Proceedings of the ACM Special Interest Group on Data Communication 2019
DOI: 10.1145/3341302.3342076
|View full text |Cite
|
Sign up to set email alerts
|

Nitrosketch

Abstract: Software switches are emerging as a vital measurement vantage point in many networked systems. Sketching algorithms or sketches, provide high-fidelity approximate measurements, and appear as a promising alternative to traditional approaches such as packet sampling. However, sketches incur significant computation overhead in software switches. Existing efforts in implementing sketches in virtual switches make sacrifices on one or more of the following dimensions: performance (handling 40 Gbps line-rate packet t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
18
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 150 publications
(18 citation statements)
references
References 63 publications
0
18
0
Order By: Relevance
“…SketchLearn [9] separates large flows from the rest based on inferred flow distributions, which incurs additional processing delay for each packet. Nitrosketch [16] reduces the insertion frequency to relieve the processing delay, but introduces uncertainty on the sketching results. OmniMon [17] seeks full accuracy and resource efficiency over collaborated network entities.…”
Section: B Related Workmentioning
confidence: 99%
“…SketchLearn [9] separates large flows from the rest based on inferred flow distributions, which incurs additional processing delay for each packet. Nitrosketch [16] reduces the insertion frequency to relieve the processing delay, but introduces uncertainty on the sketching results. OmniMon [17] seeks full accuracy and resource efficiency over collaborated network entities.…”
Section: B Related Workmentioning
confidence: 99%
“…ORAM storage for flow statistics and dedicated oblivious primitives for specific measurement logic together can make access patterns of measurement tasks oblivious [51]. Though OblivSketch [51] handles a typical link (40 Kpps, packets per second) with the same characteristic as a backbone link [52], a 10 Gbps line rate with maximum 14.88 Mpps [53] makes the measurement service hard to work with, not to mention monitoring a large-scale network with heavy traffic volume or burst traffic, where more than one switch is deployed.…”
Section: Available Solutions and Limitationsmentioning
confidence: 99%
“…There have been considerable efforts [35][36][37]53,55,56,62,63] in making a reasonable trade-off among goals: achieving a higher line-rate (such as, 40 Gbps) for 64B packets (59.2 Mpps), providing better accuracy guarantees (than CM sketch), saving more memory space, or supporting more general measurement tasks. Thanks to their higher accuracy than sampling methods [57,58], most of them use sketches [54] as their primary data structure to record flow statistics.…”
Section: Non-oblivious Network Measurement Servicementioning
confidence: 99%
See 1 more Smart Citation
“…An elastic NetCache application may be constructed from an elastic count-min sketch and an elastic key-value store. To control the relative stretch of these modules, the programmer can specify a utility function that the Key-value store/ hash table Precision [1], Sonata [8], Network-Wide HH [9], Sketchvisor [11], LinearRoad [13], NetChain [14], NetCache [15], FlowRadar [18], HashPipe [28], Elastic Sketch [31] Hash-based matrix SketchVisor [11], NetCache [15], Nitrosketch [19], UnivMon [20], Sharma et al [25], Fair Queueing [26], Elastic Sketch [31]…”
mentioning
confidence: 99%