CN109660456A - A kind of fault-tolerant adaptive routing method based on ant group algorithm - Google Patents

A kind of fault-tolerant adaptive routing method based on ant group algorithm Download PDF

Info

Publication number
CN109660456A
CN109660456A CN201811598088.6A CN201811598088A CN109660456A CN 109660456 A CN109660456 A CN 109660456A CN 201811598088 A CN201811598088 A CN 201811598088A CN 109660456 A CN109660456 A CN 109660456A
Authority
CN
China
Prior art keywords
node
routing
fault
value
ant
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.)
Granted
Application number
CN201811598088.6A
Other languages
Chinese (zh)
Other versions
CN109660456B (en
Inventor
王学香
蔡磊
李晨锋
郑阳
时龙兴
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.)
Southeast University
Original Assignee
Southeast 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 Southeast University filed Critical Southeast University
Priority to CN201811598088.6A priority Critical patent/CN109660456B/en
Publication of CN109660456A publication Critical patent/CN109660456A/en
Application granted granted Critical
Publication of CN109660456B publication Critical patent/CN109660456B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • H04L45/18—Loop-free operations
    • 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
    • H04L45/22—Alternate 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
    • H04L45/74—Address processing for routing

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses a kind of fault-tolerant adaptive routing method based on ant group algorithm, comprising: the position of each node and routing table is configured in label two-dimensional network；Malfunctioning node is detected, if faulty node, calculates the fault value and effective buffer length of the neighbouring node of malfunctioning node；If being searched using adaptive routing algorithm without malfunctioning node and obtaining one or more routed path to be selected；One optimal routed path of selection in routed path is selected to be selected according to assessment scale, pheromones routing table is updated according to Ant Routing algorithm；It detects whether the node that achieves the goal, if reaching, terminates to route；Conversely, judging whether routed path occurs deadlock, terminate if deadlock occurs；Conversely, detecting malfunctioning node again.Optimize the path selection mechanism in fault-tolerant adaptive routing algorithm based on ant group algorithm and update pheromones routing table, there are malfunctioning node, reduces the data transfer delay of entire network-on-chip, realize Network Load Balance.

