Cloudmark combats messaging threats

With messaging systems being overrun by malicious, spurious and criminal attacks, e-mail in particular is becoming as inconvenient and much less secure than its voice mail counterpart. To help stem the tide, Cloudmark introduced an automated feedback system last week to help service providers enlist their messaging customers.

The Cloudmark Network Feedback System is a reporting and information-sharing framework that automatically processes reports from end users about potentially threatening messages and gives service providers the ability to measure and demonstrate the framework's success in identifying and blocking spam, phishing and virus attacks.

Ferris Research estimated in a recent report that the cost of cleaning up after a virus is about $200 per infected system. However, the cost of transporting and storing spam is escalating, with estimates that up to 90% of messaging traffic traversing public and private networks is spam and up to 25% of stored mail is spam.

Although plenty of protection methods already exist — such as heuristic detection, overlapping anti-virus products, attachment blocking, patching and the use of virtual machines — it hasn't been enough. Many of these leave a window of vulnerability between the time a threat is identified and when it is neutralized, usually via downloadable patching.

Ferris also said in November 2005 that the window — it calls it the zero-hour problem, which is the time it takes to identify the signature of a threat — can be several hours and is not likely to be improved upon.

However, Cloudmark said its zero-hour response time is now between 20 seconds and 3 minutes. The new feedback system completes the loop by providing feedback to end users on how their efforts are contributing in the war to ensure threat-free services.

“To some service providers, anti-abuse solutions just look like a cost,” said Jamie de Guerre, technical director of program management for Cloudmark. “But with the reports, they can see all the work behind the scenes, and users can see when they have saved their company money by blocking spam.”

The system also gives users and service providers feedback on their “tracking reputation.” In other words, it lets them know how accurate they are in reporting real threats and forms a trust rating that the system can use to speed up the process.

De Guerre said the feedback also encourages users to fight back against threats as a way to “get back at spammers.”

Since the company's founding in 2001, it has grown to protect more than 100 million mailboxes. Fifty percent of the company's business comes from the service provider market, including deployments at Cable & Wireless, Cincinnati Bell, Cox Communications, NTT Communications, Vodafone and Virgin Mobile. The company has conducted 30 service provider trials in the last 21 months.

Richi Jennings, lead analyst for the e-mail security practice at Ferris, said the challenge for this and other spam filters are false positives, or the accidental deletion of legitimate mail such as newsletters or other bulk mailings.

“The concern when relying 100% on users flagging spam is whether people are making the correct choices,” Jennings said. “Spam can be a grey area, and as they say, ‘One man's spam is another man's ham.’”

He said that the accuracy of these systems is two-dimensional: How much you successfully filter out versus how much legitimate e-mail gets accidentally or intentionally filtered out.

Jennings also said the Cloudmark guys are pretty smart about the way they weight the information they get from users. “They have this concept of karma that gives them a pretty good idea of how accurate each user's determinations are.”

FAST FACTS ABOUT MESSAGING SECURITY

• 85% to 90% of e-mail is spam.¹
• More than 50% of spam is sent by compromised hosts.¹
• Fixed-line spam cost U.S. & European operators about $500 million per year.²
• Successful phishing in the U.S. has cost 2.4 million adults almost $1 billion.³
• 20% to 25% of e-mail storage is comprised of spam.¹
• 14 new viruses are launched every day.¹

¹Source: Cloudmark ²Source: Ferris Research, ³Source: Gartner