在做 node 爬虫。

目前用的是 BulkWrite :

const Book = mongoose.model('Book', bookSchema);
exports.saveAll = (from_en, books) => {
    const bulkWrite = books.map(book => ({
        replaceOne: {
            filter: {
                from_en,
                originId: book.originId
            },
            replacement: book,
            upsert: true
        }
    }))
    return Book.bulkWrite(bulkWrite).catch(error => console.error(error))
}

然后发现，这么处理 11200 条数据耗时 600s：

catId: 82 from 5040 to 5600. crawl cost: 10.1min, dataTotal: 11200, upsertTotal: 11000, matchTotal: 200
mongodb is disonnected
mongodb: 603757.883ms
✨  Done in 604.47s.

这个该如何优化？

下面是部分爬虫逻辑的代码： while 内部的代码

机子性能：I7 6700HQ / 16G RAM