How To Deploy Node.js Applications Using Systemd and Nginx

“One could use forever and crontab to take care of this. This tutorial presents a more robust, albeit more complex, solution. Using systemd (available on Arch and Fedora, and CentOS in the future), web applications can be thoroughly managed: logs, uptime, resources and security through cgroups, and advanced daemon startup can all be accessed, controlled and fine-tuned in a unified manner…”

Redis cluster tutorial

“This document is a gentle introduction to Redis Cluster, that does not use complex to understand distributed systems concepts. It provides instructions about how to setup a cluster, test, and operate it, without going into the details that are covered in the Redis Cluster specification but just describing how the system behaves from the point of view of the user.

Note that if you plan to run a serious Redis Cluster deployment, the more formal specification is an highly suggested reading.

Redis cluster is currently alpha quality code, please get in touch in the Redis mailing list or open an issue in the Redis Github repository if you find any issue…”

The Story Of Stackless Python

“This talk gives a good overview of the status of Stackless Python:

Its history from the beginning, its current status and its future development to be expected. A discussion and comparison with similar approaches like Greenlet, Eventlet and how they relate is also included.

Stackless Python 1.0 was started in 1998 as an implementation of true continuations, with all implied complications…”

BitTorrent Makes Twitter’s Server Deployment 75x Faster

“Some of the biggest Internet brands have declared their love for BitTorrent in recent months. Both Facebook and Twitter are using BitTorrent to update their networks and not without success. In Twitter’s new setup the BitTorrent-powered system has made their server deployment 75 times faster than before…”

Walmart Node.js Memory Leak

“A few weeks ago, Eran Hammer of Walmart labs came to the Node.js core team complaining of a memory leak he had been tracking down for months. By expending a lot of effort over those few months he had taken the memory leak from a couple hundred megabytes a day, down to a mere eight megabytes a day. That in and of itself is an impressive feat, however, the source and final portion of the leak was elusive.

So he spent more effort and found a test case that reproduced with only Node APIs, which placed the fault right at Node.js’ doorstep. He found that the more HTTP client requests he did, the more memory his Node process would consume, but it was really slow. Here’s the graph from his initial data:…”