What is a network-on-chip (NoC) IP and design recommendations

A Network-on-Chip (NoC) IP is a type of intellectual property (IP) that is used in System-on-Chip (SoC) designs to facilitate communication between the different components of the SoC. In other words, it is a communication architecture that allows multiple intellectual property (IP) cores within a system-on-chip (SoC) to communicate with each other over a network. NoC provides a scalable and efficient communication mechanism that enables various IPs to share data and communicate with each other.NoC IP provides a scalable, high-performance interconnect fabric that enables efficient data transfer between the different processing elements, memory controllers, and other components of an SoC.

An NoC is formed from multiple elements. First, each IP block has its own interface characteristics—such as data width and clock frequency—and employs one of the many standard protocols that have been adopted by the SoC industry: OCP, APB, AHB, AXI, STBus, and DTL. One or more sockets need to be attached to each of the functional IP blocks, which will then packetize and serialize data from source IPs into a normalized form suitable for transport over the network. Contrariwise, sockets attached to destination IPs will convert incoming packets back into whatever forms are desired.

In addition to the wires linking everything together, the main transport mechanism of the NoC is largely formed from switches and buffers. The switches act like multiplexers with associated arbiters, or demultiplexers with associated mapping logic, using the destination data in each packet’s header to route from its source to its intended destination. Meanwhile, buffers are used as storage elements to aggregate data along a path. For example, a buffer might be quickly loaded from an IP block using a fast clock, which can then turn its attention to other tasks while another IP block using a slower clock drains it.

Last, but certainly not least, pipeline registers are inserted into NoC pathways to address timing concerns. These issues may be caused by the need to traverse long distances across the SoC.

When designing an SoC with NoC IP, there are several factors that must be considered:

Power: 

Power consumption is a critical consideration in SoC design, especially for battery-powered devices. Power optimization techniques such as clock gating, power gating, and voltage scaling can be applied to NoC IP to minimize power consumption.

Performance: 

The performance of an SoC is directly related to the performance of its interconnect fabric. NoC IP must be designed with low-latency and high-bandwidth capabilities to enable efficient communication between components.

Area: 

The area occupied by NoC IP is an important consideration in SoC design, as it directly impacts the cost of production. NoC IP should be designed with a small footprint to minimize the overall area of the SoC.

Reliability: 

NoC reliability refers to the ability of the interconnect fabric to function correctly and consistently over time. In order to ensure reliable operation, NoC IP must be designed to handle various sources of noise and interference that can occur in a complex SoC. This includes providing error detection and correction mechanisms, such as parity or ECC (Error Correction Code), to protect against transmission errors or data corruption.

Fault tolerance: 

Fault tolerance is the ability of an SoC to continue operating in the event of a component failure. NoC IP must be designed to handle component failures without disrupting the overall system operation. This can be achieved through redundancy, where multiple NoC components are implemented, so that if one fails, the others can take over its duties. Additionally, fault-tolerant NoC designs may use techniques such as error correction, error detection, and graceful degradation to ensure that the system continues to function even if a component fails.

Security: 

Security is an important consideration in any SoC design, and NoC IP is no exception. NoC security measures can include authentication and encryption techniques to protect against unauthorized access or data theft. These security measures can be implemented at various levels of the NoC architecture, such as at the hardware level, in the NoC protocol, or at the software level.

Physical awareness is a key aspect for integration and overqall system efficiency:

Physical awareness in NoC means that the network architecture should consider the physical constraints and limitations of the underlying hardware during design. This is because the NoC design should take into account the physical layout of the SoC, including the position of the different IP blocks, the routing resources available, the power delivery network, and the signal integrity of the communication links.

By designing a physically aware NoC, it is possible to optimize the placement of IP blocks and minimize the distance between them, which can reduce the latency and power consumption of the system. Physical awareness also enables efficient routing and avoids congestion in the network, leading to improved performance and reliability. In addition, considering the physical constraints during NoC design helps ensure that the communication links are robust and reliable, reducing the risk of errors and increasing the overall quality of the SoC design.

Some examples of how physical awareness in NoC design can help optimize the performance of the system:

Placement of IP blocks: 

When designing a NoC, it is essential to consider the placement of IP blocks within the SoC. For example, if two IP blocks need to communicate frequently, it would be beneficial to place them close to each other. This can be achieved by designing the NoC to take into account the physical layout of the SoC and placing the communication links between these IP blocks as close as possible. This reduces the communication latency and power consumption, resulting in improved performance.

Routing resources: 

NoC designs should also take into account the available routing resources in the SoC, including the number of routing layers and vias. For instance, if the available routing resources are limited, the NoC design should avoid creating too many long-distance links, which can lead to congestion and lower performance. By optimizing the routing paths, the NoC design can minimize the number of required vias and reduce the overall complexity of the routing structure.

Power delivery network: 

The NoC design should consider the power delivery network of the SoC. For example, the network architecture can be designed to minimize the power consumption of the communication links by using lower power signaling schemes or by taking advantage of the power delivery infrastructure available in the SoC. This can help reduce the overall power consumption of the system, which is particularly important in battery-operated devices.

Signal integrity: 

The NoC design should ensure that the communication links between IP blocks are robust and reliable. This can be achieved by considering the signal integrity of the links and designing the NoC to avoid interference from other signals or noise sources. By optimizing the signal paths, the NoC can minimize signal distortions, improve reliability, and reduce the risk of errors in the communication links.