mTREE: A Customized Multicast-Enabled Tree-Based Network on Chip for AI Chips

The network on chip (NoC) used in the AI chip is required to be able to deliver data to multiple destination endpoints simultaneously. The existing solutions, such as hierarchial mesh (HMESH) and FATTREE all modify the connection between routers to achieve multicasting. Although this method can achieve delivering data to multiple destination endpoints at the same time, it significantly increases the number of routers used and, therefore, brings a huge additional area and power consumption overhead. In order to solve this problem, this work designed a customized multicast-enabled tree-based NoC, named mTREE, by modifying the microarchitecture of the router, including the routing calculation unit, credit counter, and switch allocator. When implemented on Arria10 FPGA, mTREE only brings a 1% increase in resource utilization compared with the traditional tree-topology NoC. Compared with two other multicast-enabled NoCs: 1) FATTREE and 2) HMESH, mTREE can reach a
$3.19\times $
–
$3.50\times $
and
$1.67\times $
–
$3.02\times $
reduction in resource utilization, respectively, and a
$4.81\times $
–
$8.01\times $
and
$2.85\times $
–
$3.03\times $
reduction in energy consumption, respectively.