promise-bulk
v1.0.3
Published
Write efficient bulk algorithms without modifying your existing codebase.
Downloads
8
Maintainers
Readme
Write efficient bulk algorithms without modifying your existing codebase.
const app = express()
app.post('/comments', async (req, res) => {
const bodyComment = req.body
const insertedComment = await insertBulkInstance.execute(bodyComment)
res.json(insertedComment)
})
What is insertBulkInstance
?
const insertBulkInstance = new PromiseBulk((commentsArr) => {
return db.collection('comments').insertMany(commentsArr)
})
Allows to insert items in a atomic way but doing a efficient bulk operation in the background.
When I should use this?
Imagine you have two functions:
insertInDatabaseOneItem
insertInDatabaseManyItemsAtOnce
What do you think is more efficient: call the first function 100 times or the second function one time with 100 items?
Depending on your database system, probabily the second option is much more efficient.
So, whats the problem?
Our existing logic is frequently written for 1 item so in many scenarios is complex to write it for many items.
This library allows to easily convert the calls from one item (insertInDatabaseOneItem
in the example) into calls of many items (insertInDatabaseManyItemsAtOnce
in the example).
Install
npm install --save promise-bulk
or
yarn add promise-bulk
or
pnpm add --save pnpm
Documentation
Detailed information and documentation about this library and how to use it can be found in docs.
Example
We have the scenario of a comments platform that receives so many comments per second using a REST API. Every REST API call inserts a SINGLE comment into the database.
With 100 users sending comments per second we are executing 100 database.insertOne(comment)
in parallel.
Despite of being more efficient to insert all the comments at once, refactoring this code to do database.insertMany(commentsList)
is hard to implement.
With this library we can convert the atomic insert of one comment into a grouped insert of many comments.
Atomic:
And with promise-bulk
:
The example total time is near 2x times faster in bulk scenario.
Real use case
See examples/mongo/README.md for real results and performance.