MongoDB复合分片键疑问：双玩家ID文档的分片存储规则咨询

Great question—this is a common point of confusion with compound shard keys, so let’s break it down clearly for your scenario.

Core Rule for Compound Shard Keys #

MongoDB uses the entire combination of fields in your compound shard key to determine chunk placement, not individual fields in isolation. That means:

• Documents will only end up in the same chunk if both players.0.id and players.1.id match exactly (and fall within the same chunk range, based on how MongoDB splits chunks over time as data grows).
• Having just players.0.id the same, or just players.1.id the same, does not guarantee documents are stored in the same chunk.

Example to Make It Concrete #

Let’s use sample documents to illustrate this behavior:

• Document A: { players: [{ id: "alice" }, { id: "bob" }] }
• Document B: { players: [{ id: "alice" }, { id: "charlie" }] }
• Document C: { players: [{ id: "alice" }, { id: "bob" }] }

Here’s how MongoDB treats these:

• Document A and C share the exact same compound shard key (alice + bob), so they’ll be placed in the same chunk (unless the chunk splits later due to size limits).
• Document A and B only share players.0.id (alice), but their players.1.id values differ. MongoDB sees these as distinct shard keys, so they could end up in different chunks depending on how the shard key ranges are split across your cluster.

Quick Note on Your Shard Key Choice #

Your logic here is solid: using this non-monotonic compound key should help distribute your data evenly across shards and avoid the hotspots that come with monotonic keys (like timestamps or auto-increment IDs). Just remember that MongoDB sorts chunk ranges first by players.0.id, then by players.1.id—so documents with the same players.0.id will live in contiguous chunk ranges, but not necessarily the same chunk unless their second key falls within the same interval.

内容的提问来源于stack exchange，提问作者Andrea Morandini