DNS performance and the effectiveness of caching
IEEE/ACM Transactions on Networking (TON)
King: estimating latency between arbitrary internet end hosts
Proceedings of the 2nd ACM SIGCOMM Workshop on Internet measurment
Diversity in DNS performance measures
Proceedings of the 2nd ACM SIGCOMM Workshop on Internet measurment
DEW: DNS-enhanced web for faster content delivery
WWW '03 Proceedings of the 12th international conference on World Wide Web
Proactive Caching of DNS Records: Addressing a Performance Bottleneck
SAINT '01 Proceedings of the 2001 Symposium on Applications and the Internet (SAINT 2001)
Using bundles for Web content delivery
Computer Networks: The International Journal of Computer and Telecommunications Networking
Inferring relative popularity of internet applications by actively querying DNS caches
Proceedings of the 3rd ACM SIGCOMM conference on Internet measurement
CoDNS: improving DNS performance and reliability via cooperative lookups
OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
DNS prefetching and its privacy implications: when good things go bad
LEET'10 Proceedings of the 3rd USENIX conference on Large-scale exploits and emergent threats: botnets, spyware, worms, and more
Quantitatively analyzing stealthy communication channels
ACNS'11 Proceedings of the 9th international conference on Applied cryptography and network security
Hi-index | 0.00 |
In this paper, we present a novel approach to exploit the relationships among domain names to improve the cache hit rate for a local DNS server. Using these relationships, an authoritative DNS server (ADNS) can piggyback resolutions for future queries as part of the response message for an initial query. The approach improves the cache hit rate as well as reducing the total queries and responses. The approach is particularly attractive because it can be implemented with no changes to the existing DNS protocol. Trace-based simulations show more than 50% of cache misses can be reduced in the best case while straightforward policies, using frequency and relevancy data for an ADNS, reduce cache misses by 25-40% and DNS traffic by 20-35%. These percentages improve if we focus the policies on resource records with smaller authoritative TTLs. We also show improved performance for hybrid approaches that combine the approach with renewal-based approaches. In conjunction with this work we also did a study on current DNS performance for 20 locations in the United States. The outcome of this study is that the current average DNS latency is generally in the range of 200-300 ms, but range from 500 ms to multiple seconds if we look at the 95% response time. Approaches, such as what we propose, that reduce the amount of DNS traffic will improve the overall response time for applications.