AU2020101176A4 - Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip - Google Patents

Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip Download PDF

Info

Publication number
AU2020101176A4
AU2020101176A4 AU2020101176A AU2020101176A AU2020101176A4 AU 2020101176 A4 AU2020101176 A4 AU 2020101176A4 AU 2020101176 A AU2020101176 A AU 2020101176A AU 2020101176 A AU2020101176 A AU 2020101176A AU 2020101176 A4 AU2020101176 A4 AU 2020101176A4
Authority
AU
Australia
Prior art keywords
router
transmission loss
dijkstra algorithm
power control
onocs
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
AU2020101176A
Inventor
Junxiong Chai
Shangyou Jiang
Yiyuan Xie
Mengmeng Yu
Yanli Zheng
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest University
Original Assignee
Southwest University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southwest University filed Critical Southwest University
Priority to AU2020101176A priority Critical patent/AU2020101176A4/en
Application granted granted Critical
Publication of AU2020101176A4 publication Critical patent/AU2020101176A4/en
Ceased legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/50—Transmitters
    • H04B10/564—Power control
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/76—Architectures of general purpose stored program computers
    • G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
    • G06F15/7807—System on chip, i.e. computer system on a single chip; System in package, i.e. computer system on one or more chips in a single package
    • G06F15/7825—Globally asynchronous, locally synchronous, e.g. network on chip
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04J—MULTIPLEX COMMUNICATION
    • H04J14/00—Optical multiplex systems
    • H04J14/02—Wavelength-division multiplex systems
    • H04J14/0227—Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • H04J14/0254—Optical medium access
    • H04J14/0267—Optical signaling or routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/124—Shortest path evaluation using a combination of metrics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/24—Multipath
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/10—Packet switching elements characterised by the switching fabric construction
    • H04L49/109—Integrated on microchip, e.g. switch-on-chip
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B6/00—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings
    • G02B6/24—Coupling light guides
    • G02B6/26—Optical coupling means
    • G02B6/35—Optical coupling means having switching means
    • G02B6/354—Switching arrangements, i.e. number of input/output ports and interconnection types
    • G02B6/3562—Switch of the bypass type, i.e. enabling a change of path in a network, e.g. to bypass a failed element in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/80—Optical aspects relating to the use of optical transmission for specific applications, not provided for in groups H04B10/03 - H04B10/70, e.g. optical power feeding or optical transmission through water
    • H04B10/801—Optical aspects relating to the use of optical transmission for specific applications, not provided for in groups H04B10/03 - H04B10/70, e.g. optical power feeding or optical transmission through water using optical interconnects, e.g. light coupled isolators, circuit board interconnections
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04J—MULTIPLEX COMMUNICATION
    • H04J14/00—Optical multiplex systems
    • H04J14/02—Wavelength-division multiplex systems
    • H04J14/0278—WDM optical network architectures
    • H04J14/0284—WDM mesh architectures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/123—Evaluation of link metrics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/60—Router architectures

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Electromagnetism (AREA)
  • Computing Systems (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • General Physics & Mathematics (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A new adaptive routing based on Dijkstra algorithm in Optical Networks-on-chip. In this patent, Dijkstra algorithm is adopted to select the routing path with minimum transmission loss so as to reduce the output power of the link transmitter in the Optical Networks on-Chip (ONoCs). The adaptive routing based on Dijkstra algorithm can save more power than the adaptive power control without vastly reducing network performance. Compared with traditional power control based on 5x5, 6x6, 7x7, 8x8 mesh-based on ONoCs, the invention applied in ONoCs can maximally reduce the average power of optical transmitter by 23.95%, 19.2%, 15.5%, and 13.1%. At the same time, it can save the average power of optical transimitter by 14.64%, 12.5%, 10.97%, and 14.50% compared to the adaptive power control in previous. In network performance, there are no significant performance reduction in latency and throughput compared with the traditional mesh-based ONoCs. It verifies the feasibility of the adaptive routing based on Dijkstra algorithm. This invention is of great significance for the optimized power control method in ONoCs. 1P_1 IP_2 1_3 IP_4 IP_ RouterK Router_2 Router_3 Router_4 Router_5 IP_6 IP_7 PUS U P_9 IP_10 Router_6 Router_7 RouterS Router_9 Router_10 IP_11 1P_12 * 1P_13 IP_14 IP 15 Routfr_1I Router_12 Router_13 Router_14 Router 15 IP_16 IF_17 5 P1 g 1P19 * P_20 Router_16 Router_17 Router_1S Router_19 Router_20 1P_21 IP_22 IP_23 U P_241 IP25 Router_21 Router_22 Router_23 Router_24 Router_25 Fig. 1 -0.4 -oe C> .- -0.4 -1.2- i Optimized transmission loss Traditional transmission loss -145x5 6x6 7x7 8x8 Network size Fig. 2

