Impact of Fragmentation Strategy on Ethernet Performance

Creedon, Eoin; Manzke, Michael

doi:10.1109/npc.2009.15

Cited by 2 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Redundant Array of Inexpensive Disks (RAID) systems, detailed by Hennessy et al (2002), allow the ability to fragment and distribute data across multiple disks for security purposes. Even TCP/IP networks allow for fragmentation of data into packets that comply with the Maximum Transmission Unit (MTU) of the network, although those strategies, highlighted by Creedon et al (2009), are strictly to optimize the performance of a single TCP/IP connection. Modern networking also employs the use of intermediate nodes to provide flexible relaying and routing across broad networks.…”

Section: Introductionmentioning

confidence: 99%

Probability of Data Leakage and Its Impacts on Confidentiality

Simon¹,

Graham²

2022

eccws

View full text Add to dashboard Cite

A multi-channel communication architecture featuring distributed fragments of data is presented as a method for improving security available in a communication architecture. However, measuring security remains challenging. The Quality of Secure Service (QoSS) model defines a manner by which the probability of data leakage and the probability of data corruption may be used to estimate security properties for a given communication network. These two probabilities reflect two of the three aspects of the IT security triad, specifically confidentiality and integrity. The probability of data leakage is directly related to the probability of confidentiality and may be estimated based on the probabilities of data interception, decryption, and decoding. The number of listeners who have access to the communication channels influences these probabilities, and unique to the QoSS model, the ability to fragment and distribute data messages across multiple channels between sender and receiver. To simulate the behaviors of various communication architectures and the possibility of malicious interference, the probability of data leakage and its constituent metrics require a thorough analysis. Even if a listener is aware that multiple channels exist, each intermediate node (if any) simply appears to have one input and one output. There may be one or more listeners, and they may or may not be working cooperatively. Even if the listener(s) gains access to more than one channel, there is still the challenge of decrypting, decoding, or reassembling the fragmented data. The analysis presented herein will explore the probability of confidentiality from both the authorized user’s and the adversary’s perspective.

show abstract

Section: Introductionmentioning

confidence: 99%