Shulou(Shulou.com)06/01 Report--

It is easy to use Map/Reduce for parallel "statistics" on MongoDB.

Using MapReduce to implement two functions: Map function and Reduce function, Map function calls emit (key, value)

Iterate through all the records in collection, passing key and value to the Reduce function for processing.

The Map function and Reduce function can be implemented using JavaScript, and a MapReduce operation can be performed through the db.runCommand or mapReduce command:

Db.runCommand ({mapReduce:, map:, reduce:, out:, query:, sort:, limit:, finalize:, scope:, jsMode:, verbose:})

Parameter description:

Mapreduce: the collection of targets to operate on.

Map: mapping function (generates a sequence of key-value pairs as arguments to the reduce function).

Reduce: statistical function.

Out: statistical results store collections (if not specified, temporary collections are used, which are automatically deleted when the client is disconnected).

Query: target record filtering

Sort: sort the target records.

Limit: limit the number of target records.

Finalize: final handling function (final collation of the results returned by reduce into the result set)

Scope: import external variables into map, reduce, finalize.

JsMode: whether to convert Bson format between map and reduce execution

Verbose: displays detailed time statistics.

Let's give an example:

Prepare some data:

Next, we will show you how to count the number of students in each class.

Map:

The Map function must call emit (key, value) to return the key-value pair and use this to access the Document currently to be processed.

M = function () {emit (this.classid,1);}

Value can be passed using JSON Object (multiple attribute values are supported).

For example: emit (this.classid, {count:1})

Reduce:

The parameters received by the Reduce function are similar to the Group effect, combining the sequence of keys returned by Map into {key, [value1]

Value2, value3, value...]} is passed to reduce.

R = function (key,values) {var x = 0; values.forEach (function (v) {x + = v}); return x;}

The Reduce function performs "statistical" operations on these values, and the returned results can be returned using JSON Object.

Result:

MapReduce () stores the results in the "students_res" table.

Finalize:

With finalize (), we can further process the results of reduce ().

F = function (key,value) {return {classid:key,count:value}}

Let's recalculate it again and see the results:

Options:

We can also add more control details.

Db.runCommand ({mapreduce: "stu", map:m, reduce:r, out: "stu_res", finalize:f, query: {age: {$gt:10})

You can see that only the data with age > 10 is filtered first, and then counted, so there is no number of age=9.

According to.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.