Many point tools exist to optimize particular aspects of embedded systems. However, an overall design space exploration framework is needed to combine all the decisions into a global search space, and a common interface to the optimization and evaluation tools.
The MULTICUBE project focuses on the definition of an automatic multi-objective Design Space Exploration (DSE) framework to be used to tune the System-on-Chip architecture for the target application evaluating a set of metrics (e.g. energy, latency, throughput, bandwidth, QoS, etc.) for the next generation embedded multimedia platforms.
This overall objective is two-fold. From one side, the MULTICUBE
project will define an automatic multi-objective DSE framework to find
design alternatives that best meet system constraints and cost criteria,
strongly dependent on the target application, but also to restrict the
search space to crucial parameters to enable an efficient exploration.
In the developed DSE framework, a set of heuristic optimization
algorithms must be defined to reduce the overall exploration time by
computing an approximated Pareto set of configurations with respect to
the selected figures of merit. Once the approximated Pareto set has been
built, the designer can quickly select the best system configuration
satisfying the constraints.
From the other side, the MULTICUBE project will define a run-time DSE
framework based on the applications of the results of the static
multi-objective design exploration to optimize the run-time allocation
and scheduling of different application tasks. The design exploration
flow results in a Pareto-optimal set of design alternatives with
different speed, energy, memory and communication bandwidth parameters.
This information can be used at run-time by the operation system to make
an informed decision about how the resources should be distributed over
different tasks running on the multi-processor system on-chip. This
resource distribution can not be performed during the design exploration
itself, since it depends on which tasks are active at a particular point
in time.
To support both static and run-time design exploration flows described
above, the MULTICUBE project will develop a multi-level system
specification and modeling framework to provide static and dynamic
evaluation of the system-level metrics. One of the main characteristics
of this framework consists of supporting the DSE at different
abstraction levels by providing a common tool interface composed of
several layers that are connected through standardized interfaces.
The design methodology will be implemented in a set of open-source and
proprietary Electronic Design Automation tools to guarantee a large
exploitation of the results of the MULTICUBE project in the embedded
system design community. The overall goal is to support the
competitiveness of European industries by optimizing embedded
hardware/software systems while reducing the design time and costs.
 - MULTICUBE Project Leaflet
- MULTICUBE Poster
|
