Conference Paper

Hierarchical Graph: A New Cost Effective Architecture for Network on Chip

Department of Computer Engineering, Sharif University of Technology, Teheran, Tehrān, Iran
DOI: 10.1007/11596356_33 Conference: Embedded and Ubiquitous Computing - EUC 2005, International Conference EUC 2005, Nagasaki, Japan, December 6-9, 2005, Proceedings
Source: DBLP


We purposed a new Network on Chip (NoC) architecture called Hierarchical Graph. The most interesting feature of this novel ar- chitecture is its simple implementation process. Furthermore, the flexible structure of this topology makes it suitable for use in application specified chips. To benchmark the suggested architecture with existing ones, basic models of physical implementation have been extracted and simulated using NS-2. The results compared with the common used architecture Mesh show that HG has better performance, especially in local traffics and high loads.

Full-text preview

Available from:
  • Source
    • "Design and selection of appropriate architecture for on-chip communication has a key role in the design and implementation of the complete platform for NoC. Authors in Faraydon et al. (2001), Alireza et al. (2005), Partha et al. (2005), Hemayet et al. (2005), Sheraz et al. (2011) (2009) and Yi et al. (2002) have proposed, evaluated, or analyzed different on-chip architectures. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The Network on Chip (NoC) paradigm is rapidly replacing bus based System on Chip (SoC) designs due to their inherent disadvantages such as non-scalability, saturation and congestion. Currently very few tools are available for the simulation and evaluation of on-chip architectures. This study proposes a generic object oriented model for performance evaluation of on-chip interconnect architectures and algorithms. The generic nature of the proposed model can help the researchers in evaluation of any kind of on-chip switching networks. The model was applied on 2D-Mesh and 2D-Diagonal-Mesh on-chip switching networks for verification and selection of best out of both the analyzed architectures. The results show the superiority of 2D-Diagonal-Mesh over 2D-Mesh in terms of average packet delay.
    Full-text · Article · Mar 2013
  • Source
    • "In Ref. [28], the authors have proposed a design methodology for automating application specific NoC synthesis. In Ref. [29], the authors have proposed a NoC based hierarchical graph (HG) as a flexible communication architecture which can be modified for application specific need. HG employs less number of routers than mesh network and thus, occupies less chip area and can be implemented easily and costs less. "
    [Show abstract] [Hide abstract]
    ABSTRACT: There are many design challenges in the hardware-software co-design approach for performance improvement of data-intensive streaming applications with a general-purpose microprocessor and a hardware accelerator. These design challenges are mainly to prevent hardware area fragmentation to increase resource utilization, to reduce hardware reconfiguration cost and to partition and schedule the tasks between the microprocessor and the hardware accelerator efficiently for performance improvement and power savings of the applications.In this paper a modular and block based hardware configuration architecture named memory-aware run-time reconfigurable embedded system (MARTRES) is proposed for efficient resource management and performance improvement of streaming applications. Subsequently we design a task placement algorithm named hierarchical best fit ascending (HBFA) algorithm to prove that MARTRES configuration architecture is very efficient in increased resource utilization and flexible in task mapping and power savings. The time complexity of HBFA algorithm is reduced to O(n) compared to traditional Best Fit (BF) algorithm’s time complexity of O(n2), when the quality of the placement solution by HBFA is better than that of BF algorithm. Finally we design an efficient task partitioning and scheduling algorithm named balanced partitioned and placement-aware partitioning and scheduling algorithm (BPASA). In BPASA we exploit the temporal parallelism in streaming applications to reduce reconfiguration cost of the hardware, while keeping in mind the required throughput of the output data. We balance the exploitation of spatial parallelism and temporal parallelism in streaming applications by considering the reconfiguration cost vs. the data transfer cost. The scheduler refers to the HBFA placement algorithm to check whether contiguous area on FPGA is available before scheduling the task for HW or for SW.
    Full-text · Article · Mar 2009 · Computers & Electrical Engineering
  • Source
    • "Lacking specific tools for evaluating networks on chip, the performance of NoC architectures leads us to use Network Simulator (NS-2) [8]. In [9] and [10] the authors use the NS- 2 for different performance metrics comparing different architectures. Every topology has been evaluated with 16 IPs. "
    [Show abstract] [Hide abstract]
    ABSTRACT: NoC is a potent solution to address design complexity and productivity problems whose its key component is the interconnect architecture which directly affects both cost and performance parameters. The purpose of this paper is to present the basic ideas behind the development of our new hierarchical network-on-chip (NoC) architecture, called ldquoNormardquo that its most distinguished characteristic is its hierarchical nature. We present two types of Norma, Norma-I and Norma-II which have been compared to 2D Mesh. Results illustrate that Norma-I and Norma-II have far more proper functionality than 2D Mesh, when 75 percent of whole traffic interacts in each subnet.
    Full-text · Conference Paper · Jan 2009
Show more