Scholarlink

A 2016 study by Wei and Heidemann showed that anycast routing of DNS queries to root name servers is fairly stable, with only 1% of RIPE Atlas vantage points “flipping” back and forth between different root name server sites. Continuing this study longitudinally, however, we observe that among the vantage points that collected data continuously from 2016 to 2024 the fraction that experience flipping has increased from 0.8% to 3.2%. Given this apparent increase, it is natural to ask how much anycast flipping impacts the performance of everyday tasks such as web browsing. To measure this impact, we established a mock web page incorporating many embedded objects on an anycast-based CDN and downloaded the page from geographically distributed BrightData vantage points. We observed that packets within individual TCP flows almost always reach the same site, but different flows may flip to different sites. We found that 1,988 (10.9%) of 18,294 <vantage point_="" anycast="" IP=""> pairs suffer from frequent anycast flipping (i.e., a vantage point is directed to sites other than the most visited one of the anycast IP more than 50% of the time) and that 1,030 of these (5.6% of the total) suffer a median increase in round-trip time larger than 50&#xa0;ms when directed to a site other than the most visited one. We then used Mahimahi to emulate downloads of popular web sites, randomly applying the above-mentioned flipping probability (50%) and flipping latency penalty (50&#xa0;ms) to CDN downloads. We found, for example, that there was a median increase in the First Contentful Paint metric ranging, across 3 vantage points and 20 web sites, from 20.7% to 52.6% for HTTP/1.1 browsers and from 18.3% to 46.6% for HTTP/2 browsers. These results suggest that for a small, but not negligible portion of clients, the impact of anycast flipping on web performance may be significant.</vantage>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Characterizing Anycast Flipping: Prevalence and Impact

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xiao Zhang,
  • Shihan Lin,
  • Tingshan Huang,
  • Bruce M. Maggs,
  • Kyle Schomp,
  • Xiaowei Yang

摘要

A 2016 study by Wei and Heidemann showed that anycast routing of DNS queries to root name servers is fairly stable, with only 1% of RIPE Atlas vantage points “flipping” back and forth between different root name server sites. Continuing this study longitudinally, however, we observe that among the vantage points that collected data continuously from 2016 to 2024 the fraction that experience flipping has increased from 0.8% to 3.2%. Given this apparent increase, it is natural to ask how much anycast flipping impacts the performance of everyday tasks such as web browsing. To measure this impact, we established a mock web page incorporating many embedded objects on an anycast-based CDN and downloaded the page from geographically distributed BrightData vantage points. We observed that packets within individual TCP flows almost always reach the same site, but different flows may flip to different sites. We found that 1,988 (10.9%) of 18,294 pairs suffer from frequent anycast flipping (i.e., a vantage point is directed to sites other than the most visited one of the anycast IP more than 50% of the time) and that 1,030 of these (5.6% of the total) suffer a median increase in round-trip time larger than 50 ms when directed to a site other than the most visited one. We then used Mahimahi to emulate downloads of popular web sites, randomly applying the above-mentioned flipping probability (50%) and flipping latency penalty (50 ms) to CDN downloads. We found, for example, that there was a median increase in the First Contentful Paint metric ranging, across 3 vantage points and 20 web sites, from 20.7% to 52.6% for HTTP/1.1 browsers and from 18.3% to 46.6% for HTTP/2 browsers. These results suggest that for a small, but not negligible portion of clients, the impact of anycast flipping on web performance may be significant.