Description

1P_1 IP_2 1_3 IP_4 IP_
RouterK Router_2 Router_3 Router_4 Router_5
IP_6 IP_7 PUS U P_9 IP_10
Router_6 Router_7 RouterS Router_9 Router_10
IP_11 1P_12 * 1P_13 IP_14 IP 15
Routfr_1I Router_12 Router_13 Router_14 Router 15 IP_16 IF_17 5 P1 g 1P19 * P_20
Router_16 Router_17 Router_1S Router_19 Router_20
1P_21 IP_22 IP_23 U P_241 IP25
Router_21 Router_22 Router_23 Router_24 Router_25
Fig. 1
C> -0.4
.- -0.4 -oe
-1.2- i Optimized transmission loss Traditional transmission loss
-145x5 6x6 7x7 8x8 Network size
Fig. 2
Editorial Note 2020101176 There is only four pages of the description
EXPLORING A NEW ADAPTIVE ROUTING BASED ON DIJKSTRA ALGORITHM IN OPTICAL NETWORKS-ON-CHIP TECHNICAL FIELD
The invention relates to the field of Optical Networks-on-Chip (ONoCs) and nano-scale optical logic devices.
BACKGROUND
Multiprocessor system on chip (MPSoC) attaches great importance to achieve high performance and low power consumption because of consuming power by providing a lot of parallelism. With constantly improving manufacturing and integration of on-chip CMOS-compatible, a number of processing cores on a die are increasing dramatically so that the traditional networks on chip (NoC) has been not enough to meet the demand of large bandwidth and low latency. Thus, ONoCs, which equip with higher bandwidth, lower delay and higher energy efficiency, come to be a efiective solution to chip multiprocessors (CMPs) systems. The optical router is the key component in the ONoCs communication system. Basic optical switching elements, waveguide crossings, optical terminal, and waveguide consisting of optical router will cause crosstalk noise and transmission loss. Besides, larger transmission loss will lead to more power consumption. Thus reducing power consumption attaches great significance of improving network performance. In traditional power control, the power of the source node is allocated in terms of the maximum link transmission loss obtained from the link with the source node to all destination nodes. Compared to the traditional power control, the adaptive power control (APC) is to adjust each link's signal power transmission loss according to the ratio between signal power transmission loss and the maximum signal power transmission loss. Thus, APC is of great significance in reducing the total energy consumption of different communication paths. The APC power control can save power on the basis of traditional power. However, in addition that the link with the largest transmission loss makes full use of the power, there is still extra power wasted in other links. To make the most of link transmission loss and reduce power, Dijkstra algorithm is adopted to realize the routing path selection of with the minimum transmission loss from the source node to the destination node. When the transmission loss of link is minimum, the output power of the optical transmitter can be achieved minimum. Thus it can realize the optimized power control. On the basis of ensuring that the network performance does not decrease significantly, the power of optical transmitters in different communication paths are minimized.
SUMMARY OF THE INVENTION
The present invention proposes a new adaptive routing based on Dijkstra algorithm realizing the selection of routing path with minimum transmission loss. On the basis of keeping the sensitivity of the receiver unchanged, the optical signal power can be reduced when the transmission loss achieved the minimum. In terms of network performance, we verify the feasibility of this algorithm to realize the selection of minimum transmission loss routing path with the aid of OPNET simulation software. The network performance simulation results show that the End-to-end (ETE) delay is slightly larger than that in the traditional network. Furthermore, the throughput of the optimized network is nearly same as the throughput in the traditional network with the network size expanding. Thus, the adaptive routing based on Dijkstra algorithm saves more power than the adaptive power control without vastly sacrificing the network performance.
SPECIFIC IMPLEMENTATION METHODS
In order to better understand the present invention, the implementation of the present invention is further described in detail below with reference to the accompanying drawings: The present invention introduces a new adaptive routing based on Dijkstra algorithm which realizes the selection of routing path with minimum transmission loss. It is analyzed in 5x5 mesh model based ONoCs shown in Fig. 1. Being appropriate for undirected and directed graphs with positive weights, Dijkstra algorithm also could be adopted to network architecture studied in ONoCs. Dijkstra algorithm is the shortest path algorithm from one node to the other nodes. The transmission loss from the i th port to the j th port in router L J)(x, y) is used as the
weight of the selection for routing path.
For example, if processor core IP_1 communicates with processor core
IP 13,wechoose L,(0,2) as the weight value between processor core IP_ 1 and
processorcore IP_ 2, treating L,(0,3) as the weight value between processor core
IP_ 1 and processor core IP _ 6 . Firstly, compare the L,(0,2) and L,,(0,3) , then
