Exploration and design of a synchronous message passing framework for a CPU-NPU heterogeneous architecture
- Pennefather, Sean, Bradshaw, Karen, Irwin, Barry
- Authors: Pennefather, Sean , Bradshaw, Karen , Irwin, Barry
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460038 , vital:75886 , ISBN 9781538655559 , https://doi.org/10.1109/IPDPSW.2018.00017
- Description: In this paper we present the development of a framework for communication between an NPU (network processing unit) and CPU through synchronous message passing that is compliant with the synchronous communication events of the CSP formalisms. This framework is designed to be used for passing generic information between application components operating on both architectures and is intended to operate in conjunction with existing datapaths present on the NPU which in turn are responsible for network traffic transmission. An investigation of different message passing topologies is covered before the proposed message passing fabric is presented. As a proof of concept, an initial implementation of the fabric is developed and tested to determine its viability and correctness. Through testing it is shown that the implemented framework operates as intended. However, it is noted the throughput of the exploratory implementation is not considered suitable for high-performance applications and further evaluation is required.
- Full Text:
- Date Issued: 2018
- Authors: Pennefather, Sean , Bradshaw, Karen , Irwin, Barry
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460038 , vital:75886 , ISBN 9781538655559 , https://doi.org/10.1109/IPDPSW.2018.00017
- Description: In this paper we present the development of a framework for communication between an NPU (network processing unit) and CPU through synchronous message passing that is compliant with the synchronous communication events of the CSP formalisms. This framework is designed to be used for passing generic information between application components operating on both architectures and is intended to operate in conjunction with existing datapaths present on the NPU which in turn are responsible for network traffic transmission. An investigation of different message passing topologies is covered before the proposed message passing fabric is presented. As a proof of concept, an initial implementation of the fabric is developed and tested to determine its viability and correctness. Through testing it is shown that the implemented framework operates as intended. However, it is noted the throughput of the exploratory implementation is not considered suitable for high-performance applications and further evaluation is required.
- Full Text:
- Date Issued: 2018
Extending the NFComms framework for bulk data transfers
- Pennefather, Sean, Bradshaw, Karen, Irwin, Barry
- Authors: Pennefather, Sean , Bradshaw, Karen , Irwin, Barry
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460050 , vital:75887 , ISBN 9781450366472 , https://doi.org/10.1145/3278681.3278686
- Description: In this paper we present the design and implementation of an indirect messaging extension for the existing NFComms framework that provides communication between a network flow processor and host CPU. This extension addresses the bulk throughput limitations of the framework and is intended to work in conjunction with existing communication mediums. Testing of the framework extensions shows an increase in throughput performance of up to 300× that of the current direct message passing framework at the cost of increased single message latency of up to 2×. This trade-off is considered acceptable as the proposed extensions are intended for bulk data transfer only while the existing message passing functionality of the framework is preserved and can be used in situations where low latency is required for small messages.
- Full Text:
- Date Issued: 2018
- Authors: Pennefather, Sean , Bradshaw, Karen , Irwin, Barry
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460050 , vital:75887 , ISBN 9781450366472 , https://doi.org/10.1145/3278681.3278686
- Description: In this paper we present the design and implementation of an indirect messaging extension for the existing NFComms framework that provides communication between a network flow processor and host CPU. This extension addresses the bulk throughput limitations of the framework and is intended to work in conjunction with existing communication mediums. Testing of the framework extensions shows an increase in throughput performance of up to 300× that of the current direct message passing framework at the cost of increased single message latency of up to 2×. This trade-off is considered acceptable as the proposed extensions are intended for bulk data transfer only while the existing message passing functionality of the framework is preserved and can be used in situations where low latency is required for small messages.
- Full Text:
- Date Issued: 2018
Health education on diabetes at a South African national science festival
- Bradshaw, Karen, Mhlongo, M, Srinivas, S.C., Marara P.
- Authors: Bradshaw, Karen , Mhlongo, M , Srinivas, S.C. , Marara P.
