How to build your own public key infrastructure

Published on by Nick Sullivan.

A major part of securing a network as geographically diverse as CloudFlare’s is protecting data as it travels between datacenters. Customer data and logs are important to protect but so is all the control data that our applications use to communicate with each other. For example, our application servers need to securely communicate with our new datacenter in Osaka, Japan. CC BY-SA 2.0 image by kris…

Osaka, Japan: CloudFlare's 35th Data Center

Published on by Joshua Motta.

Move over Jurassic World, the long awaited sequel to our Tokyo deployment is here. Our Osaka data center is our 2nd in Japan, 5th in Asia (following deployments in Hong Kong, Tokyo, Singapore, and Seoul), and 35th globally. This latest deployment serves not only Osaka, Japan's second largest city, but also Nagoya, the 3rd largest, and the entire Keihanshin metropolitan area including Kyoto and Kobe. This means faster…

EFF, CloudFlare Ask Federal Court Not To Force Internet Companies To Enforce Music Labels’ Trademarks

Published on by Kenneth R. Carter.

This blog was originally posted by the Electronic Frontier Foundation who represents CloudFlare in this case. JUNE 18, 2015 | BY MITCH STOLTZ This month, CloudFlare and EFF pushed back against major music labels’ latest strategy to force Internet infrastructure companies like CloudFlare to become trademark and copyright enforcers, by challenging a broad court order that the labels obtained in secret. Unfortunately, the court denied CloudFlare’s challenge and…

Go has a debugger—and it's awesome!

Published on by Filippo Valsorda.

Something that often, uh... bugs[1] Go developers is the lack of a proper debugger. Sure, builds are ridiculously fast and easy, and println(hex.Dump(b)) is your friend, but sometimes it would be nice to just set a breakpoint and step through that endless if chain or print a bunch of values without recompiling ten times. CC BY 2.0 image by Carl Milner You could…

How to receive a million packets per second

Published on by Marek Majkowski.

Last week during a casual conversation I overheard a colleague saying: "The Linux network stack is slow! You can't expect it to do more than 50 thousand packets per second per core!" That got me thinking. While I agree that 50kpps per core is probably the limit for any practical application, what is the Linux networking stack capable of? Let's rephrase that to make it more…