Perhaps nowhere is this need more prevalent than for multi-core architectures, since this is an areas where traditional processor simulation and system analysis methodologies tend to break down when faced with issues such as concurrent programming and multiple processor flows, asynchronous operations, core-to-core integration issues, etc.

Real (either prototype or production) hardware provides the best source of information for this type of analysis, but often suffers from visibility into functionality of the system, such as processor instruction flow and data movement on embedded buses, which are not typically made available at the pin IO. Increasing the visibility of these embedded operations typically requires an instrumentation subsystem.

Debug solutions vary greatly with different IP vendors, which can make integrating debug and instrumentation systems for heterogeneous multi-core architectures difficult. In this is dynamic area, the most mature standardized solutions addressing system instrumentation and debug are those based on IEEE 5001, also widely referred to as Nexus.

Some background on Nexus 5001
The initial IEEE 5001 Nexus specification was developed in 1999 is an IEEE standard for debug interfaces of embedded systems and processors, and has since been implemented in numerous devices.

Unlike other debug solutions, that are typically proprietary to a processor or tools vender, Nexus is openly defined and consistently supported by a variety of tools vendor and across a range of processor cores. Nexus standards are commercially supported by the Nexus 5001 Forum, which includes silicon suppliers, tools developers, IP companies, and end users. [1]

Nexus was developed to provide extended debug capability, addressing the limitations of JTAG (IEEE 1149.1) when used for data intensive analysis operations such as trace and calibration. While JTAG remains a widely used interface for on-chip debug, its serial architecture has inherent limitations in supporting the speed and bandwidth needed in analysis of modern devices.

Nexus is compatible with JTAG, and in basic run control modes, can be implemented using only a JTAG serial port; but more significantly, Nexus supports the use of high bandwidth interfaces to efficiently transport data between the silicon targets and debug tools. Wider bandwidth and higher speed data interfaces allow for simpler and more powerful debug interfaces than JTAG based debug architectures.

This has significant advantages for multi-core systems, where larger mounts of data can be needed for analysis.. Multi-core debug is a significantly more complex and data intensive analysis problem than single core debug, since in addition to having to debug the application software for each processor core, there are new considerations related to the communications between processors, multiple processors sharing of common resources (memory peripherals, etc.), synchronization for systems where processors or buses may not be running the same speed, etc.