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/460074 , vital:75890
- Description: Diabetes is one of the non-communicable diseases with a major negative impact on the health and development of South Africans. Empowering the population’s understanding of the condition, with health-literacy appropriate approaches, is one of the interventions that allows discussions around the prevention of diabetes. Objective. To determine the effects of a health education programme on increasing knowledge about diabetes and encouraging preventive measures. Method. A public health education exhibition was held by a pharmacy student at a national science festival. It incorporated presentations, posters, health models, word-search games, information leaflets and a computer-based quiz consisting of pre- and post-intervention questions. Results. Junior and senior school learners participated in the computer-based quiz. Results from the junior school pre-intervention phase showed that learners had a fair prior knowledge of diabetes, with an overall score of 52.8%. Improvement in their overall mean score at the 5% significance level was noted (p=0.020). There was a significant difference in the mean score after the intervention at the 1% level (government schools: 65.5 (standard error (SE) 3.1)%, independent schools: 45.9 (6.2)%; p=0.006). Of the senior learners 53.7% (n=137) indicated that they use computers at school, while 118 (46.3%) did not have access to computers. The improvement in overall knowledge of the senior participants after the intervention was significant at the 0.1% level (p is less than 0.001). Conclusion. The health education offered by the pharmacy student's project was interactive and used an interdisciplinary approach to improve health literacy and raise awareness of diabetes. This is a tested intervention that may be adopted for improving health literacy among schoolchildren.
- Full Text:
- Date Issued: 2018
- Authors: Bradshaw, Karen , Mhlongo, M , Srinivas, S.C. , Marara P.
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/460074 , vital:75890
- Description: Diabetes is one of the non-communicable diseases with a major negative impact on the health and development of South Africans. Empowering the population’s understanding of the condition, with health-literacy appropriate approaches, is one of the interventions that allows discussions around the prevention of diabetes. Objective. To determine the effects of a health education programme on increasing knowledge about diabetes and encouraging preventive measures. Method. A public health education exhibition was held by a pharmacy student at a national science festival. It incorporated presentations, posters, health models, word-search games, information leaflets and a computer-based quiz consisting of pre- and post-intervention questions. Results. Junior and senior school learners participated in the computer-based quiz. Results from the junior school pre-intervention phase showed that learners had a fair prior knowledge of diabetes, with an overall score of 52.8%. Improvement in their overall mean score at the 5% significance level was noted (p=0.020). There was a significant difference in the mean score after the intervention at the 1% level (government schools: 65.5 (standard error (SE) 3.1)%, independent schools: 45.9 (6.2)%; p=0.006). Of the senior learners 53.7% (n=137) indicated that they use computers at school, while 118 (46.3%) did not have access to computers. The improvement in overall knowledge of the senior participants after the intervention was significant at the 0.1% level (p is less than 0.001). Conclusion. The health education offered by the pharmacy student's project was interactive and used an interdisciplinary approach to improve health literacy and raise awareness of diabetes. This is a tested intervention that may be adopted for improving health literacy among schoolchildren.
- Full Text:
- Date Issued: 2018
Real-time geotagging and filtering of network data using a heterogeneous NPU-CPU architecture
- Pennefather, Sean, Bradshaw, Karen, Barry, Irwin
- Authors: Pennefather, Sean , Bradshaw, Karen , Barry, Irwin
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460099 , vital:75892 , ISBN 9780620810227
- Description: In this paper, we present the design and implementation of a NPU-CPU heterogeneous network monitoring application. This application allows for both filtering and monitoring operations to be performed on network traffic based on country of origin or destination of IP traffic in real-time at wire speeds up to 1 Gbit/s. This is achievable by distributing the application components to the relevant candidate architectures, leveraging the strengths of each. Communication between architectures is handled at runtime by a low latency synchronous message passing library. Testing of the implemented application indicates that the system can perform geolocation lookups on network traffic in real-time without impacting network throughput.
- Full Text:
- Date Issued: 2018
- Authors: Pennefather, Sean , Bradshaw, Karen , Barry, Irwin
- Date: 2018
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460099 , vital:75892 , ISBN 9780620810227
- Description: In this paper, we present the design and implementation of a NPU-CPU heterogeneous network monitoring application. This application allows for both filtering and monitoring operations to be performed on network traffic based on country of origin or destination of IP traffic in real-time at wire speeds up to 1 Gbit/s. This is achievable by distributing the application components to the relevant candidate architectures, leveraging the strengths of each. Communication between architectures is handled at runtime by a low latency synchronous message passing library. Testing of the implemented application indicates that the system can perform geolocation lookups on network traffic in real-time without impacting network throughput.
- Full Text:
- Date Issued: 2018
Design of a Message Passing Model for Use in a Heterogeneous CPU-NFP Framework for Network Analytics. Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2017, 3-10 September 2017
- Pennefather, Sean, Bradshaw, Karen, Barry, Irwin
- Authors: Pennefather, Sean , Bradshaw, Karen , Barry, Irwin
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460011 , vital:75884 , ISBN 9780620767569 , http://dx.doi.org/10.18489/sacj.v31i2.692
- Description: Currently, network analytics requires direct access to network packets, normally through a third-party application, which means that obtaining realtime results is difficult. We propose the NFP-CPU heterogeneous framework to allow parts of applications written in the Go programming language to be executed on a Network Flow Processor (NFP) for enhanced performance. This paper explores the need and feasibility of implementing a message passing model for data transmission between the NFP and CPU, which is the crux of such a heterogeneous framework. Architectural differences between the two domains are highlighted within this context and we present a solution to bridging these differences.
- Full Text:
- Date Issued: 2017
- Authors: Pennefather, Sean , Bradshaw, Karen , Barry, Irwin
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460011 , vital:75884 , ISBN 9780620767569 , http://dx.doi.org/10.18489/sacj.v31i2.692
- Description: Currently, network analytics requires direct access to network packets, normally through a third-party application, which means that obtaining realtime results is difficult. We propose the NFP-CPU heterogeneous framework to allow parts of applications written in the Go programming language to be executed on a Network Flow Processor (NFP) for enhanced performance. This paper explores the need and feasibility of implementing a message passing model for data transmission between the NFP and CPU, which is the crux of such a heterogeneous framework. Architectural differences between the two domains are highlighted within this context and we present a solution to bridging these differences.
- Full Text:
- Date Issued: 2017
Feature-fusion guidelines for image-based multi-modal biometric fusion
- Brown, Dane, Bradshaw, Karen
- Authors: Brown, Dane , Bradshaw, Karen
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/460063 , vital:75889 , xlink:href="https://doi.org/10.18489/sacj.v29i1.436"
- Description: The feature level, unlike the match score level, lacks multi-modal fusion guidelines. This work demonstrates a newapproach for improved image-based biometric feature-fusion. The approach extracts and combines the face, fingerprintand palmprint at the feature level for improved human identification accuracy. Feature-fusion guidelines, proposed inour recent work, are extended by adding a new face segmentation method and the support vector machine classifier.The new face segmentation method improves the face identification equal error rate (EER) by 10%. The support vectormachine classifier combined with the new feature selection approach, proposed in our recent work, outperforms otherclassifiers when using a single training sample. Feature-fusion guidelines take the form of strengths and weaknessesas observed in the applied feature processing modules during preliminary experiments. The guidelines are used toimplement an effective biometric fusion system at the feature level, using a novel feature-fusion methodology, reducingthe EER of two groups of three datasets namely: SDUMLA face, SDUMLA fingerprint and IITD palmprint; MUCT Face,MCYT Fingerprint and CASIA Palmprint.
- Full Text:
- Date Issued: 2017
- Authors: Brown, Dane , Bradshaw, Karen
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , article
- Identifier: http://hdl.handle.net/10962/460063 , vital:75889 , xlink:href="https://doi.org/10.18489/sacj.v29i1.436"
- Description: The feature level, unlike the match score level, lacks multi-modal fusion guidelines. This work demonstrates a newapproach for improved image-based biometric feature-fusion. The approach extracts and combines the face, fingerprintand palmprint at the feature level for improved human identification accuracy. Feature-fusion guidelines, proposed inour recent work, are extended by adding a new face segmentation method and the support vector machine classifier.The new face segmentation method improves the face identification equal error rate (EER) by 10%. The support vectormachine classifier combined with the new feature selection approach, proposed in our recent work, outperforms otherclassifiers when using a single training sample. Feature-fusion guidelines take the form of strengths and weaknessesas observed in the applied feature processing modules during preliminary experiments. The guidelines are used toimplement an effective biometric fusion system at the feature level, using a novel feature-fusion methodology, reducingthe EER of two groups of three datasets namely: SDUMLA face, SDUMLA fingerprint and IITD palmprint; MUCT Face,MCYT Fingerprint and CASIA Palmprint.
- Full Text:
- Date Issued: 2017
Improved Automatic Face Segmentation and Recognition for Applications with Limited Training Data
- Bradshaw, Karen, Brown, Dane
- Authors: Bradshaw, Karen , Brown, Dane
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460085 , vital:75891 , ISBN 9783319582740 , https://doi.org/10.1007/978-3-319-58274-0_33
- Description: This paper introduces varied pose angle, a new approach to improve face identification given large pose angles and limited training data. Face landmarks are extracted and used to normalize and segment the face. Our approach does not require face frontalization and achieves consistent results. Results are compared using frontal and non-frontal training images for Eigen and Fisher classification of various face pose angles. Fisher scales better with more training samples only with a high quality dataset. Our approach achieves promising results for three well-known face datasets.
- Full Text:
- Date Issued: 2017
- Authors: Bradshaw, Karen , Brown, Dane
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460085 , vital:75891 , ISBN 9783319582740 , https://doi.org/10.1007/978-3-319-58274-0_33
- Description: This paper introduces varied pose angle, a new approach to improve face identification given large pose angles and limited training data. Face landmarks are extracted and used to normalize and segment the face. Our approach does not require face frontalization and achieves consistent results. Results are compared using frontal and non-frontal training images for Eigen and Fisher classification of various face pose angles. Fisher scales better with more training samples only with a high quality dataset. Our approach achieves promising results for three well-known face datasets.
- Full Text:
- Date Issued: 2017
“Enhanced biometric access control for mobile devices,” in Proceedings of the 20th Southern Africa Telecommunication Networks and Applications Conference
- Bradshaw, Karen, Brown, Dane
- Authors: Bradshaw, Karen , Brown, Dane
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460025 , vital:75885 , ISBN 9780620767569
- Description: In the new Digital Economy, mobile devices are increasingly being used for tasks that involve sensitive and/or f inancial data. Hitherto, security on smartphones has not been a priority and furthermore, users tend to ignore the security features in favour of more rapid access to the device. We propose an authentication system that can provide enhanced security by utilizing multi-modal biometrics from a single image, captured at arm’s length, containing unique face and iris data. The system is compared to state-of-the-art face and iris recognition systems, in related studies using the CASIA-Iris-Distance dataset and the IITD iris dataset. The proposed system outperforms the related studies in all experiments and shows promising advancements to at-a-distance iris recognition on mobile devices.
- Full Text:
- Date Issued: 2017
- Authors: Bradshaw, Karen , Brown, Dane
- Date: 2017
- Subjects: To be catalogued
- Language: English
- Type: text , book
- Identifier: http://hdl.handle.net/10962/460025 , vital:75885 , ISBN 9780620767569
- Description: In the new Digital Economy, mobile devices are increasingly being used for tasks that involve sensitive and/or f inancial data. Hitherto, security on smartphones has not been a priority and furthermore, users tend to ignore the security features in favour of more rapid access to the device. We propose an authentication system that can provide enhanced security by utilizing multi-modal biometrics from a single image, captured at arm’s length, containing unique face and iris data. The system is compared to state-of-the-art face and iris recognition systems, in related studies using the CASIA-Iris-Distance dataset and the IITD iris dataset. The proposed system outperforms the related studies in all experiments and shows promising advancements to at-a-distance iris recognition on mobile devices.
- Full Text:
- Date Issued: 2017
- «
- ‹
- 1
- ›
- »