A highly distributable computational framework for fast cloud data retrieval

Amir H. Basirat, Asad I. Khan, Bala Srinivasan

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    Abstract

    Unlike the existing relational, hierarchical and object-oriented schemes, associative models can analyze data in similar ways to which our brain links information. Such interactions when implemented in voluminous data clouds can assist in searching for overarching relations in complex and highly distributed data sets with speed and accuracy. In this paper, a different perspective of data recognition will be considered. Rather than looking at conventional approaches, such as statistical computations and deterministic learning schemes, this paper will be focusing on distributed processing approach for scalable data recognition and processing through applying an access scheme that will enable fast data retrieval across multiple records and data segments associatively, utilizing a parallel approach. Doing so will yield a new form of databaselike functionality that can scale up or down over the available infrastructure without interruption or degradation, dynamically and automatically. In our proposed model, data records are treated as patterns. As a result, data storage and retrieval is performed using a distributed pattern recognition approach that is implemented through the integration of loosely-coupled computational networks, followed by a divide-and-distribute approach that facilitates distribution of these networks within the cloud dynamically.

    LanguageEnglish
    Title of host publication2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015
    Subtitle of host publication9-11 December 2015, Miami, Florida, USA, Proceedings
    EditorsRandy Goebel, Andreas Holzinger, Karin Verspoor
    Place of PublicationPiscataway, NJ
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages246-250
    Number of pages5
    ISBN (Electronic)9781509002870
    DOIs
    Publication statusPublished - 3 Mar 2016
    EventInternational Conference on Machine Learning and Applications 2015 - Miami, United States
    Duration: 9 Dec 201511 Dec 2015
    Conference number: 14th
    http://www.icmla-conference.org/icmla15/

    Conference

    ConferenceInternational Conference on Machine Learning and Applications 2015
    Abbreviated titleICMLA 2015
    CountryUnited States
    CityMiami
    Period9/12/1511/12/15
    Internet address

    Keywords

    • Distributed computing
    • Parallel processing
    • Associative memory
    • Cloud computing
    • Neural networks
    • Hierachical graph neuron

    Cite this

    Basirat, A. H., Khan, A. I., & Srinivasan, B. (2016). A highly distributable computational framework for fast cloud data retrieval. In R. Goebel, A. Holzinger, & K. Verspoor (Eds.), 2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015: 9-11 December 2015, Miami, Florida, USA, Proceedings (pp. 246-250). [7424316] Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICMLA.2015.96
    Basirat, Amir H. ; Khan, Asad I. ; Srinivasan, Bala. / A highly distributable computational framework for fast cloud data retrieval. 2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015: 9-11 December 2015, Miami, Florida, USA, Proceedings. editor / Randy Goebel ; Andreas Holzinger ; Karin Verspoor. Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers, 2016. pp. 246-250
    @inproceedings{84efc9a928344ad9aeec528b8c56cd67,
    title = "A highly distributable computational framework for fast cloud data retrieval",
    abstract = "Unlike the existing relational, hierarchical and object-oriented schemes, associative models can analyze data in similar ways to which our brain links information. Such interactions when implemented in voluminous data clouds can assist in searching for overarching relations in complex and highly distributed data sets with speed and accuracy. In this paper, a different perspective of data recognition will be considered. Rather than looking at conventional approaches, such as statistical computations and deterministic learning schemes, this paper will be focusing on distributed processing approach for scalable data recognition and processing through applying an access scheme that will enable fast data retrieval across multiple records and data segments associatively, utilizing a parallel approach. Doing so will yield a new form of databaselike functionality that can scale up or down over the available infrastructure without interruption or degradation, dynamically and automatically. In our proposed model, data records are treated as patterns. As a result, data storage and retrieval is performed using a distributed pattern recognition approach that is implemented through the integration of loosely-coupled computational networks, followed by a divide-and-distribute approach that facilitates distribution of these networks within the cloud dynamically.",
    keywords = "Distributed computing, Parallel processing, Associative memory, Cloud computing, Neural networks, Hierachical graph neuron",
    author = "Basirat, {Amir H.} and Khan, {Asad I.} and Bala Srinivasan",
    year = "2016",
    month = "3",
    day = "3",
    doi = "10.1109/ICMLA.2015.96",
    language = "English",
    pages = "246--250",
    editor = "Randy Goebel and Andreas Holzinger and Karin Verspoor",
    booktitle = "2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    address = "United States",

