Beruflich Dokumente
Kultur Dokumente
Manel Langar
Department of Electrical Engineering, National Engineering School of Tunis, Tunisia.
Riadh Bourguiba
Department of Electrical Engineering, National Engineering School of Tunis, Tunisia.
Jaouhar Mouine
Department of Electrical Engineering, Prince Sattam Bin Abdulaziz University, Saudi Arabia.
4392
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 11, Number 6 (2016) pp 4392-4397
© Research India Publications. http://www.ripublication.com
Only non-existing functionalities are specifically designed, 2. Arbitration: the arbiter analyzes the requests of all the
then deeply verified and documented, before being cataloged masters connected to the bus and designates a winner,
for future designs. depending on a specific arbitration policy (fixed priority,
To foster IPs reuse, secure investments and improve the round robin, time division multiple access,...).
overall quality, several standard have been specified that 3. Address decoding: the winning master places an address
precisely define IP interfaces (connectivity and protocol). on the bus for the decoder. This latter decodes the high
These later considerably facilitate IPs exchange between order bits of the address to select a slave, which in turn
design projects and companies. decodes the low order bits of the address to select an
internal storage location and prepares itself for
P2P link responding.
In early devices, communication was quite simple: a producer 4. Data exchange: Depending on the transfer direction (read
synchronizes with a consumer using a simple handshake, in or write), data is either placed on the bus by the slave and
order to transmit data. Today, IPs still implement the same read by the master, or placed by the master and written in
principles, but adapt them to the SoC flavour. Interfaces can the slave.
be master or slave, depending on who order data exchange.
Also, signals have specific meanings and can be optional, In practice, these four phases do not take place sequentially in
depending on the kind of IP (for example, a processor can the same clock cycle, but are rather distributed on four
eventually have cache management signals in its interface). consecutive cycles, in a pipelined fashion. With this
Given the complexity of real components, often IPs have technique, overall latency stays the same, but the throughput
multiple interfaces dedicated to different usages. Thus, a is almost multiplied by four, since four data transfers can
Direct Memory Access (DMA) coprocessor can have both a overlay at a single cycle. Pipelined transfers are supported by
slave interface to receive instruction from the main processor, a panel of burst transfers that avoids pipeline stalls during re-
and a master interface to read/write data from/to memory. arbitration.
4393
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 11, Number 6 (2016) pp 4392-4397
© Research India Publications. http://www.ripublication.com
Switching techniques
The Switching technique describes how the packets are
transmitted between switches from source to destination
nodes.
There are three main switching techniques:
- Store-and-forward: packets are transferred to the next
switch only if the whole packet has been received [19]. So
switches must have enough buffer resources to receive the
longest packet circulating in the NoC. Also, packets
latency depends on their length.
- Virtual cut-through: It is similar to the store-and-forward
but it allows the packet to be transferred to the next switch
once its header is received. This technique reduces the
packet latency and the required buffering resources [10]. Figure 2: Structure of Wormhole switch
4394
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 11, Number 6 (2016) pp 4392-4397
© Research India Publications. http://www.ripublication.com
4395
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 11, Number 6 (2016) pp 4392-4397
© Research India Publications. http://www.ripublication.com
4396
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 11, Number 6 (2016) pp 4392-4397
© Research India Publications. http://www.ripublication.com
[3] Evgeny Bolotin, Israel Cidon, Ran Ginosar, and [16] Masoud Oveis-Gharan and Gul N. Khan. Statically
Avinoam Kolodny. Qnoc: Qos architecture and adaptive multi fifo buffer architecture for network on
design process for network on chip. Journal of chip. Microprocess. Microsyst., 39(1):11-26,
systems architecture, 50:105-128, 2004. February 2015.
[4] William Dally and Brian Towles. Principles and [17] Li-Shiuan Peh and William J. Dally. A delay model
Practices of Interconnection Networks. Morgan and speculative architecture for pipelined routers. In
Kaufmann Publishers Inc., San Francisco,CA, USA, Proceedings of the 7th International Symposium on
2003. High-Performance Computer Architecture, HPCA
[5] William J. Dally. Virtual-channel flow control. ’01, pages 255-, Washington, DC, USA, 2001. IEEE
SIGARCH Comput.Archit. News, 18(2SI):60-68, Computer Society.
