由于您似乎熟悉 ORM 模式,我建议您使用“mongoose”模块。虽然我猜想你会在 NodeJS 和 Mongo 方面有很长的学习曲线来制作一个可靠的应用程序。  
这是一个可以帮助您入门的工作示例:
#! /usr/bin/node
var mongoose = require('mongoose');
mongoose.connect('localhost', 'test');
var fs = require('fs');
var lineList = fs.readFileSync('mytest.csv').toString().split('\n');
lineList.shift(); // Shift the headings off the list of records.
var schemaKeyList = ['RepName', 'OppID', 'OppName', 'PriorAmount', 'Amount'];
var RepOppSchema = new mongoose.Schema({
    RepName: String,
    OppID: String,
    OppName: String,
    PriorAmount: Number,
    Amount: Number
});
var RepOppDoc = mongoose.model('RepOpp', RepOppSchema);
function queryAllEntries () {
    RepOppDoc.aggregate(
        {$group: {_id: '$RepName', oppArray: {$push: {
            OppID: '$OppID', 
            OppName: '$OppName',
            PriorAmount: '$PriorAmount',
            Amount: '$Amount'
            }}
        }}, function(err, qDocList) {
        console.log(util.inspect(qDocList, false, 10));
        process.exit(0);
    });
}
// Recursively go through list adding documents.
// (This will overload the stack when lots of entries
// are inserted.  In practice I make heavy use the NodeJS 
// "async" module to avoid such situations.)
function createDocRecurse (err) {
    if (err) {
        console.log(err);
        process.exit(1);
    }
    if (lineList.length) {
        var line = lineList.shift();
        var doc = new RepOppDoc();
        line.split(',').forEach(function (entry, i) {
            doc[schemaKeyList[i]] = entry;
        });
        doc.save(createDocRecurse);
    } else {
        // After the last entry query to show the result.
        queryAllEntries();
    }
}
createDocRecurse(null);
您在“mytest.csv”中的数据:
Rep Name,Opp ID,Opp Name,Prior Amount,Amount
Rep 1,1234561,Opp 1,10000,8000
Rep 1,1234562,Opp 2,15000,9000
Rep 2,1234563,Opp 3,20000,10000
Rep 1,1234564,Opp 4,25000,11000
Rep 2,1234565,Opp 5,30000,12000