AppNexus operated about 1,500 global bidding server instances
Catherine Williams · Michael Rubenstein and Catherine Williams, App Nexus // Data Driven #31 // Nov 2014 · Nov 20, 2014 · at 8:23
Catherine Williams, Head of Data Science at AppNexus, describes the scale of the company's ad tech infrastructure during a presentation at Data Driven NYC.
0:00 / 0:05exact quote · 5.7s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“These bidding instances, there's actually about 1500 of them spread in data centers around the world”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Catherine Williams
Product teams lacking data science, engineering, or product management are unstable
“The data science product engineering trifecta was just a very, very powerful combination. And when you were lacking one, it didn't, it was unstable.”
AppNexus leveraged Google's trillion-word corpus for domain word-splitting algorithms
“The data here is the Google trillion word corpus, right, where they scrape the entire internet, look for the frequencies of all the different strings, including misspellings, right, and that basically gives you a probability of any given string being, like, a …”
AppNexus processes 30 billion daily impressions on a 16-node Hadoop cluster
“We have a 16 node Hadoop cluster currently, and I have on my proposed budget for 2015, a 200 node Hadoop cluster so that we can really get our hands on all that raw data of the thirty billion impressions we're transacting daily.”