the next hop of router from router Router_1 could be determined. In terms of the internal structure of router in paper, L,,(0,2) is more small so that we consider
Router_2 as the next-hop router. Secondly, compare the weight value L, 2 (4,3)
between the processor core Router_2 and Router_7 and the weight value L, 2 (4,2)
between processor cores Router_2 and Router_3, we choose the next-hop router Router_7. Thirdly, compare the weight value L2 2 (1,2) between the processor core
Router_7 and Router_8 and the weight value L 2 2(1,3) between processor cores
Router_7 and Router_12. We choose the next-hop router Router_8. Finally, optical signal reaches the processor cores IP_13. What's more, in order to recognize whether the algorithm realizes the minimum of link transmission loss or not in selecting a routing path, the algorithm will traverse again to compare whether the distance of choosing the shortest node as a transit node is more small or not. For example, the algorithm will calculate and compare between L 1 (0,2)+ L 2 (4,3) and
Le1 (0,3)+L2 1 (1,2) to determine whether the routing path achieves the minimum
transmission loss or not. Ultimately, when processor core IP_1 communicates with processor core IP_13, the routing path selected by the above algorithm is: Router_1 - Router_2 - Router_7 Router_8 - Router_13, which acknowledges the minimum of the transmission loss. According to the fact that the optical signal power received from the receiver should not be less than the sensitivity of the receiver, the output power of the link
transmitter P7, could be calculated in equation (1), in which Lsw is the optical
power loss of the optical switching device, LWG is the optical power loss of the
waveguide, and SRX is the sensitivity of the receiver. Therefore, on the premise that
the receiver sensitivity remains the same, the output power P7, of the link
transmitter is aim to be minimum. PTx Lsw + LWG + S RX (1) On the premise of not vastly sacrificing the network performance, adaptive routing based on Dijkstra algorithm in ONoCs has obvious advantage than adaptive power control in previous study. As is shown in equation (2), the average transmission
loss L(s,d) of all links from the source node to the others is considered as the
transmission loss evaluation value under different network sizes. Fig. 2 shows that under different network sizes 5x5, 6x6, 7x7 and 8x8, the optimized average link transmission loss is respectively 0.193 dB, 0.194dB, 0.183 dB, 0.161 dB lower than the traditional average link transmission loss.
=1 =1 L (d L (s,d ) - (2)
Besides, the performance of optimized power control is better than the adaptive power control and traditional power control. The average output power of the
transmitter in network links named PT(,) is allocated to the each source node, and
the calculation formula is shown in equation (3). M, N represent the network size and
(x,y) represents the location of link source node. PY represents the output power
required by the optical transmitter at the source node in network link. The average output power of optical transmitter under the optimized power control and the adaptive power control are reduced relatively obvious, and it can maximally reduce by 23.95% in 5x5 mesh-based ONoCs in Fig. 3. M N
P YxT (-x ,y
P T (x , y ) -- = (3) M x N End-to-end (ETE) delay and throughput are important indicators for evaluating network performance. With the aid of OPNET simulation software based on discrete event scheduling works, the latency and throughput are obtained in different network sizes. Fig. 4 demonstrates that when the packet size is 1024 B, the ETE-delay of optimized network is a little bit bigger than that in the tradition network. The network throughput is shown in Fig. 5. The network based on Dijkstra algorithm is nearly the same throughput as the traditional network based on the dimensional-order routing algorithm with the increasing of the network size. Furthermore, the throughput of the network based on Dijkstra algorithm and the traditional network based on the dimensional-order routing algorithm gradually increase to saturation with the offered load increasing. It is apparent that the above-described simulation examples of the present invention are merely examples for clearly demonstrating the superiority of the present invention, and are not intended to limit the application of the present invention. For applications in ONoCs, the invention can also be applied to various network topologies based on the above descriptions. There is no need and no way to exhaust all of the implementations. Any modifications made within the spirit and scope of the present invention, equivalents, modifications, and the like in accordance with the present invention are intended to be included within the scope of the appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 is 5x 5 mesh model based ONoCs. Fig. 2 is the average reduction of link transmission loss in MxN mesh-based ONoCs. Fig. 3 is the improvement of the optical transmitter average power in MxN mesh-based ONoCs. Fig. 4 is the ETE Delay in MxN mesh-based ONoCs. Fig. 5 is the network throughput in MxN mesh-based ONoCs.

