CogniServe: Heterogeneous Server Architecture for Large-Scale Recognition

IEEE Micro (Impact Factor: 1.81). 07/2011; DOI: 10.1109/MM.2011.37
Source: IEEE Xplore

ABSTRACT As smart mobile devices become pervasive, vendors are offering rich features supported by cloud-based servers to enhance the user experience. Such servers implement large-scale computing environments, where target data is compared to a massive preloaded database. CogniServe is a highly efficient recognition server for large-scale recognition that employs a heterogeneous architecture to provide low-power, high-throughput cores, along with application-specific accelerators.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The latest developments in mobile computing technology have enabled intensive applications on the modern Smartphones. However, such applications are still constrained by limitations in processing potentials, storage capacity and battery lifetime of the Smart Mobile Devices (SMDs). Therefore, Mobile Cloud Computing (MCC) leverages the application processing services of computational clouds for mitigating resources limitations in SMDs. Currently, a number of computational offloading frameworks are proposed for MCC wherein the intensive components of the application are outsourced to computational clouds. Nevertheless, such frameworks focus on runtime partitioning of the application for computational offloading, which is time consuming and resources intensive. The resource constraint nature of SMDs require lightweight procedures for leveraging computational clouds. Therefore, this paper presents a lightweight framework which focuses on minimizing additional resources utilization in computational offloading for MCC. The framework employs features of centralized monitoring, high availability and on demand access services of computational clouds for computational offloading. As a result, the turnaround time and execution cost of the application are reduced. The framework is evaluated by testing prototype application in the real MCC environment. The lightweight nature of the proposed framework is validated by employing computational offloading for the proposed framework and the latest existing frameworks. Analysis shows that by employing the proposed framework for computational offloading, the size of data transmission is reduced by 91%, energy consumption cost is minimized by 81% and turnaround time of the application is decreased by 83.5% as compared to the existing offloading frameworks. Hence, the proposed framework minimizes additional resources utilization and therefore offers lightweight solution for computational offloading in MCC.
    PLoS ONE 08/2014; · 3.53 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Neuromorphic vision algorithms are biologically inspired models that follow the processing that takes place in the primate visual cortex. Despite their efficiency and robustness, the complexity of these algorithms results in reduced performance when executed on general purpose processors. This paper proposes an application-specific system for accelerating a neuromorphic vision system for object recognition. The system is based on HMAX, a biologically-inspired model of the visual cortex. The neuromorphic accelerators are validated on a multi-FPGA system. Results show that the neuromorphic accelerators are 13.8× (2.6×) more power efficient when compared to CPU (GPU) implementation.
    Journal of Signal Processing Systems 02/2013; 70(2). · 0.56 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The latest developments in mobile devices technology have made smartphones as the future computing and service access devices. Users expect to run computational intensive applications on Smart Mobile Devices (SMDs) in the same way as powerful stationary computers. However in spite of all the advancements in recent years, SMDs are still low potential computing devices, which are constrained by CPU potentials, memory capacity and battery life time. Mobile Cloud Computing (MCC) is the latest practical solution for alleviating this incapacitation by extending the services and resources of computational clouds to SMDs on demand basis. In MCC, application offloading is ascertained as a software level solution for augmenting application processing capabilities of SMDs. The current offloading algorithms offload computational intensive applications to remote servers by employing different cloud models. A challenging aspect of such algorithms is the establishment of distributed application processing platform at runtime which requires additional computing resources on SMDs. This paper reviews existing Distributed Application Processing Frameworks (DAPFs) for SMDs in MCC domain. The objective is to highlight issues and challenges to existing DAPFs in developing, implementing, and executing computational intensive mobile applications within MCC domain. It proposes thematic taxonomy of current DAPFs, reviews current offloading frameworks by using thematic taxonomy and analyzes the implications and critical aspects of current offloading frameworks. Further, it investigates commonalities and deviations in such frameworks on the basis significant parameters such as offloading scope, migration granularity, partitioning approach, and migration pattern. Finally, we put forward open research issues in distributed application processing for MCC that remain to be addressed.
    IEEE Communications Surveys &amp Tutorials 01/2013; 15(3):1294-1313. · 6.49 Impact Factor

Full-text (4 Sources)

Available from
Jul 22, 2014