Old but Gold: Prospecting TCP to Engineer DNS Anycast (extended)

Old but Gold: Prospecting TCP to Engineer DNS Anycast (extended)
复制标题

老而黄金:探索 TCP 来设计 DNS 选播(扩展)

DOI:
--
复制
发表时间:
2020
期刊:
影响因子:
--
通讯作者:
Cristian Hesselman
Cristian Hesselman
中科院分区:
--
文献类型:
--
作者:
G. Moura;J. Heidemann;W. Hardaker;Jeroen Bulten;J. Ceron;Cristian Hesselman

文献摘要

参考文献

被引文献

相似文献

DNS延迟是许多服务运营商关注的问题:cdn的存在是为了减少对最终用户的服务延迟,但必须依赖全局DNS来实现可达性和负载平衡。我们展示了递归DNS解析器对低延迟的偏好在tld和DNS根转移流量。目前,DNS延迟是通过分布式基础设施(如RIPE Atlas)或使用Verfploeter进行主动探测来监控的。虽然Atlas覆盖范围很广,但它是不完整的,而Verfploeter在IPv6中的覆盖是有限的。在本文中,我们展示了被动观察TCP握手提供了一种测量DNS延迟的机制。从TCP进行被动RTT估计是一个老想法,但它以前从未用于检查DNS。我们表明,今天有足够的TCP DNS流量来提供比现有方法更大的覆盖范围,并且是使用IPv6观察DNS延迟的最佳方法。我们表明,从TCP估计的DNS延迟与UDP延迟是一致的。我们的方法发现了真正的问题:我们定义了DNS极化,这是一个新问题,一个超巨型将全球流量发送到一个任意播送站点,而不是利用全球任意播送部署——我们发现谷歌流量极化,并将其延迟从100毫秒减少到10毫秒,对于微软来说,由于流量去极化,延迟从90毫秒减少到20毫秒。我们的方法在一个欧洲国家的顶级域名中得到了实际应用,使用我们的工具进行监控有助于找到并纠正将欧洲流量发送到澳大利亚的路由绕道。
DNS latency is a concern for many service operators: CDNs exist to reduce service latency to end-users, but must rely on global DNS for reachability and load-balancing. We show that a recursive DNS resolver’s preference for low latency shifts traffic at TLDs and the DNS root. DNS latency today is monitored with distributed infrastructure such as RIPE Atlas, or with active probing using Verfploeter. While Atlas coverage is wide, it is incomplete, and Verfploeter coverage in IPv6 is limited. In this paper we show that passive observation of TCP handshakes provides a mechanism to measure DNS latency . Passive RTT estimation from TCP is an old idea, but it has never been used to examine DNS before. We show that there is sufficient TCP DNS traffic today to provide greater coverage than existing approaches, and is the best method to observe latency of DNS using IPv6. We show that estimates of DNS latency from TCP is consistent with UDP latency. Our approach finds real problems: We define DNS polarization , a new problem where a hypergiant sends global traffic to one anycast site rather than taking advantage of the global anycast deployment—we found Google traffic polarized and cut its latency from 100 ms to 10 ms, and for Microsoft, the latency cut due to traffic being depolarized was from 90 ms to 20 ms. Our approach is in operational use for a European country’s top-level domain, and monitoring with our tool helped find and correct a routing detour sending European traffic to Australia.
DOI: 10.1145/3365609.3365865
发表时间: 2019-11
期刊: Proceedings of the 18th ACM Workshop on Hot Topics in Networks
影响因子: --
作者:
T. Arnold;Matt Calder;Ítalo F. S. Cunha;Arpit Gupta;H. Madhyastha;Michael Schapira;Ethan Katz-Bassett
通讯作者: T. Arnold;Matt Calder;Ítalo F. S. Cunha;Arpit Gupta;H. Madhyastha;Michael Schapira;Ethan Katz-Bassett
如果可以的话缓存我:DNS 生存时间的影响
DOI: 10.1145/3355369.3355568
发表时间: 2019
期刊: ACM Interent Measurements Conference
影响因子: --
作者:
Moura, Giovane C.;Heidemann, John;Schmidt, Ricardo de;Hardaker, Wes
通讯作者: Hardaker, Wes
DOI: 10.1145/3379479
发表时间: 2020
期刊: Proceedings of the ACM on Measurement and Analysis of Computing Systems
影响因子: --
作者:
Zhu, Pengxiong;Man, Keyu;Wang, Zhongjie;Qian, Zhiyun;Ensafi, Roya;Halderman, J. Alex;Duan, Haixin
通讯作者: Duan, Haixin