[PYTHON] Mongodb Shortest Introduction (3) I tried to speed up even millions

Mongodb Shortest Introduction (1) Install & start on EC2 & suddenly put tens of thousands of items continued The shortest introduction to Mongodb (2) I searched for tens of thousands of cases.

Now that we have an API, let's try a lot more. As expected, t2.micro is impossible, so go to m3.medium. It became a tricycle → a bicycle.

After all it is slow if you put a lot, so I will put an index.

So far, it's the minimum as a casual DB. Searching for files is indexing.

once again. Useful at the development stage.

When refilling everything.

sudo service mongod stop

Did you stop? It looks like Mysql's innoDB, but it seems that journaling files are getting bigger and bigger. Even if you erase the data, something seems to grow under / var / lib / mongo. It's pretty big data, so it's good to have a clean state every time you develop. I'll try again with ** Crispy, Tturn! **. (Yoiko must not imitate)

sudo rm -rf /var/lib/mongo/*

Delete the files under / var / lib / mongo /. Keep the directory.

Start up

sudo service mongod start

The empty / var / lib / mongo / has been initialized.

Erase by collection (Click here for Yoiko)

db.honyarara.remove({});

Farewell.

The main subject. mongotan has an index.

https://docs.mongodb.org/manual/indexes/

(Assumption that data is not entered or deleted again)

I'll fit it all. When there is nothing to look for.

db.honyarara.createIndex( { "$**": "text" } )

When the text type and numbers are separated.

--Ordinary index

db.honyarara.createIndex ({"column name": 1}) value is 1 ascending order-1 descending order

--Text type (when it is a language)

db.honyarara.createIndex ({"column name": "text", "column name 2": "text", ....})

It can be all, or it can be specified

Click here for the text index ↓ https://docs.mongodb.org/manual/core/index-text/#create-text-index

Now, in most cases it should be faster. It takes time, but it's better to be quick.

When it's still late

If you have more complex queries or tunings, we recommend that you no longer use mongodb.

After all when quitting.

Start over with the above procedure
Delete only the index (Click here for Yoiko)

Check current index and name

db.honyarara.getIndexes()

After checking the "name" field, specify and delete

db.tracks.dropIndex (index name)