Area for Public Dissemination of INFN in SHAPES

Since 1984, the APE (Array Processor Experiment) group designed four generations of massive parallel numerical computers based on custom VLIW processors dedicated to Theoretical Physics, mainly High Energy Physics.

In SHAPES, the INFN Roma APE group is mainly involved in the activities of the SHAPES_HW working group.

DNP: Distributed Network Processor

The INFN Roma group's contribution to the SHAPES project consist of a RTL-SystemC, a TLM-SystemC and a VHDL model of the DNP.

DNP stands for Distributed Network Processor, i.e. an IP optimized to off-load networking related tasks from the computation processors. The main task of the DNP is to provide inter-tile communication services. A secondary service is to act as a DMA controller for intra-tile communications. The DNP offers its services to other masters in the tile, typically a RISC or a DSP processor.

Research Positions available

A few research positions are available in INFN Roma funded by the SHAPES project. Look into the APE for detail

Topic revision: r3 - 2008-02-11 - 16:42:38 - DavideRossetti
This site is powered by the TWiki collaboration platformCopyright © by the SHAPES collaboration. All material on this collaboration platform is the property of the SHAPES project partners.
Ideas, requests, problems regarding TWiki? Send feedback