Performance driven design space exploration for multi-objective optimization using stability in competition

Design Space ◽

Parametric Design ◽

Space Exploration ◽

Systems Development ◽

High Level Synthesis ◽

Performance Parameters ◽

Systems Architecture ◽

Multi Parametric Design Space Exploration (DSE) for optimal micro-architecture synthesis is an extremely complex yet crucial stage in embedded systems development. Often it is very time complex to find the best suitable configuration to map the inherently contradictory performance parameters into systems silicon real estate. Owing to its exponentially exploding design space and multi way combinatorial mapping, DSE has proven to be notoriously hard and intractable for VLSI CAD tools. The presented work introduces a highly scalable and generalized analytical approach to identify the best configuration of systems architecture while maintaining prime accuracy resolution. This PSE approach coupled with Stability in Competition principles has been applied to a number of well known benchmark High Level Synthesis (HLS) applications, with an impressive 71.80% aggregate speedup and results being more pronounced for larger design space HLS applications.

Performance Driven Design Space Exploration for Multi-Objective Optimization Using Stability in Competition

10.32920/ryerson.14663730.v1 ◽

2021 ◽

Author(s):

Summit Sehgal

Keyword(s):

Real Estate ◽

Analytical Approach ◽

Design Space ◽

Parametric Design ◽

Space Exploration ◽

Systems Development ◽

Performance Parameters ◽

Systems Architecture

Multi Parametric Design Space Exploration (DSE) for optimal micro-architecture synthesis is an extremely complex yet crucial stage in embedded systems development. Often it is very time complex to find the best suitable configuration to map the inherently contradictory performance parameters into systems silicon real estate. Owing to its exponentially exploding design space and multi way combinatorial mapping, DSE has proven to be notoriously hard and intractable for VLSI CAD tools. The presented work introduces a highly scalable and generalized analytical approach to identify the best configuration of systems architecture while maintaining prime accuracy resolution. This DSE approach coupled with

Performance Driven Design Space Exploration for Multi-Objective Optimization Using Stability in Competition

10.32920/ryerson.14663730 ◽

2021 ◽

Author(s):

Summit Sehgal

Keyword(s):

Real Estate ◽

Analytical Approach ◽

Design Space ◽

Parametric Design ◽

Space Exploration ◽

Systems Development ◽

Performance Parameters ◽

Systems Architecture

Multi Parametric Design Space Exploration (DSE) for optimal micro-architecture synthesis is an extremely complex yet crucial stage in embedded systems development. Often it is very time complex to find the best suitable configuration to map the inherently contradictory performance parameters into systems silicon real estate. Owing to its exponentially exploding design space and multi way combinatorial mapping, DSE has proven to be notoriously hard and intractable for VLSI CAD tools. The presented work introduces a highly scalable and generalized analytical approach to identify the best configuration of systems architecture while maintaining prime accuracy resolution. This DSE approach coupled with

Implementation and Design Space Exploration of a Turbo Decoder in High-Level Synthesis

2019 International Conference on ReConFigurable Computing and FPGAs (ReConFig) ◽

10.1109/reconfig48160.2019.8994787 ◽

2019 ◽

Author(s):

Wesley Stirk ◽

Jeff Goeders

Keyword(s):

Design Space ◽

Space Exploration ◽

High Level Synthesis ◽

Turbo Decoder ◽

Distributed design-space exploration for high-level synthesis systems

[1992] Proceedings 29th ACM/IEEE Design Automation Conference ◽

10.1109/dac.1992.227806 ◽

2003 ◽

Cited By ~ 24

Author(s):

R. Dutta ◽

J. Roy ◽

R. Vemuri

Keyword(s):

Design Space ◽

Space Exploration ◽

High Level Synthesis ◽

Distributed Design ◽

Divide and conquer high-level synthesis design space exploration

ACM Transactions on Design Automation of Electronic Systems ◽

10.1145/2209291.2209302 ◽

2012 ◽

Vol 17 (3) ◽

pp. 1-19 ◽

Cited By ~ 24

Author(s):

Benjamin Carrion Schafer ◽

Kazutoshi Wakabayashi

Keyword(s):

Design Space ◽

Space Exploration ◽

Divide And Conquer ◽

High Level Synthesis ◽

Synthesis Design ◽

A rapid design space exploration approach for multi-objective optimization of DSP filter designs

10.32920/ryerson.14653857.v1 ◽

2021 ◽

Author(s):

Aakriti Tarun Sharma

Keyword(s):

Design Space ◽

Space Exploration ◽

Systems Design ◽

Design Solution ◽

Multi Objective ◽

Time Period ◽

And Performance ◽

The process of converting a behavioral specification of an application to its equivalent system architecture is referred to as High Level-Synthesis (HLS). A crucial stage in embedded systems design involves finding the trade off between resource utilization and performance. An exhaustive search would yield the required results, but would take a huge amount of time to arrive at the solution even for smaller designs. This would result in a high time complexity. We employ the use of Design Space Exploration (DSE) in order to reduce the complexity of the design space and to reach the desired results in less time. In reality, there are multiple constraints defined by the user that need to be satisfied simultaneously. Thus, the nature of the task at hand is referred to as Multi-Objective Optimization. In this thesis, the design process of DSP benchmarks was analyzed based on user defined constraints such as power and execution time. The analyzed outcome was compared with the existing approaches in DSE and an optimal design solution was derived in a shorter time period.