The Platform 2012 system is suitable for research in the following areas:
- High-performance parallel computing with unconventional Network-on-chip multi-core architectures
- Parallel algorithms for multi-core processors (e.g. imaging, video, and next-generation immersive applications such as computational photography and augmented reality)
- Task partitioning and mapping algorithms for multi-core processors
- Thread-level parallelism in streaming applications dataflow models
This platform will be distributed to researchers in Canada through the Embedded Systems Canada project that CMC manages on behalf of 37 institutions located across the country. It will also be available to international collaborators involved in the initiative. Computer simulation capability is available now and FPGA-based capability is expected in early 2012.
P2012 Core Platform
The Platform 2012 (P2012) is an accelerator system-on-chip targeted to imaging, video, and next-generation immersive applications such as computational photography and augmented reality. The P2012 computing fabric is highly modular, as it is based on multiple clusters implemented with independent power and clock domains.
More...
The P2012 Software Development Kit (SDK)
The programming environment relies on the services of the system infrastructure and runtime layer, which supports execution engines, dynamic deployment of the applications, platform quality service, and power management. Finally, the base of the stack consists of the platform modeling layer, including a functional simulator, a transaction-level simulator integrating performance modeling of the architecture platform, and architectural power estimation models. The overall environment is integrated into an Eclipse-based integrated development environment (IDE).
More...
FPGA-Based Emulation of Platform P2012 (coming soon)
We are currently developing an FPGA-based emulator for Platform P2012 with the following features:
- Hardware configuration: one cluster with 8 PEs
- Communication between the x86 host and the computing fabric: PCIe
- High level parallel programming language: OpenCL
More...