Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>It looks like the example query you give for an id of 6 matches 47402 documents:</p> <pre><code>&gt; db.results.find({id:6}).explain() "n" : 47402, </code></pre> <p>and the average document size is about 14654 bytes:</p> <pre><code>&gt; db.results.stats() "avgObjSize" : 14654.313273911172, </code></pre> <p>Does this match your view of the situation?</p> <p>This means that the results from this query will be about 662 MB in size. If the 662 MB of documents returned by the query are not in memory and must be fetched from disk then a lot of disk activity will result and the query will take a long time.</p> <p>There are a few ways you could improve performance, depending on your application requirements:</p> <ul> <li><p>use a more specific query that matches fewer documents, for example use _id to fetch a single document or specify a datetime as well as an id in your query.</p></li> <li><p>refactor your document schema to make the documents you wish to query smaller.</p></li> <li><p>use a projection in the query to select only a subset of the 14KB document to return and build an index that contains the fields in that query projection, so that the query can be satisfied entirely from the index. Note however that this will only work if you select a small enough subset of the document so that your index does not grow to be large, putting you back in the same situation of having to page the index in from disk. The db.results.stats() output tells you how big your indexes are.</p></li> <li><p>equip your server with enough memory so that the collection will remain entirely in memory. The collection size is about 26 GB so you would need at least that much memory for this approach:</p> <blockquote> <p>db.results.stats() "size" : 26662218768,</p> </blockquote></li> </ul> <p>Hope this helps,</p> <p>Bruce</p>
    singulars
    1. This table or related slice is empty.
    1. POMongoDB: very slow queries inside index
      singulars
      1. PTQuestion
    1. PTAnswer
    1. This table or related slice is empty.
    1. USBruce Lucas
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COThanks for your really helpful answer. Documents are big because I save in some of them a lot of text content (not in all) - maybe I should move them do another collection and only save references. But as I commented above, I found real reason of slow explain and that's why my question is resolved for now (time of the explain is not time of the real query).
      singulars
      1. PO
      1. USSerge
    2. COKeep in mind as you interpret your timing experiments that you will see different times for the same operation depending on whether the information is currently in memory. For example, if you do query that returns 600MB of results, the first time it will be slow because it has to read the 600MB from disk, but the second time it will be fast because that 600MB will be in memory. The same may apply when you compare explain() results and find() results.
      singulars
      1. PO
      1. USBruce Lucas
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload