H. Peixoto, M. Jacome, Algorithm and architecture-level design space exploration using hierarchical data flows, in: IEEE International Conference on Applications-Specific Systems, Architectures and Processors, 1997, pp. 272-- 282.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....we discuss the results and give an outlook for further work in this area in section 5. 2 RELATED WORK Although there is a large body of work in the area of design space exploration nearly all of it has been in specific domains such as embedded system design [2] hardwaresoftware co design [17], or control systems [13] Even though, at an abstract level, the tasks and the algorithms involved in design space exploration tend to be similar, frameworks continue to be application domain specific. Moreover, they distinguish between the model, and the system that executes and simulates ....

H. P. Peixoto and M. F. Jacome. Algorithm and architecture-level design space exploration using hierarchical data flows. In Proceedings of the IEEE International Conference on Application-Specific Systems, Architectures and Processors (ASAP '97), 1997.

....in the performance of a system s components may have a variety of origins. For example, for high level system representations, such as those used in system level and hardware software codesign, uncertainty may be due to a looping index whose exact initial value is unknown, e.g. data dependent [1, 3, 6, 11]. A number of current system level models use hierarchy and aggregation as a means of controlling complexity [3, 11] If such approaches are # This work was supported by NSF Career Grants NCR 9624230 and MIP 9624321 and a Texas Higher Education Coordinating Board Grant ATP 088. to succeed, and ....

.... such as those used in system level and hardware software codesign, uncertainty may be due to a looping index whose exact initial value is unknown, e.g. data dependent [1, 3, 6, 11] A number of current system level models use hierarchy and aggregation as a means of controlling complexity [3, 11]. If such approaches are # This work was supported by NSF Career Grants NCR 9624230 and MIP 9624321 and a Texas Higher Education Coordinating Board Grant ATP 088. to succeed, and one is to reason efficiently about the required performance of such systems during design space exploration, it will ....

[Article contains additional citation context not shown here]

H. Peixoto and M. Jacome. Algorithm and architecture level design space exploration using hierarchical data flows. In Proc. 11th Intern. Conf. on Application-specific Systems, Architectures and Processors, pages 271--82, July 1997.

....such as MPEG decoding discussed in this paper, exhibit tasks with uncertain processing times, making design subject to timing constraints quite challenging. Current system level models use hierarchy and aggregation as a means of controlling complexity during early design space exploration [5, 24]. If such approaches are to succeed, and one is to reason efficiently about the required performance of such systems during early design space exploration, it will become increasingly critical to capture the performance variability of such complex system components. In this paper we use ....

....synchronization reductions. In general, hierarchical high level system models (e.g. data control flow graphs, see [3] comprise a variety of nodes and arc s representing computations and flow of control constructs, including branching, looping, and possible synchronization requirements, see e.g. [16, 13, 5, 21, 24]. Rather than formally defining such a framework we will exhibit some cases that arise and the manner in which they are reduced to a corresponding DAG. Below we show how the delay weights associated with traversing the nodes in Fig. 2 can be reduced to equivalent path weights. Generalizations of ....

[Article contains additional citation context not shown here]

H. Peixoto and M. Jacome. Algorithm and architecture level design space exploration using hierarchical data flows. In Proc. 11th Intern. Conf. on Application-specific Systems, Architectures and Processors, pages 271--82, July 1997.

....in the performance of a system s components may have a variety of origins. For example, for high level system representations, such as those used in system level and hardware software codesign, uncertainty may be due to a looping index whose exact initial value is unknown, e.g. data dependent [1, 4, 7, 6, 9, 15, 16]. A number of current system level models use hierarchy and aggregation as a means of controlling complexity [4, 15] If such approaches are to succeed, and one is to reason efficiently about the required performance of such systems during design space exploration, it will become increasingly ....

.... as those used in system level and hardware software codesign, uncertainty may be due to a looping index whose exact initial value is unknown, e.g. data dependent [1, 4, 7, 6, 9, 15, 16] A number of current system level models use hierarchy and aggregation as a means of controlling complexity [4, 15]. If such approaches are to succeed, and one is to reason efficiently about the required performance of such systems during design space exploration, it will become increasingly critical to capture the performance variability of aggregated system elements. We propose to use probability ....

[Article contains additional citation context not shown here]

H. Peixoto and M. Jacome. Algorithm and architecture level design space exploration using hierarchical data flows. In Proc. of the 11th Intern. Conf. on Application-specific Systems, Architectures and Processors, pages 271--82, July 1997.

No context found.

H. Peixoto, M. Jacome, Algorithm and architecture-level design space exploration using hierarchical data flows, in: IEEE International Conference on Applications-Specific Systems, Architectures and Processors, 1997, pp. 272-- 282.

No context found.

H. Peixoto and M. Jacome. Algorithm and architecture-level design space exploration using hierarchical data flows. In IEEE International Conference on Applications-Specific Systems, Architectures and Processors, pages 272--282, 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC