Dynamic Reconfiguration of Tasks Applied to an UAV System Using Aspect Orientation

Freitas, E. de; Binotto, Alécio Pedro Delazari; Pereira, Carlos Eduardo; Sourander, André; Larsson, Tony

doi:10.1109/ispa.2008.69

Cited by 6 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, before the fmal implementation, there is some additional work to be done in order to include task (re)allocation features, presented in [12], to support additional aspects of the adaptation capability.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Supporting platform for heterogeneous sensor network operation based on unmanned vehicles systems and wireless sensor nodes

Freitas

Allgayer

Wehrmeister

et al. 2009

2009 IEEE Intelligent Vehicles Symposium

Self Cite

View full text Add to dashboard Cite

Advances in vehicle intelligence technology is enabling the development of systems composed of unmanned vehicles, which are able to interact with devices spread on the environment in order to take decisions related to their movements. Sensor networks represent an area that can profit a lot of this new possibilities, as autonomous vehicles can be used to carry sensor devices, which interacting with static sensor nodes can enhance the results provided by the overall system. However, some problems arise in applications' development in such systems due to the network nodes heterogeneity, and also the dynamicity of the environment in which they are deployed, which changes constantly. Thus, new platform solutions are necessary to handle the heterogeneous nodes capabilities in order to facilitate coordination and integration among them. This paper proposes a supporting infrastructure to address these problems composed of an adaptive middleware and a customizable sensor node platform. The goal is to support cooperation in heterogeneous sensor networks, which are composed by static and mobile nodes with different capabilities. The middleware adapts itself in order to manage the very distinct com puting resources of the nodes, and also changes in the environment and in the application demands. The customizable sensor node platform allows optimizations in hw/sw modules to meet specific application requirements, allowing the creation of low-end and resource rich nodes that work in an integrated network. In order to illustrate the proposed approach, a system for military surveillance applications is presented as case study.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Required features are included through adding components or weaving aspects to handle real-time and other non-functional cross-cutting requirements in the minimal middleware. Using aspects to tune the middleware is not focused in this paper, for more information the readers are referred to [9] and [12].…”

Section: The Proposed Middlewarementioning

confidence: 99%

Supporting platform for heterogeneous sensor network operation based on unmanned vehicles systems and wireless sensor nodes

Freitas

Allgayer

Wehrmeister

et al. 2009

2009 IEEE Intelligent Vehicles Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…Scheduling on a CPU-GPU platform: Involving loadbalancing of tasks over a CPU-GPU platform, our previous work [14] firstly identified non-functional requirements that can be used as support for the assignment decision of tasks at high-level granularity. It was followed by [15], where they together observed a modern desktop as a heterogenous cluster composed of the CPU and several asymmetric PUs, like the GPU.…”

Section: Related Workmentioning

confidence: 99%

“…Although they can be extended to attend multiple PUs (which are not in a large number, since our focus is a desktop), we compare them with a straightforward implementation of a greedy technique (Algorithm 5), called here EFT (Earliest First Termination) and partially inspired on [14]. It performs the assignment of a task to the PU that will finish its pending task at the earliest time and, at the same time, can terminate Find a single task T that if switched to the other PU, it yields a maximum gain 5: end for 6: Swap the allocation of T in X 7: Return X; the processing of the new task at first.…”

Section: Dynamic Schedulermentioning

confidence: 99%

An Effective Dynamic Scheduling Runtime and Tuning System for Heterogeneous Multi and Many-Core Desktop Platforms

Binotto

Pereira

Kuijper

et al. 2011

2011 IEEE International Conference on High Performance Computing and Communications

Self Cite

View full text Add to dashboard Cite

A personal computer can be considered as a one-node heterogeneous cluster that simultaneously processes several application tasks. It can be composed by, for example, asymmetric CPU and GPUs. This way, a high-performance heterogeneous platform is built on a desktop for data intensive engineering calculations. In our perspective, a workload distribution over the Processing Units (PUs) plays a key role in such systems. This issue presents challenges since the cost of a task at a PU is non-deterministic and can be affected by parameters not known a priori. This paper presents a context-aware runtime and tuning system based on a compromise between reducing the execution time of engineering applications -due to appropriate dynamic scheduling -and the cost of computing such scheduling applied on a platform composed of CPU and GPUs. Results obtained in experimental case studies are encouraging and a performance gain of 21.77% was achieved in comparison to the static assignment of all tasks to the GPU.

show abstract

“…Scheduling on a CPU-GPU platform: Involving loadbalancing of GPU tasks, our previous work [13] observed modern desktops as heterogenous clusters composed of the CPU and several asymmetric PUs, like the GPU. Complemented by [14], it proposed a dynamic load-balancing framework for high-level tasks with performance profiling support.…”

Section: Related Workmentioning

confidence: 99%

Effective Dynamic Scheduling on Heterogeneous Multi/Manycore Desktop Platforms

Binotto

Pedras

Goetz

et al. 2010

2010 22nd International Symposium on Computer Architecture and High Performance Computing Workshops

Self Cite

View full text Add to dashboard Cite

GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to desktops towards high performance computing. Together with multicore CPUs and other co-processors, a powerful heterogeneous execution platform is built on a desktop for data intensive calculations. In our perspective, we see the modern desktop as a heterogeneous cluster that can deal with several applications' tasks at the same time. To improve application performance and explore such heterogeneity, a distribution of workload over the asymmetric PUs (Processing Units) plays an important role for the system. However, this problem faces challenges since the cost of a task at a PU is non-deterministic and can be influenced by several parameters not known a priori, like the problem size domain. We present a context-aware architecture that maximizes application performance on such platforms. This approach combines a model for a first scheduling based on an offline performance benchmark with a runtime model that keeps track of tasks' real performance. We carried a demonstration using a CPU-GPU platform for computing iterative SLEs (Systems of Linear Equations) solvers using the number of unknowns as the main parameter for assignment decision. We achieved a gain of 38.3% in comparison to the static assignment of all tasks to the GPU (which is done by current programming models, such as OpenCL and CUDA for Nvidia).

show abstract

Dynamic Reconfiguration of Tasks Applied to an UAV System Using Aspect Orientation

Abstract: Abstract

Cited by 6 publications

References 11 publications

Supporting platform for heterogeneous sensor network operation based on unmanned vehicles systems and wireless sensor nodes

Supporting platform for heterogeneous sensor network operation based on unmanned vehicles systems and wireless sensor nodes

An Effective Dynamic Scheduling Runtime and Tuning System for Heterogeneous Multi and Many-Core Desktop Platforms

Effective Dynamic Scheduling on Heterogeneous Multi/Manycore Desktop Platforms

Contact Info

Product

Resources

About