Feeds

Yahoo! open sources uber web server

'400-terabytes-a-day' Traffic Server lands at Apache

Secure remote control for conventional and virtual desktops

Yahoo! has open sourced the back-end software platform that underpins the company's webmail client and countless other applications offered up across its sweeping web portal.

Known as Traffic Server, the platform handles general edge caching, edge processing, and load balancing at Yahoo!, but it's also used to manage traffic on the company's internal storage and server-virtualization services.

"It's used at the edge, but it also gets used almost like an application server," Chuck Neerdaels, Yahoo! vp of content storage, delivery, and edge, tells The Reg. It serves up the latest version of Yahoo! Mail, for instance, which the company calls "Candy Gram."

Acquired with Yahoo!'s purchase of Inktomi, Traffic Server has been in active use at the two companies for the past eight years. According to Yahoo!, it now handles 30,000 requests per second, serving 30 billion Web objects and 400 terabytes of data a day.

"It is a very mature, very reliable piece of technology," says Shelton Shugar, Yahoo!'s senior vp of cloud computing. "In some form, it supports more than half of Yahoo!'s traffic."

The company donated a version of the platform to The Apache Software Foundation last week through the Apache Incubator program. "This is part of our overhaul strategy to open source cloud services that are mature and not laden with Yahoo!-specific stuff that wouldn't make sense for open source," says Shugar.

In June, the company open sourced its internal Hadoop distro, an internet-scale distributed data-crunching platform based the Apache project of the same name.

Neerdaels says Traffic Server was originally designed as a proxy cache. But it's Yahoo!'s go-to tool for http session management, and it includes an API for tweaking content all the way down at the protocol level. "You can poke around with various headers and inject content and direct client requests to different backends, all through a relatively clean API," he says.

Neerdaels and Shuger also describe Traffic Server as an "extensible framework" that lets you tweak the architecture according to the task at hand. And, according to the company, it suits web operations both large and small. "It's simple enough for a small operation to pick it up quickly," says Neerdaels. "But the big players can pick it up and it will scale to meet their needs pretty impressively."

Those 400 terabytes are served up from between 100 and 150 "commodity" machines. ®

Beginner's guide to SSL certificates

Whitepapers

Go beyond APM with real-time IT operations analytics
How IT operations teams can harness the wealth of wire data already flowing through their environment for real-time operational intelligence.
Forging a new future with identity relationship management
Learn about ForgeRock's next generation IRM platform and how it is designed to empower CEOS's and enterprises to engage with consumers.
10 threats to successful enterprise endpoint backup
10 threats to a successful backup including issues with BYOD, slow backups and ineffective security.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Security and trust: The backbone of doing business over the internet
Explores the current state of website security and the contributions Symantec is making to help organizations protect critical data and build trust with customers.