chroma，未来换 Milvus 会不会很痛？</p>
]]></description><link>https://localaihub.com/post/161</link><guid isPermaLink="true">https://localaihub.com/post/161</guid><dc:creator><![CDATA[小周]]></dc:creator><pubDate>Sun, 03 May 2026 10:37:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 07:50:00 GMT]]></title><description><![CDATA[<p dir="auto">我们 Milvus 用得还行，但排错门槛比 Chroma/Qdrant 高。新人进来先学半天概念。</p>
]]></description><link>https://localaihub.com/post/160</link><guid isPermaLink="true">https://localaihub.com/post/160</guid><dc:creator><![CDATA[hello_zhou]]></dc:creator><pubDate>Sun, 03 May 2026 07:50:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 06:54:00 GMT]]></title><description><![CDATA[<p dir="auto">不天然。召回质量更多看 embedding、切块、过滤、索引参数。数据库不是魔法。</p>
]]></description><link>https://localaihub.com/post/159</link><guid isPermaLink="true">https://localaihub.com/post/159</guid><dc:creator><![CDATA[林小北]]></dc:creator><pubDate>Sun, 03 May 2026 06:54:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 05:40:00 GMT]]></title><description><![CDATA[<p dir="auto">专门向量库召回会不会天然比 pgvector 好？</p>
]]></description><link>https://localaihub.com/post/158</link><guid isPermaLink="true">https://localaihub.com/post/158</guid><dc:creator><![CDATA[小树]]></dc:creator><pubDate>Sun, 03 May 2026 05:40:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 03:16:00 GMT]]></title><description><![CDATA[<p dir="auto">你们是否需要 multi-vector、混合检索、分区、复杂标量过滤？没有就别为了名字上。</p>
]]></description><link>https://localaihub.com/post/157</link><guid isPermaLink="true">https://localaihub.com/post/157</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Sun, 03 May 2026 03:16:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 02:57:00 GMT]]></title><description><![CDATA[<p dir="auto">如果以后多业务、多 collection、大规模更新，Milvus 有价值。只有一个内部问答，可能过度。</p>
]]></description><link>https://localaihub.com/post/156</link><guid isPermaLink="true">https://localaihub.com/post/156</guid><dc:creator><![CDATA[阿树]]></dc:creator><pubDate>Sun, 03 May 2026 02:57:00 GMT</pubDate></item><item><title><![CDATA[Reply to Milvus 对我们是不是太重了？ on Sun, 03 May 2026 00:51:00 GMT]]></title><description><![CDATA[<p dir="auto">先问谁运维。Milvus 能力强，但组件和排障成本也是真成本。</p>
]]></description><link>https://localaihub.com/post/155</link><guid isPermaLink="true">https://localaihub.com/post/155</guid><dc:creator><![CDATA[小乔同学]]></dc:creator><pubDate>Sun, 03 May 2026 00:51:00 GMT</pubDate></item></channel></rss>