STAR uses server TEEs for anonymous rate limiting at scale
Original title: STAR: Scalable TEE-based Anonymous Rate-limiter
The design reports authentication in tens of microseconds and about 13.9K requests per second per CPU core.
STAR is a research design for anonymous rate limiting that places trusted execution environments on servers so services can throttle TLS and HTTP floods without identifying users. The authors argue prior schemes either fail to scale or assume services and authorities do not collude. TEEs speed authentication but let a malicious service watch memory-access patterns; STAR adds countermeasures against secret-dependent accesses and checks them with dynamic analysis. A secure sharding mechanism spreads client requests across servers.
- Authentication is reported in tens of microseconds.
- Throughput is about 13.9K requests per second per CPU core.
- Countermeasures target secret-dependent memory accesses observed by a malicious service.
- Sharding is designed to carry the efficiency gains across servers.
From the original article
1 min read at arxiv.orgProtecting Internet services from attacks such as TLS/HTTP floods requires rate limiting, yet identifying users to enforce these limits can compromise their privacy. Anonymous rate limiting promises to reconcile these goals, but existing schemes either struggle to scale or assume that services and authorities do not collude.
By Kosei Akama · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.