Inferring Spammers in the Network Core

Abstract

Despite a large amount of effort devoted in the past years trying to limit unsolicited mail, spam is still a major global concern. Content-analysis techniques and blacklists, the most popular methods used to identify and block spam, are beginning to lose their edge in the battle. We argue here that one not only needs to look into the network-related characteristics of spam traffic, as has been recently suggested, but also to look deeper into the network core, to counter the increasing sophistication of spammers. At the same time, local knowledge available at a given server can often be irreplaceable in identifying specific spammers.

To this end, in this paper we show how the local intelligence of mail servers can be gathered and correlated passively, scalably, and with low-processing cost at the ISP-level providing valuable network-wide information. First, we use a large network flow trace from a major national ISP, to demonstrate that the pre-filtering decisions and thus spammer-related knowledge of individual mail servers can be easily and accurately tracked and combined at the flow level. Then, we argue that such aggregated knowledge not only allows ISPs to monitor remotely what their “own” servers are doing, but also to develop new methods for fighting spam.