Privad: Practical Privacy in Online Advertising

Abstract

Online advertising is a major economic force in the Internet today. Today’s deployments, however, erode privacy and degrade performance as browsers wait for ad networks to deliver ads. We present Privad, a practical private online advertising system. Privad serves ads from the endhost; this is attractive from three standpoints — privacy, profit, and performance. Tracking the user’s profile on their computer and not at a third-party improves privacy. Better targeting and potentially lower operating costs can improve profits. And relying more on the local endhost rather than a distant central third-party improves performance. We have implemented Privad and deployed it on a small scale. This paper focuses on the scalability aspects of Privad. It describes the Privad architecture and protocols, analyzes its scalability characteristics, and presents measurements that substantiate our scalability claims. 1.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 197 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.