Recording Venue: QCon 2007
Guest(s): Randy Shoup
Host(s): Markus
In this episode we discuss with Randy Shoup, Distinguished Architect at eBay, about architectural pinciples and patterns used for building the highly scalable eBay infrastructure. The discussion is structured into four main ideas: partition everything, use asynchrony everywhere, automate everything, and design the system keeping in mind that everything fails at some point in a large distributed system.
SE Radio 109: eBay’s Architecture Principles with Randy Shoup
Join the discussion
You must be logged in to post a comment.
12 comments
More from this show
Matthew Skelton joins host Giovanni Asproni to talk about team topologiesâan approach to organizing teams for fast flow of value. The...
Vinay Tripathi, a senior network engineer in Google Backbone Engineering and an 18-year network engineering veteran, discusses BGP...
Tim McNamara, a well-known Rust educator, author of Rust in Action (Manning), and a recipient of a Rust Foundation Fellowship in 2023...
A very interesting listen. Wonderful podcast as usual.
I like this episode, the content was really interesting and great,
also the way that the episode is structured is very efficient.
Apart from these insightfull ideas about transactions,
I was reminded very much of erlang. I wonder if it would make a diffrence if ebay used erlang.
It also seems like even ebay is struck on some legacy JSP pages đ
Also ebay seems to have an interesting development methodology.
bis denn – sven
Very interesting talk. I liked to listen again to the episode with Werner Vogels, as there are several aspects in common (scalability, etc).
I like all these “practical” episodes (such as Amazon, eBay, the Guardian…) but would it be possible to talk about some other applications than Internet? We could enjoy some concrete example of medical or safety-critical applications…
Thank you for this excellent content!
… with the more practical episodes. It is on our radar to do more practical stuff, also outside the web and enterprise stuff (personally, I am more interested in these thigns). But it’s not as easy to get those people in front of the mic.
We are working on it!
Markus
I meant to correct this before, but of course, a petabyte is 10^15 bytes, not 10^12 bytes ;-). That is what I get for trying to reason it out in the middle of an interview!
Thanks very much to everyone who found this episode useful. Markus and crew, keep up the great work!
— Randy
Thanks a lot they were really very good links . good job dude .
It is a big mystery for me when I start thinking about the scalability of eBay after reading this article. Really canât imagine the volume handled. [26 billion SQL queries, 100 million items available, 212 million registered users, 1 billion photos, 1 billion page views a day, 2 petabytes of data]. Awesome, just keep going
Hope Randy Shoup and his co-workers have done a extraordinary work and this should be a reference point for some upcoming projects. Their statistical data of traffic per day really shocks me. The word saturation should be taken out of dictionary seeing this.
Building on previous QCon presentations about eBay’s architecture, this session will concentrate on how eBay leverages the power of the extended eBay community to improve the user experience in a systematic and automated way, combining both real-time and analytical approaches.
I had some cool browse with e-bay, but never thought of the volume they handle. Great job done by Randy Shoup and his team, I am just carried away by the way they handle this volume. I guess this volume would increase further in proportion with net users.
Wow, I knew that eBay was big but I had never imagined that it was so big. Who hasn’t wondered how eBay does their business? 100 million items available for purchase…that is really massive. Not to mention the amount of page views.
[…] http://www.se-radio.net/2008/09/episode-109-ebays-architecture-principles-with-randy-shoup/ […]