ComScore's Big Data Deployment In Detail - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
Software // Information Management
Commentary
11/24/2010
11:04 AM
Doug Henschen
Doug Henschen
Commentary
Connect Directly
Google+
LinkedIn
Twitter
RSS
E-Mail
50%
50%

ComScore's Big Data Deployment In Detail

Crunching tens of terabytes, a leading digital marketing measurement firm delivers insight -- quickly.

When your slogan is "we measure the digital world," you better have serious data-crunching capabilities to back up your claims.

comScore has been delivering big-data intelligence for ten years, and a recent refresh of its data warehousing environment is expected to take the company well into another decade.

Read the stream of press releases churned out by comScore and you'll quickly understand its business. One day it might report on the top-50 Web sites on the Internet. The next it might divulge the market shares of the Internet's leading search engines.

comScore reports on markets across the globe, detailing, for example, the number of mobile phone users in the top-five EU countries receiving SMS-based advertisements -- answer: more than 100 million.

In short, comScore sells data, and the faster it can collect and crunch high-volume samples of Internet and mobile-device usage data, the more numerous and valuable the company's insights become.

"If we can sell last-month's data this month, it has a certain value, but if we can sell last week's data this week, it has even more value," explains Scott Smith, comScore's vice president of data warehousing.

To enhance that turnaround time, Smith has overseen multiple refreshes of comScore's data warehousing platform over the last 10 years. All of them have run on the Sybase IQ column-store database.

comScore's first deployment, way back in 2000, was built on eight Dell severs. The latest, installed in March 2010, runs on 12 Dell R710 servers and an EMC storage area network (SAN).

"I've gone through almost the entire Dell rack server line, from the old 6800s to today's R710s," says Smith. "We try to keep it to $10,000 a pop so you can bring another [database] reader or writer onto a node."

comScore had 56 terabytes of compressed user data on its old deployment while the new system can handle up to 150 terabytes with plenty of room to grow. Smith says he's already planning to add 10 more R710 servers for processing power, and the SAN can grow as needed. Sybase IQ's Multiplex grid lets comScore scale up processing power and storage incrementally and independently.

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Previous
1 of 2
Next
Comment  | 
Print  | 
More Insights
Slideshows
10 Top Cloud Computing Startups
Cynthia Harvey, Freelance Journalist, InformationWeek,  8/3/2020
Commentary
How Enterprises Can Adopt Video Game Cloud Strategy
Joao-Pierre S. Ruth, Senior Writer,  7/28/2020
Commentary
Conversational AI Comes of Age
Guest Commentary, Guest Commentary,  8/7/2020
White Papers
Register for InformationWeek Newsletters
Video
Current Issue
Special Report: Why Performance Testing is Crucial Today
This special report will help enterprises determine what they should expect from performance testing solutions and how to put them to work most efficiently. Get it today!
Slideshows
Flash Poll