We want to inform the community that the grant committee has approved the following grant request on May 26th, 2025:
University of Lisbon - Practical performance-anonymity tradeoffs in gossip protocols
Total grant: $21000
Thank you 
—Grants Committee
Progress update — SPREAD: Extending GossipSub with Efficient Anonymous Dissemination
Hi all — a progress report on the work funded under this grant 
Our main outcome is SPREAD (Secure Peer-to-Peer Relay for Efficient
Anonymous Dissemination), a GossipSub extension that targets exactly the
tradeoff the grant defined, getting the best of both worlds:
it raises the bar against sender deanonymization (the passive, timing-based attacks that can map validators to their peer IDs and IP addresses)
while improving dissemination efficiency rather than trading it away.
This contrasts with prior anonymity designs such as Dandelion++, whose high latency cost has kept them out of Ethereum’s consensus layer.
SPREAD combines a local random walk (for anonymity) with a topology-aware,
geographically directed propagation (for low-latency global reach). We evaluated it
against the real GossipSub go-libp2p-pubsub implementation and against
Dandelion++. It lowers mean stretch by ~23% vs GossipSub (and ~67% vs
Dandelion++) and reduces deanonymization accuracy relative to
GossipSub, while also considerably improving tail behavior and reaching distant peers quickly.
The work is public in three places:
A full paper is on the way and we’ll share it here once it’s out. We’re also actively engaging with the networking teams in the Ethereum ecosystem toward real-world adoption.
Thank you to the SSV Grants Committee for supporting this research. 