Mercer-Lee's Space
Search K

English

简体中文

English

简体中文

Appearance

Traditional Pagination Query ​

javascript
db.users
  .find(conditions)
  .sort({ _id: -1 })
  .skip((page - 1) * numberOfPage)
  .limit(numberOfPage);

From the code above, this is a very traditional, unoptimized pagination query that retrieves data based on the page number and items per page. There's an obvious drawback: skip() traverses documents one by one. When the collection has a large amount of data, this becomes extremely slow.

Optimization ​

In typical pagination queries, the frontend maintains a page number variable as the key parameter for backend queries. But as discussed, this approach has significant downsides. Many optimizations involve using a unique value as the query condition with an index to speed up the query. The optimal choice is _id — as long as you have the _id of the last document from the previous query. This article focuses on this special MongoDB value.

javascript
db.users
  .find(...conditions, {'$lt': _id})
  .sort({_id: -1})
  .limit(numberOfPage)

ObjectID ​

I originally wrote quite a bit more, but then found someone had already written a better article explaining ObjectID, so I'll defer to that instead.