Description

A kind of fault-tolerant adaptive routing method based on ant group algorithm
Technical field
The present invention relates to the route technology fields of network-on-chip, more particularly to a kind of fault-tolerant adaptive based on ant group algorithm Answer method for routing.
Background technique
With the development of semiconductor technology, chip also gradually from monokaryon to multicore so that many-core develops, this requires Increasingly focus on the interconnection mode of system when chip architecture designs.However the interconnection mode based on bus of mainstream is not now It is able to satisfy the demand of growing chip-scale, for this status, researcher proposes the chip interconnection of network-on-chip Mode.It in network-on-chip, if network size is smaller, is generally routed using certainty, because simply and being able to satisfy when realizing Performance requirement；With the increase of network size, using adaptive routing algorithm, because there is a strong possibility makes at this moment certainty routing Network performance is largely effected at network congestion.But with the increase of network size, node failure just inevitably will appear. Then how to realize can be normally carried out occurring a small amount of malfunctioning node in network route this idea become important grind Study carefully direction.Fault-tolerant adaptive routing algorithm is exactly to solve the problems, such as this from the level of routing algorithm, and traditional is fault-tolerant adaptive Algorithm generally carries out failure prompting using single value, while routing table storage overhead is too big, and in routed path selection Also expense is taken a significant amount of time.Although achieving the effect that reduce delay to a certain extent, there is also certain limitations.
Summary of the invention
In order to solve above-mentioned technical problem, the purpose of the present invention is to propose to a kind of based on ant group algorithm it is fault-tolerant from Method for routing is adapted to, optimize the path selection mechanism in existing fault-tolerant adaptive routing algorithm based on ant group algorithm and updates letter Plain routing table is ceased, the data transfer delay of entire network-on-chip can be reduced there are malfunctioning node in this way, is realized Network Load Balance.
The technical scheme is that
A kind of fault-tolerant adaptive routing method based on ant group algorithm, comprising the following steps:
S01: the position of each node in two-dimensional network is marked using two-dimensional coordinate system and routing table is configured；
S02: detection malfunctioning node, if faulty node enters step S03, if entering step S04 without malfunctioning node；
S03: the fault value FV and effective buffer length EBL of the neighbouring node of malfunctioning node are calculated；
S04: being searched using adaptive routing algorithm and obtains one or more routed path to be selected；
S05: one optimal routed path of selection in routed path is selected to be selected according to assessment scale, according to Ant Routing Algorithm updates pheromones routing table；
S06: detecting whether the node that achieves the goal, if reaching, terminates to route；Conversely, going to step S07；
S07: judging whether routed path occurs deadlock, terminates if deadlock occurs；Conversely, then going to step S02.
In preferred technical solution, routing table setting includes storing entrance and Egress node in the step S01, is added Pheromones, FPD and FV.
In preferred technical solution, the distance of node and malfunctioning node is indicated in the step S03 using fault value FV, therefore The calculation formula of barrier value FV are as follows:
Wherein, dxAnd dyIt indicates in the horizontal and vertical directions away from the distance value of malfunctioning node, FV value is that 0 expression keeps ant Group's pheromones do not change；1 indicates slight reduction ant colony pheromones；2 indicate to be significantly reduced ant colony pheromones；3 indicate ant The pheromones of group are set as 0.
In preferred technical solution, output is filtered out from multiple candidate channels using EBL value in the step S03 and is led to Road；When there are two output channels to be selected of due east E and due south S, the value of EBL calculates as follows:
Wherein, freeslotIt (o) is remaining buffer area number in the o of channel, COC is a series of standby of routing function offer Output channel is selected, FPD refers to failure path complexity.
In preferred technical solution, when source mesh node is related to malfunctioning node, formula is as follows:
Wherein, the subscript of FPD indicates the position of routing node, and subscript indicates the direction of routing, if routing node is failure Node, then the value of FPD is reset to 0, and
When source mesh node is uncorrelated to malfunctioning node, formula is as follows:
Wherein
In preferred technical solution, in the step S05, when there are a plurality of when routing path, FV value is used first It is selected, the preferential routed path for selecting FV value big；It is a plurality of when the FV peak of routing path is identical if it exists, selection The big routed path of EBL value.
Compared with prior art, the invention has the advantages that
Optimize the path selection mechanism in existing fault-tolerant adaptive routing algorithm based on ant group algorithm and updates pheromones Routing table, the foundation for using multiple values to remind as failure can reduce entire piece there are malfunctioning node in this way The data transfer delay of upper network realizes Network Load Balance.
Detailed description of the invention
The invention will be further described with reference to the accompanying drawings and embodiments:
Fig. 1 is the flow chart of the fault-tolerant adaptive routing method the present invention is based on ant group algorithm；
Fig. 2 is that EBL of the present invention calculates schematic diagram；
Fig. 3 is the schematic diagram that malfunctioning node of the present invention is located between the mesh node of source；
Fig. 4 is schematic diagram of the malfunctioning node of the present invention not between the mesh node of source.
Specific embodiment
In order to make the objectives, technical solutions and advantages of the present invention clearer, With reference to embodiment and join According to attached drawing, the present invention is described in more detail.It should be understood that these descriptions are merely illustrative, and it is not intended to limit this hair Bright range.In addition, in the following description, descriptions of well-known structures and technologies are omitted, to avoid this is unnecessarily obscured The concept of invention.
Embodiment:
As shown in Figure 1, a kind of fault-tolerant adaptive routing method based on ant group algorithm comprising the steps of:
Step 1 initializes routing table:
The position of each node is marked using two-dimensional coordinate system in two-dimensional network, routing table is stored into addition to conventional Outside mouth and Egress node also need that pheromones, FPD and FV is added.
Wherein, pheromones are used for ant group algorithm；FPD is for calculating EBL；FV is used to record the distance of node and malfunctioning node Whether have an impact to pheromones.
Step 2 checks malfunctioning node:
Self-checking circuit in bootrom is all included self-checking circuit in general chip, is checked whether using self-checking circuit Faulty node.If so, then marking these malfunctioning nodes, and enter step 3；If no, entering step four.
Step 3 startup separator notification mechanisms:
Startup separator notification mechanisms, the i.e. neighbouring node failure value (FaultValue, FV) of calculating malfunctioning node and effective Buffer length (EffectiveBufferLength, EBL)；Malfunctioning node pair is indicated using the form that FV is combined with EBL The influence of other nodes in network, specific calculations incorporated Fig. 2 are as follows:
FV value is calculated, calculation formula is as follows:
Wherein, dxAnd dyWhat is indicated is distance value in the horizontal and vertical directions away from malfunctioning node.
FV value has 0,1,2,3 these four situations respectively, and respectively 0 indicates that ant colony pheromones is kept not change；1 indicates slight Reduction ant colony pheromones；2 indicate to be significantly reduced ant colony pheromones；3 indicate the pheromones of ant colony to be set as 0.
EBL value is calculated, filters out better output channel from multiple candidate channels using the value of EBL, it in this way can be with Preferably avoid congestion.When there are two output channels to be selected of due east E and due south S, the value of EBL calculates as follows:
Wherein, freeslot(o)Refer to remaining buffer area number in the o of channel, it is a series of that COC refers to that routing function provides Alternative output channel (CandidateOutputChannels, COC), FPD refer to failure path complexity (Fault- Location-BasePD, FPD)；
It is specific calculate divide that source mesh node is related to malfunctioning node again and source mesh node it is uncorrelated with malfunctioning node both Situation, different situations calculate as follows:
1, as follows using formula when source mesh node is related to malfunctioning node:
Wherein, the subscript of FPD indicates the position of routing node, and subscript indicates the direction of routing.If routing node is failure Node, then the value of FPD is reset to 0, and
2, as follows using formula when source mesh node is uncorrelated to malfunctioning node:
Wherein
Step 4 routed path is searched:
Carry out accessed path using adaptive routing algorithm, one or more routed path can be found.Here also there are two types of Situation: one is when malfunctioning node blocks Shortest path routing, as shown in figure 3, using non-Shortest path routing, that is, around Capable mode；Another kind is when malfunctioning node does not block Shortest path routing, as shown in figure 4, using Shortest path routing.
The selection of step 5 routed path:
After step 4, one or more can be obtained to routing path.Come using the form that FV and EBL are combined An optimal paths are therefrom selected, specific rule is as follows:
(1) when only one when routing path, this is selected to route road path；
(2) it when there are a plurality of when routing path, is selected first using FV value, FV value is bigger, more preferential selection；If There are a plurality of when the FV peak of routing path is identical, is selected to routing path using EBL value, and value is got over Big more preferential selection.
After making the selection of routed path, pheromones routing table is updated using Ant Routing algorithm.
Step 6 route iteration is completed:
It checks whether data reach destination node, if reaching, completes this routing；It carries out if not reaching from step Rapid two start iteration.
It should be understood that above-mentioned specific embodiment of the invention is used only for exemplary illustration or explains of the invention Principle, but not to limit the present invention.Therefore, that is done without departing from the spirit and scope of the present invention is any Modification, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.In addition, appended claims purport of the present invention Covering the whole variations fallen into attached claim scope and boundary or this range and the equivalent form on boundary and is repairing Change example.