Claims (4)

  1. Editorial Note 2020101176 There is only one page of the claim
    The claims defining the invention are as follows: 1. A New Adaptive Routing based on Dijkstra algorithm in Optical Networks-on-chip aims to achieve the routing path with minimum transmission loss. Dijkstra algorithm is used in mesh-based Optical Networks-on-Chip (ONoCs). On the premise of ensuring the proper operation of ONoCs, the output power of the optical transmitter on the link achieves the minimum. The optimized power control based on Dijkstra algorithm saves more power than the adaptive power control without vastly sacrificing the network performance.
  2. 2. The minimum transmission loss selection of routing path based on Dijkstra algorithm (subject to claim 1) Dijkstra algorithm could be used for achieving the routing path selection of the minimum transmission loss. The port-to-port transmission loss of five port non-blocking on-chip optical router serves as the weight of the selection of routing path. By comparing the transmission loss of each port of the router, the minimum transmission loss path from the source node to the destination node is selected.
  3. 3. The realization of optimized power control (subject to claim 1) The output power of the optical transmitter on the link is determined on the transmission loss of the optical switching device, the transmission loss of the waveguide, and the sensitivity of the receiver. On the premise of keeping the receiver's sensitivity unchanged, the output power of the optical transmitter could be the minimum when the link transmission loss can be minimized. Compared with traditional power control and adaptive power control, optimized power control can save more power and improve performance.
  4. 4. The network performance of the adaptive routing based on Dijkstra algorithm (subject to claim 1) End-to-end (ETE) delay and throughput are important indicators for evaluating network performance. The ETE delay of optimized network is is slightly larger than that in the tradition network. Meanwhile, the throughput of optimized network based on Dijkstra algorithm has almost the same as the throughput of traditional network.
    Fig. 2 Fig. 1
    Fig. 4 Fig. 3
    Fig. 5
AU2020101176A 2020-06-28 2020-06-28 Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip Ceased AU2020101176A4 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2020101176A AU2020101176A4 (en) 2020-06-28 2020-06-28 Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
AU2020101176A AU2020101176A4 (en) 2020-06-28 2020-06-28 Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip

Publications (1)

Publication Number Publication Date
AU2020101176A4 true AU2020101176A4 (en) 2020-07-30

Family

ID=71738684

Family Applications (1)

Application Number Title Priority Date Filing Date
AU2020101176A Ceased AU2020101176A4 (en) 2020-06-28 2020-06-28 Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip

Country Status (1)

Country Link
AU (1) AU2020101176A4 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114598946A (en) * 2022-01-24 2022-06-07 西安电子科技大学 Fuzzy logic-based on-chip optical network self-adaptive routing planning method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114598946A (en) * 2022-01-24 2022-06-07 西安电子科技大学 Fuzzy logic-based on-chip optical network self-adaptive routing planning method
CN114598946B (en) * 2022-01-24 2023-02-10 西安电子科技大学 Fuzzy logic-based on-chip optical network adaptive routing planning method

Similar Documents

Publication Publication Date Title
US11057300B2 (en) Energy-efficient traffic scheduling algorithm for SDN data center network based on multi-layer virtual topology
Bolotin et al. Routing table minimization for irregular mesh NoCs
CN105743804B (en) Data flow control method and system
JP2012105265A (en) Table-driven routing in dragonfly processor interconnect network
JP2012105266A (en) Progressive adaptive routing in dragonfly processor interconnect network
CN102437953B (en) Low-power-consumption adaptive routing method in network on chip
CN103973564B (en) The adaptive routing method of interconnected network system
CN105187313A (en) Network-on-chip topology structure and adaptive routing method thereof
CN113839878B (en) Network-on-chip approximate communication system for data intensive application
AU2020101176A4 (en) Exploring a new adaptive routing based on dijkstra algorithm in optical networks-on-chip
CN108833271A (en) A kind of power grid wide-area control service communication routing resource and server
Hull et al. Bandwidth management in wireless sensor networks
CN114116596A (en) Dynamic relay-based infinite routing method and architecture for neural network on chip
CN103546397A (en) Self-routing Omega network structure supporting random ordering
CN112506850B (en) On-chip and inter-chip interconnection network
CN117135059A (en) Network topology structure, construction method, routing algorithm, equipment and medium
CN103729332A (en) MoT (mesh-of-tree) structure based low-power-consumption NoC (network-on-chip) routing method
Wang et al. Improved power of two choices for fat-tree routing
CN113079093B (en) Routing method based on hierarchical Q-routing planning
CN117221212B (en) Optical network on chip low congestion routing method and related equipment
Wang et al. Flow control mechanism for wireless network-on-chip
CN117938750B (en) Method, device, equipment, storage medium and product for processing scheduling route information
Wang et al. Performance analysis of multi-path routing in wireless ad hoc networks
Azampanah et al. LATEX: New selection policy for adaptive routing in application-specific NoC
Li et al. Data Center Traffic Rescheduling Algorithm Based on Ant Colony Optimization Algorithm

Legal Events

Date Code Title Description
FGI Letters patent sealed or granted (innovation patent)
MK22 Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry