查找操作缓慢

Question

I have the following query:

query = [{
    $match: {
      $and: [{
        feed: {
          $in: feeds
        }
      }, {
        updatedAt: {
          $gte: new Date(date)
        }
      }
    ]
    }
  },
  {
    $lookup: {
      from: "feed",
      localField: "feed",
      foreignField: "_id",
      as: "feed"
    }
  },
  {
    $unwind: {
      path: "$feed",
      preserveNullAndEmptyArrays: true
    }
  },
  {
    $lookup: {
      from: "source",
      localField: "feed.source",
      foreignField: "_id",
      as: "feed.source",
    }
  },
  {
    $sort: {
      updatedAt: -1
    }
  },
  {
    $limit: limit
  },
  {
    $skip: skip
  }
];

The query is taking too long, about 10 seconds to be specific.

I read https://docs.mongodb.com/manual/core/aggregation-pipeline-optimization/ but it is still rather slow.

I am looking to optimize as much as possible, is there a way? The collection is indexed for $updatedAt: -1

Any ideas?

Answer 1

If you want this to run fast I'd advise removing the $lookup pipeline stages and modeling your document based on its usage.

You could however add an index on the first match part of the aggregation query (updatedAt, feed). But you'll still have the problem that you are still looking up from 2 other collections which wont use any indexes.

The $match and $sort pipeline operators can take advantage of an index when they occur at the beginning of the pipeline.

https://docs.mongodb.com/master/core/aggregation-pipeline/#pipeline-operators-and-indexes