A Canonical Implementation Of The Advanced Encryption Standard On The Graphics Processing Unit
- Pilkington, Nick, Irwin, Barry V W
- Authors: Pilkington, Nick , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/430007 , vital:72659 , https://digifors.cs.up.ac.za/issa/2008/Proceedings/Research/47.pdf
- Description: This paper will present an implementation of the Advanced Encryption Standard (AES) on the graphics processing unit (GPU). It investigates the ease of implementation from first principles and the difficulties encountered. It also presents a performance analysis to evaluate if the GPU is a viable option for a cryptographics platform. The AES implementation is found to yield orders of maginitude increased performance when compared to CPU based implementations. Although the implementation introduces complica-tions, these are quickly becoming mitigated by the growing accessibility pro-vided by general programming on graphics processing units (GPGPU) frameworks like NVIDIA’s Compute Uniform Device Architechture (CUDA) and AMD/ATI’s Close to Metal (CTM).
- Full Text:
- Authors: Pilkington, Nick , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/430007 , vital:72659 , https://digifors.cs.up.ac.za/issa/2008/Proceedings/Research/47.pdf
- Description: This paper will present an implementation of the Advanced Encryption Standard (AES) on the graphics processing unit (GPU). It investigates the ease of implementation from first principles and the difficulties encountered. It also presents a performance analysis to evaluate if the GPU is a viable option for a cryptographics platform. The AES implementation is found to yield orders of maginitude increased performance when compared to CPU based implementations. Although the implementation introduces complica-tions, these are quickly becoming mitigated by the growing accessibility pro-vided by general programming on graphics processing units (GPGPU) frameworks like NVIDIA’s Compute Uniform Device Architechture (CUDA) and AMD/ATI’s Close to Metal (CTM).
- Full Text:
An Investigation into the Performance of General Sorting on Graphics Processing Units
- Pilkington, Nick, Irwin, Barry V W
- Authors: Pilkington, Nick , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429881 , vital:72648 , https://doi.org/10.1007/978-1-4020-8741-7_65
- Description: Sorting is a fundamental operation in computing and there is a constant need to push the boundaries of performance with different sorting algo-rithms. With the advent of the programmable graphics pipeline, the par-allel nature of graphics processing units has been exposed allowing programmers to take advantage of it. By transforming the way that data is represented and operated on parallel sorting algorithms can be im-plemented on graphics processing units where previously only graphics processing could be performed. This paradigm of programming exhibits potentially large speedups for algorithms.
- Full Text:
- Authors: Pilkington, Nick , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429881 , vital:72648 , https://doi.org/10.1007/978-1-4020-8741-7_65
- Description: Sorting is a fundamental operation in computing and there is a constant need to push the boundaries of performance with different sorting algo-rithms. With the advent of the programmable graphics pipeline, the par-allel nature of graphics processing units has been exposed allowing programmers to take advantage of it. By transforming the way that data is represented and operated on parallel sorting algorithms can be im-plemented on graphics processing units where previously only graphics processing could be performed. This paradigm of programming exhibits potentially large speedups for algorithms.
- Full Text:
Guidelines for Constructing Robust Discrete-Time Computer Network Simulations
- Richter, John, Irwin, Barry V W
- Authors: Richter, John , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429896 , vital:72649 , https://doi.org/10.1007/978-1-4020-8737-0_69
- Description: Developing network simulations is a complex task that is often per-formed in research and testing. The components required to build a network simulator are common to many solutions. In order to expedite further simulation development, these components have been outlined and detailed in this paper. The process for generating and using these components is then detailed, and an example of a simulator that has been implemented using this system, is detailed
- Full Text:
- Authors: Richter, John , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429896 , vital:72649 , https://doi.org/10.1007/978-1-4020-8737-0_69
- Description: Developing network simulations is a complex task that is often per-formed in research and testing. The components required to build a network simulator are common to many solutions. In order to expedite further simulation development, these components have been outlined and detailed in this paper. The process for generating and using these components is then detailed, and an example of a simulator that has been implemented using this system, is detailed
- Full Text:
High level internet scale traffic visualization using hilbert curve mapping
- Irwin, Barry V W, Pilkington, Nick
- Authors: Irwin, Barry V W , Pilkington, Nick
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429911 , vital:72650 , https://doi.org/10.1007/978-3-540-78243-8_10
- Description: A high level analysis tool was developed for aiding in the analysis of large volumes of network telescope traffic, and in particular the comparisons of data col-lected from multiple telescope sources. Providing a visual means for the evaluation of worm propagation algorithms has also been achieved. By using a Hilbert curve as a means of ordering points within the visual-ization space, the concept of nearness between nu-merically sequential network blocks was preserved. The design premise and initial results obtained using the tool developed are discussed, and a number of fu-ture extensions proposed.