May 1990. [18] Rohit Sunkam Ramanujam, Vassos Soteriou, Bill
[6] William J. Dally and Brian Towles. Route packets, Lin, and Li-Shiuan Peh. Design of a high-throughput
not wires: On-chip inteconnection networks. In distributed shared-buffer noc router. In Proceedings
Proceedings of the 38th Annual Design Automation of the 2010 Fourth ACM/IEEE International
Conference, DAC ’01, pages 684-689, New York, Symposium on Networks-on-Chip, NOCS ’10, pages
NY, USA, 2001. ACM. 69-78, Washington, DC, USA, 2010. IEEE Computer
[7] Ming Fang, Songqiao Chen, and Zhang Heying. Society.
Design a dynamically shared buffer with prefetching [19] Andrew Tanenbaum. Computer Networks. Prentice
for multiple virtual channel network on chip. Journal Hall Professional Technical Reference, 4th edition,
of Convergence Information Technology, 8(5):235- 2002.
242, March 2015. [20] Anh T. Tran and Bevan M. Baas. Roshaq: High-
[8] Christopher J. Glass and Lionel M. Ni. The turn performance on-chip router with shared queues. In
model for adaptive routing. J. ACM, 41(5):874-902, Proceedings of the 2011 IEEE 29th International
September 1994. Conference on Computer Design, ICCD ’11, pp.
[9] Yatin Hoskote, Sriram Vangal, Arvind Singh, Nitin 232-238, Washington, DC, USA, 2011. IEEE
Borkar, and Shekhar Borkar. A 5-ghz mesh Computer Society.
interconnect for a teraflops processor. IEEE Micro, [21] Hangsheng Wang, Li-Shiuan Peh, and Sharad Malik.
27(5):51-61, September 2007. Power driven design of router microarchitectures in
[10] Parviz Kermani and Leonard Kleinrock. Virtual cut- on-chip networks. In Proceedings of the 36th Annual
through: a new computer communication switching IEEE/ACM International Symposium on
technique. Computer Networks, 3:267-286, 1979. Microarchitecture, MICRO 36, pages 105-,
[11] Khalid Latif, Amir-Mohammad Rahmani, Ethiopia Washington, DC, USA, 2003. IEEE Computer
Nigussie, Tiberiu Seceleanu, Martin Radetzki, and Society.
Hannu Tenhunen. Partial virtual channel sharing: A [22] Cesar Albenes Zeferino and Altamiro Amadeu Susin.
generic methodology to enhance resource Socin: A parametric and scalable network-on-chip. In
management and fault tolerance in networks-on-chip. Proceedings of the 16th Symposium on Integrated
J. Electron. Test., 29(3):431-452, June 2013. Circuits and Systems Design, SBCCI ’03, pp. 169-,
[12] De Micheli and L. Benini. Networks on Chips. Washington, DC, USA, 2003. IEEE Computer
Morgan Kaufmann Publishers Inc., San Francisco, Society.
CA, USA, 2006.
[13] Robert Mullins, Andrew West, and Simon Moore.
Low-latency virtual-channel routers for on-chip
networks. In Proceedings of the 31st Annual
International Symposium on Computer Architecture,
ISCA ’04, pages188-, Washington, DC, USA, 2004.
IEEE Computer Society.
[14] M. H. Neishaburi and Zeljko Zilic. Reliability aware
noc router architecture using input channel buffer
sharing. In Proceedings of the 19th ACM Great
Lakes Symposium on VLSI, GLSVLSI ’09, pages
511-516, New York, NY, USA, 2009. ACM.
[15] Chrysostomos A. Nicopoulos, Dongkook Park,
Jongman Kim, N. Vijaykrishnan, Mazin S. Yousif,
and Chita R. Das. Vichar: A dynamic virtual channel
regulator for network-on-chip routers. In Proceedings
of the 39th Annual IEEE/ACM International
Symposium on Microarchitecture, MICRO 39, pages
333-346, Washington, DC, USA, 2006. IEEE
Computer Society.
4397