Improved Packet Reordering Metrics

Jayasumana, Anura P.; Piratla, Nischal M.; Banka, Tarun; Bare, Abhijit A.; Whitner, Rick

doi:10.17487/rfc5236

Cited by 42 publications

(41 citation statements)

References 10 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In that paper it was shown experimentally that Restored is able to recover several measures of quality of service, among them the following metric [10].…”

Section: Explaining Restored Behaviormentioning

confidence: 99%

“…However, Theorem 1 forces us to reevaluate the real-life applicability of this simple example: since the vast majority of traces used in the experimental benchmarking from [9] had SUS ≤ 3 the "theoretical" inconsistency of RD displayed by the example in Observation 2 is less stringent "in practice": in particular since no two permutations map to the same reordering pattern and, crucially, repeat packets are discarded before computing RD [10], it is no longer that surprising that RD could be largely consistent "in real life", even though theoretically inconsistent.…”

Section: Definition 7 Reorder Density (Rd)mentioning

confidence: 99%

“…In an experimental paper [9] we showed that Restored is able to regenerate sequences similar to the original sequences with respect to several reordering metrics. One metric for which this result is true was reorder density (RD) from [10,15,16]. We found the experimental result for RD paradoxical for the following reason: Restored guarantees that the regenerated trace is (locally) similar to the original sequence for a precise notion called ≡ F B -equivalent (rigorously explained below).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Identifying Almost Sorted Permutations from TCP Buffer Dynamics

Istrate¹

2015

SACS

View full text Add to dashboard Cite

Associate to each sequence A of integers (intending to model packet IDs in a TCP/IP stream) a sequence of positive integers of the same length M(A). The i'th entry of M(A) is the size (at time i) of the smallest buffer needed to hold out-of-order packets, where space is accounted for unreceived packets as well. Call two sequences A, B equivalent (writtenFor a sequence of integers A define SUS(A) to be the shuffled-upsequences reordering measure defined as the smallest possible number of classes in a partition of the original sequence into increasing subsequences. We prove the following result: any two permutations A, B of the same length with SUS(A), SUS(B) ≤ 3 such that A ≡ F B B are identical. The result is no longer valid if we replace the upper bound 3 by 4.We also consider a similar problem for permutations with repeats. In this case the uniqueness of the preimage is no longer true, but we obtain a characterization of all the preimages of a given sequence, which in particular allows us to count them in polynomial time.The results were motivated by explaining the behavior and engineering Restored, a receiver-oriented model of traffic we introduced and experimentally validated in earlier work.

show abstract

“…In that paper it was shown experimentally that Restored is able to recover several measures of quality of service, among them the following metric [10].…”

Section: Explaining Restored Behaviormentioning

confidence: 99%

Section: Definition 7 Reorder Density (Rd)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Almost Sorted Permutations from TCP Buffer Dynamics

Istrate¹

2015

SACS

View full text Add to dashboard Cite

show abstract

“…It was first introduced by Piratla et al [133] and later specified in RFC 5236 [85]. To compute the reorder density M RD , the sequence number of each packet is compared to the position in which it arrived at the destination.…”

Section: Reorder Density and Entropymentioning

confidence: 99%

Multipath aggregation of heterogeneous access networks

Kaspar¹

2012

SIGMultimedia Rec.

View full text Add to dashboard Cite

The explosive deployment of wired and wireless communication infrastructure has recently enabled many novel applications and sparked new research problems. One of the unsolved issues in today's Internet - the main topic of this thesis - is the goal of increasing data transfer speeds of end hosts by aggregating and simultaneously using multiple network interfaces. This objective is most interesting when the Internet is accessible through several, relatively slow and variable (typically wireless) networks, which are unable to single-handedly provide the required data rate for resource-intensive applications, such as bulk file transfers and high-definition multimedia streaming.

show abstract

“…In order to measure the incidence and relevance of the reorderings observed in our tests, we use the metrics proposed in RFC 5236 [12] and in [21], adapting them to the needs of our specific application when necessary.…”

Section: Metricsmentioning

confidence: 99%

Implementing Information-Theoretically Secure Oblivious Transfer from Packet Reordering

Palmieri

Pereira

2012

Information Security and Cryptology - ICISC 2011

View full text Add to dashboard Cite

Abstract. If we assume that adversaries have unlimited computational capabilities, secure computation between mutually distrusting players can not be achieved using an error-free communication medium. However, secure multi-party computation becomes possible when a noisy channel is available to the parties. For instance, the Binary Symmetric Channel (BSC) has been used to implement Oblivious Transfer (OT), a fundamental primitive in secure multi-party computation. Current research is aimed at designing protocols based on real-world noise sources, in order to make the actual use of information-theoretically secure computation a more realistic prospect for the future. In this paper, we introduce a modified version of the recently proposed Binary Discrete-time Delaying Channel (BDDC), a noisy channel based on communication delays. We call our variant Reordering Channel (RC), and we show that it successfully models packet reordering, the common behavior of packet switching networks that results in the reordering of the packets in a stream during their transit over the network. We also show that the protocol implementing oblivious transfer on the BDDC can be adapted to the new channel by using a different sending strategy, and we provide a functioning implementation of this modified protocol. Finally, we present strong experimental evidence that reordering occurrences between two remote Internet hosts are enough for our construction to achieve statistical security against honest-but-curious adversaries.

show abstract

Improved Packet Reordering Metrics

Cited by 42 publications

References 10 publications

Identifying Almost Sorted Permutations from TCP Buffer Dynamics

Identifying Almost Sorted Permutations from TCP Buffer Dynamics

Multipath aggregation of heterogeneous access networks

Implementing Information-Theoretically Secure Oblivious Transfer from Packet Reordering

Contact Info

Product

Resources

About