Claims (6)

1. a kind of fault-tolerant adaptive routing method based on ant group algorithm, which comprises the following steps:
S01: the position of each node in two-dimensional network is marked using two-dimensional coordinate system and routing table is configured；
S02: detection malfunctioning node, if faulty node enters step S03, if entering step S04 without malfunctioning node；
S03: the fault value FV and effective buffer length EBL of the neighbouring node of malfunctioning node are calculated；
S04: being searched using adaptive routing algorithm and obtains one or more routed path to be selected；
S05: one optimal routed path of selection in routed path is selected to be selected according to assessment scale, according to Ant Routing algorithm Update pheromones routing table；
S06: detecting whether the node that achieves the goal, if reaching, terminates to route；Conversely, going to step S07；
S07: judging whether routed path occurs deadlock, terminates if deadlock occurs；Conversely, then going to step S02.
2. the fault-tolerant adaptive routing method according to claim 1 based on ant group algorithm, which is characterized in that the step Routing table setting includes storing entrance and Egress node in S01, and pheromones, FPD and FV is added.
3. the fault-tolerant adaptive routing method according to claim 1 based on ant group algorithm, which is characterized in that the step The distance of node and malfunctioning node, the calculation formula of fault value FV are indicated in S03 using fault value FV are as follows:
Wherein, dxAnd dyIt indicates in the horizontal and vertical directions away from the distance value of malfunctioning node, FV value is that 0 expression keeps ant colony letter Breath element does not change；1 indicates slight reduction ant colony pheromones；2 indicate to be significantly reduced ant colony pheromones；3 indicate ant colony Pheromones are set as 0.
4. the fault-tolerant adaptive routing method according to claim 1 based on ant group algorithm, which is characterized in that the step Output channel is filtered out from multiple candidate channels using EBL value in S03；When there are the output to be selected of due east E and due south S two is logical When road, the value of EBL calculates as follows:
Wherein, freeslot(o)It is remaining buffer area number in the o of channel, COC is a series of alternative outputs for routing function and providing Channel, FPD refer to failure path complexity.
5. the fault-tolerant adaptive routing method according to claim 4 based on ant group algorithm, which is characterized in that when source mesh section When point is related to malfunctioning node, formula is as follows:
Wherein, the subscript of FPD indicates the position of routing node, and subscript indicates the direction of routing, if routing node is malfunctioning node, Then the value of FPD is reset to 0, and
When source mesh node is uncorrelated to malfunctioning node, formula is as follows:
Wherein
6. the fault-tolerant adaptive routing method according to claim 1 based on ant group algorithm, which is characterized in that the step It in S05, when there are a plurality of when routing path, is selected first using FV value, the preferential routed path for selecting FV value big； Routed path that is a plurality of when the FV peak of routing path is identical if it exists, selecting EBL value big.
CN201811598088.6A 2018-12-26 2018-12-26 Fault-tolerant self-adaptive routing method based on ant colony algorithm Active CN109660456B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811598088.6A CN109660456B (en) 2018-12-26 2018-12-26 Fault-tolerant self-adaptive routing method based on ant colony algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811598088.6A CN109660456B (en) 2018-12-26 2018-12-26 Fault-tolerant self-adaptive routing method based on ant colony algorithm