    }

    Basirat, AH, Khan, AI & Srinivasan, B 2016, A highly distributable computational framework for fast cloud data retrieval. in R Goebel, A Holzinger & K Verspoor (eds), 2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015: 9-11 December 2015, Miami, Florida, USA, Proceedings., 7424316, IEEE, Institute of Electrical and Electronics Engineers, Piscataway, NJ , pp. 246-250, International Conference on Machine Learning and Applications 2015, Miami, United States, 9/12/15. https://doi.org/10.1109/ICMLA.2015.96

    A highly distributable computational framework for fast cloud data retrieval. / Basirat, Amir H.; Khan, Asad I.; Srinivasan, Bala.

    2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015: 9-11 December 2015, Miami, Florida, USA, Proceedings. ed. / Randy Goebel; Andreas Holzinger; Karin Verspoor. Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers, 2016. p. 246-250 7424316.

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    TY - GEN

    T1 - A highly distributable computational framework for fast cloud data retrieval

    AU - Basirat, Amir H.

    AU - Khan, Asad I.

    AU - Srinivasan, Bala

    PY - 2016/3/3

    Y1 - 2016/3/3

    N2 - Unlike the existing relational, hierarchical and object-oriented schemes, associative models can analyze data in similar ways to which our brain links information. Such interactions when implemented in voluminous data clouds can assist in searching for overarching relations in complex and highly distributed data sets with speed and accuracy. In this paper, a different perspective of data recognition will be considered. Rather than looking at conventional approaches, such as statistical computations and deterministic learning schemes, this paper will be focusing on distributed processing approach for scalable data recognition and processing through applying an access scheme that will enable fast data retrieval across multiple records and data segments associatively, utilizing a parallel approach. Doing so will yield a new form of databaselike functionality that can scale up or down over the available infrastructure without interruption or degradation, dynamically and automatically. In our proposed model, data records are treated as patterns. As a result, data storage and retrieval is performed using a distributed pattern recognition approach that is implemented through the integration of loosely-coupled computational networks, followed by a divide-and-distribute approach that facilitates distribution of these networks within the cloud dynamically.

    AB - Unlike the existing relational, hierarchical and object-oriented schemes, associative models can analyze data in similar ways to which our brain links information. Such interactions when implemented in voluminous data clouds can assist in searching for overarching relations in complex and highly distributed data sets with speed and accuracy. In this paper, a different perspective of data recognition will be considered. Rather than looking at conventional approaches, such as statistical computations and deterministic learning schemes, this paper will be focusing on distributed processing approach for scalable data recognition and processing through applying an access scheme that will enable fast data retrieval across multiple records and data segments associatively, utilizing a parallel approach. Doing so will yield a new form of databaselike functionality that can scale up or down over the available infrastructure without interruption or degradation, dynamically and automatically. In our proposed model, data records are treated as patterns. As a result, data storage and retrieval is performed using a distributed pattern recognition approach that is implemented through the integration of loosely-coupled computational networks, followed by a divide-and-distribute approach that facilitates distribution of these networks within the cloud dynamically.

    KW - Distributed computing

    KW - Parallel processing

    KW - Associative memory

    KW - Cloud computing

    KW - Neural networks

    KW - Hierachical graph neuron

    UR - http://www.scopus.com/inward/record.url?scp=84969663971&partnerID=8YFLogxK

    U2 - 10.1109/ICMLA.2015.96

    DO - 10.1109/ICMLA.2015.96

    M3 - Conference Paper

    SP - 246

    EP - 250

    BT - 2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015

    A2 - Goebel, Randy

    A2 - Holzinger, Andreas

    A2 - Verspoor, Karin

    PB - IEEE, Institute of Electrical and Electronics Engineers

    CY - Piscataway, NJ

    ER -

    Basirat AH, Khan AI, Srinivasan B. A highly distributable computational framework for fast cloud data retrieval. In Goebel R, Holzinger A, Verspoor K, editors, 2015 IEEE 14th International Conference on Machine Learning and Applications, ICMLA 2015: 9-11 December 2015, Miami, Florida, USA, Proceedings. Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers. 2016. p. 246-250. 7424316 https://doi.org/10.1109/ICMLA.2015.96