NocStudio is a physically-aware auto NoC (Network on Chip) synthesis platform. It is capable of specifying, designing, analyzing, and generating NetSpeed NoC IP.
NocStudio accepts user input interactively through a sophisticated interactive GUI or by means of script files that contain the SoC specification.
NocStudio works in a simple 3 step flow—
Physically aware, quick TAT, algorithm based optimization, built-in performance simulator, deadlock-free construction, etc.
NetSpeed IP is cognizant of multiple security risks at the SoC level and provides architectural security support at different levels. We are confident that we will satisfy your need. Please contact NetSpeed support for further detail.
Yes. NocStudio supports TrustZone feature. Please contact NetSpeed Support for further detail.
Yes. NocStudio generates complete IPXACT files for all NetSpeed IP.
NocStudio uses sophisticated methods for detecting, analyzing, and avoiding deadlocks. The NoC enerated is guaranteed to be correct-by-construction with protocol-level and network-level deadlock avoidance.
NocStudio is blazingly fast; in minutes for all the configurations we have worked on. We believe NocStudio would be a great help to your architects for SoC architecture planning & design.
There is no penalty for Deadlock avoidance. We are deadlock-free by construction. Orion uses graph theory based approach and formal techniques to ensure that there are no cycles in the entire message dependency chain of the system. NocStudio captures inter-dependencies between various messages and interfaces in the system. Then, it augments it with additional dependency information interpreted from the protocol definition (AXI, ACE, etc.) and infer from the system traffic. Finally the combined dependency specification is used to ensure full deadlock avoidance. We are deadlock free both at protocol level and network level and NocStudio would not generate a NoC if it cannot avoid the deadlock.
While performing deadlock avoidance we generate the system dependency graph as well so that users can crosscheck and make sure that there are no deadlocks. If we detect a protocol deadlock that cannot be solved we notify the user of the deadlock and the primary cause so they can fix it by changing their protocol.