Publications (2)

Publication Number Publication Date
CN109660456A true CN109660456A (en) 2019-04-19
CN109660456B CN109660456B (en) 2021-04-02

Family

ID=66116715

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811598088.6A Active CN109660456B (en) 2018-12-26 2018-12-26 Fault-tolerant self-adaptive routing method based on ant colony algorithm

Country Status (1)

Country Link
CN (1) CN109660456B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110808911A (en) * 2019-11-13 2020-02-18 天津津航计算技术研究所 Networking communication routing method based on ant colony pheromone

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103490933A (en) * 2013-09-27 2014-01-01 国家电网公司 Method for protecting and restoring service comprising inevitable nodes
CN105610707A (en) * 2016-02-01 2016-05-25 东南大学 Implementation method of AntNet routing algorithm in two-dimensional mesh topology network-on-chip

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103490933A (en) * 2013-09-27 2014-01-01 国家电网公司 Method for protecting and restoring service comprising inevitable nodes
CN105610707A (en) * 2016-02-01 2016-05-25 东南大学 Implementation method of AntNet routing algorithm in two-dimensional mesh topology network-on-chip

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
CHIA-AN LIN等: "ACO-based fault-aware routing algorithm for Network-on-Chip systems", 《IEEE》 *
HSIEN-KAI HSIN等: "Ant Colony Optimization-Based Adaptive Network-on-Chip Routing Framework Using Network Information Region", 《IEEE》 *
HSIEN-KAI HSIN等: "Ant Colony Optimization-Based Fault-Aware Routing in Mesh-Based Network-on-Chip Systems", 《IEEE》 *
杜文峰等: "基于缓冲区反馈的差异化多路径并发传输数据分配算法", 《小型微型计算机系统》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110808911A (en) * 2019-11-13 2020-02-18 天津津航计算技术研究所 Networking communication routing method based on ant colony pheromone
CN110808911B (en) * 2019-11-13 2022-05-17 天津津航计算技术研究所 Networking communication routing method based on ant colony pheromone

Also Published As

Publication number Publication date
CN109660456B (en) 2021-04-02

Similar Documents

Publication Publication Date Title
ES2735006T3 (en) Driver placement for quick failure in split architecture
US10284465B2 (en) Efficient algorithmic forwarding in fat-tree networks
US9379971B2 (en) Method and apparatus for determining paths between source/destination pairs
US8537677B2 (en) Incremental adaptive packet routing in a multi-dimensional network
US10623303B2 (en) Method of routing data and switch in a network
US9137098B2 (en) T-Star interconnection network topology
US7200117B2 (en) Method of optimizing network capacity and fault tolerance in deadlock-free routing
US8243604B2 (en) Fast computation of alterative packet routes
US8964559B2 (en) Deadlock prevention in direct networks of arbitrary topology
US9577956B2 (en) System and method for supporting multi-homed fat-tree routing in a middleware machine environment
Ebrahimi et al. Minimal-path fault-tolerant approach using connection-retaining structure in networks-on-chip
WO2016091029A1 (en) Method and apparatus for forwarding traffic of stacking system
KR20140053265A (en) Method and apparatus for resilient routing of control traffic in a split-architecture system
CN101262298B (en) Multi-failure protection method for multiple service levels in WDM network
US8825986B2 (en) Switches and a network of switches
CN110380960A (en) A kind of backup path determines method, apparatus, the network equipment and storage medium
US10498631B2 (en) Routing packets using distance classes
US10374943B2 (en) Routing packets in dimensional order in multidimensional networks
CN105794157B (en) Stacking system
EP3328008A1 (en) Deadlock-free routing in lossless multidimensional cartesian topologies with minimal number of virtual buffers
US20190312805A1 (en) Dynamic flooding for link state protocols
ES2448791T3 (en) Method and system for automatic selection of diversion paths in a wireless mesh network
CN109660456A (en) A kind of fault-tolerant adaptive routing method based on ant group algorithm
Priya et al. Fault tolerance in network on chip using bypass path establishing packets
KR102024515B1 (en) How to Add Transmitters and Routes

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant