I suggest taking a look at node.io (https://github.com/chriso/node.io) - a scraping framework written for NodeJS. It uses htmlParser rather than jsdom and scales nicely. It also has support for handling timeouts, retries, etc.