Abstract

If by accident “google.com” would not send users to Google servers but to your personal webserver, it would break down under traffic after a few seconds. A minute later, your internet service provider would go offline. If they are unlucky, even their upstream network provider would fold under the load.

10⁹ users impose a daily traffic load on popular Google services that can be best described as distributed denial of service attack. Coping with that torrent of user requests is where the story of Reliability Engineering begins. It is a story of keeping the site up under all peek traffic, while upgrading the software stack, while upgrading the network fabric, and replacing broken hardware without a user noticing, 24/7.

This talk is about distributed software architectures, their failure modes, and why load balancers need load balancers themselves. The talk concludes with a brief discussion on how to scale not software-wise but organization-wise, namely Google’s hiring practices for engineers and interns.

Biography

Clemens Fruhwirth is a senior site reliability engineer at Google Zürich, a TU Wien master graduate, and a free software contributor. He is the inventor of LUKS, the most popular harddisk encryption standard with GNU/Linux. At Google, he and his colleagues run systems that invoice 10⁵$ per minute. He likes sub-millisecond latency and his favorite storage unit is a petabyte.

Note

Following the talk, your hot questions can find their answers over a cold buffet.

Speakers

• Clemens Fruhwirth, Senior site reliability engineer at Google Zürich

Links

• Google Formular, für alle, die an Infos zu Veranstaltungen und Jobs bei google intreressiert sind

Curious about our other news? Subscribe to our news feed, calendar, or newsletter, or follow us on social media.

Note: This is one of the thousands of items we imported from the old website. We’re in the process of reviewing each and every one, but if you notice something strange about this particular one, please let us know. — Thanks!