Proceedings of the Applied Networking Research Workshop 2019
DOI: 10.1145/3340301.3341123
|View full text |Cite
|
Sign up to set email alerts
|

A performance perspective on web optimized protocol stacks

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
15
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
3
2
1

Relationship

1
5

Authors

Journals

citations
Cited by 24 publications
(17 citation statements)
references
References 14 publications
2
15
0
Order By: Relevance
“…All the tests in the tsA have the maximum requests success rate, thus not being pictured. The achieved results demonstrate that HTTP/3 provides better performance in networks with lossy and with delay variations, which is in line with results achieved by other authors [21][22][23] in previous versions of QUIC protocol.…”
Section: F I G U R Esupporting
confidence: 90%
See 3 more Smart Citations
“…All the tests in the tsA have the maximum requests success rate, thus not being pictured. The achieved results demonstrate that HTTP/3 provides better performance in networks with lossy and with delay variations, which is in line with results achieved by other authors [21][22][23] in previous versions of QUIC protocol.…”
Section: F I G U R Esupporting
confidence: 90%
“…The performance of the HTTPs is affected by the size and frequency of objects being transferred, and by the network conditions, where HTTP/3 gains a clearer advantage in scenarios with high latency. Such results, considering the PLT metric, are also achieved by other researchers, 21,22 which put in evidence that HTTP/3 can be considered as a choice for networks with considerable latency.…”
Section: Related Worksupporting
confidence: 68%
See 2 more Smart Citations
“…Since there is no support for TLS 1.3 early-data in Chromium (as of June 2019) and a limited and challenging deployment of TFO in the Internet, this still gives a 1-RTT advantage of QUIC over TCP+TLS+HTTP/2. In [24], we found that this 1-RTT advantage is the primary factor for QUIC outperforming the traditional Web stack in non-lossy environments when looking at technical metrics. QUIC, similar to early-data and TFO, suffers from replay attacks, which have an especially large surface in distributed clusters [5] when requests are non-idempotent.…”
Section: Repeatable Protocol Performance Evaluationsmentioning
confidence: 88%