- Full Text:
- Authors: Irwin, Barry V W , Pilkington, Nick
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429911 , vital:72650 , https://doi.org/10.1007/978-3-540-78243-8_10
- Description: A high level analysis tool was developed for aiding in the analysis of large volumes of network telescope traffic, and in particular the comparisons of data col-lected from multiple telescope sources. Providing a visual means for the evaluation of worm propagation algorithms has also been achieved. By using a Hilbert curve as a means of ordering points within the visual-ization space, the concept of nearness between nu-merically sequential network blocks was preserved. The design premise and initial results obtained using the tool developed are discussed, and a number of fu-ture extensions proposed.
- Full Text:
Towards a taxonomy of network scanning techniques
- Barnett, Richard J, Irwin, Barry V W
- Authors: Barnett, Richard J , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/430310 , vital:72682 , https://doi.org/10.1145/1456659.1456660
- Description: Network scanning is a common reconnaissance activity in network in-trusion. Despite this, it's classification remains vague and detection sys-tems in current Network Intrusion Detection Systems are incapable of detecting many forms of scanning traffic. This paper presents a classi-fication of network scanning and illustrates how complex and varied this activity is. The presented classification extends previous, well known, definitions of scanning traffic in a manner which reflects this complexity.
- Full Text:
- Authors: Barnett, Richard J , Irwin, Barry V W
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/430310 , vital:72682 , https://doi.org/10.1145/1456659.1456660
- Description: Network scanning is a common reconnaissance activity in network in-trusion. Despite this, it's classification remains vague and detection sys-tems in current Network Intrusion Detection Systems are incapable of detecting many forms of scanning traffic. This paper presents a classi-fication of network scanning and illustrates how complex and varied this activity is. The presented classification extends previous, well known, definitions of scanning traffic in a manner which reflects this complexity.
- Full Text:
Using inetvis to evaluate snort and bro scan detection on a network telescope
- Irwin, Barry V W, van Riel, J P
- Authors: Irwin, Barry V W , van Riel, J P
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429981 , vital:72656 , https://doi.org/10.1007/978-3-540-78243-8_17
- Description: This paper presents an investigative analysis of net-work scans and scan detection algorithms. Visualisa-tion is employed to review network telescope traffic and identify incidents of scan activity. Some of the identified phenomena appear to be novel forms of host discovery. Scan detection algorithms used by the Snort and Bro intrusion detection systems are cri-tiqued by comparing the visualised scans with alert output. Where human assessment disagrees with the alert output, explanations are sought by analysing the detection algorithms. The Snort and Bro algorithms are based on counting unique connection attempts to destination addresses and ports. For Snort, notable false positive and false negative cases result due to a grossly oversimplified method of counting unique destination addresses and ports.
- Full Text:
- Authors: Irwin, Barry V W , van Riel, J P
- Date: 2008
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/429981 , vital:72656 , https://doi.org/10.1007/978-3-540-78243-8_17
- Description: This paper presents an investigative analysis of net-work scans and scan detection algorithms. Visualisa-tion is employed to review network telescope traffic and identify incidents of scan activity. Some of the identified phenomena appear to be novel forms of host discovery. Scan detection algorithms used by the Snort and Bro intrusion detection systems are cri-tiqued by comparing the visualised scans with alert output. Where human assessment disagrees with the alert output, explanations are sought by analysing the detection algorithms. The Snort and Bro algorithms are based on counting unique connection attempts to destination addresses and ports. For Snort, notable false positive and false negative cases result due to a grossly oversimplified method of counting unique destination addresses and ports.
- Full Text:
- «
- ‹
- 1
- ›
- »