CN101778047A - Method for resolving optical switching grouping competition and photoelectric hybrid cache device - Google Patents
Method for resolving optical switching grouping competition and photoelectric hybrid cache device Download PDFInfo
- Publication number
- CN101778047A CN101778047A CN201010107927A CN201010107927A CN101778047A CN 101778047 A CN101778047 A CN 101778047A CN 201010107927 A CN201010107927 A CN 201010107927A CN 201010107927 A CN201010107927 A CN 201010107927A CN 101778047 A CN101778047 A CN 101778047A
- Authority
- CN
- China
- Prior art keywords
- optical
- output
- grouping
- light
- buffer
- 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
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Optical Communication System (AREA)
Abstract
The invention requests to protect a photoelectric hybrid conflict device and a method for conflict control and cache dispatching of optical grouping, and relates to the technical field of fiber-optical communication. Aiming at the defects of low cache utilization factor, huge volume and high cost in fiber-optical delay lines of the conventional optical switch, the invention adopts a technical scheme that: the fiber-optical delay lines are used as main sharing output cache, N output terminals share B limited fiber-optical delay lines, an electronic storage is used as a photoelectric hybrid cache structure for an assisted cache, and an output channel of conflict optical grouping is dispatched by an improved first-fit-void-filling algorithm. When the optical grouping fails to be dispatched to an output port, the shortest available FDL cache is searched first, and if the shortest available FDL cache is not found, the optical grouping is cached in the electronic storage; and when the output port is idle or the outputted FDL line is available, the stored grouping is dispatched to the output port. The method reduces the packet loss rate of the optical grouping and achieves the aim of improving the performance of optical switching.
Description
Technical field
The present invention relates to fiber optic communication field, be specifically related to a kind of photoelectric hybrid cache device and competition solution that optical switching grouping competition solves that be used for.
Technical background
The light switching technology is a kind of novel switching technology that wavelength division multiplexing in the communication network (WDM) and dense wave division multipurpose (DWDM) transmission technology are used and the customer service development is risen development under communication network bandwidth and the quick condition that increases of rate requirement, it is expected to break through the information bottleneck problem that the electronic type exchange causes, and makes communication network develop into service-oriented Driven Dynamic optical-fiber network.Optical Switch Node place at communication network, when a plurality of light groupings that arrive when same wavelength need from the same output port output of switching node simultaneously, will compete, and very big based on packet burst in the Internet network of IP operation, then professional probability in the generation competition of light exchcange core node place is also bigger.Owing to also there is not the random asccess memory (RAM) of light at present, light grouping competition solution commonly used mainly is: fibre delay line (FDL) buffer memory, wavelength Conversion and penetrate routing plan partially, they are respectively on the time-domain, on the wavelength domain and solve the race problem of light grouping on the spatial domain.Because optical transponder unit prematurity still only is fit to specific lattice l network and penetrate route partially, therefore, the FDL buffer memory is to think the easiest realization and the most frequently used light territory caching technology at present.Because the cache-time of FDL is a discrete values, can not provide and insert storage at random, the FDL framework and the controlling mechanism of Optical Switch Node are still waiting further research, to improve the performance that finite population FDL buffer efficiency and competition solve.
Solve in the light grouping competition device at existing FDL, there is hol blocking in the buffer memory that FDL is configured in OPS input node place and solves competitive problem that can be relatively poor; The buffer memory that FDL is configured in output node solves the grouping competition better performances, but exists intra-node need quicken and the bigger problem of demand FDL number; FDL is configured in that the internodal feedback of output-input is shared and the output type shared buffer memory has reduced demand FDL number, but there is repeatedly circulation losses in light signal and reduces the risk of signal to noise ratio.In traditional electron exchange field, the electronics random asccess memory (RAM) can provide the random cache of continuous time, but be used for the grouping of buffer memory light, and the electricity/light conversion when needing light/electricity conversion and grouping cache to finish has reduced packet switching speed.Therefore, study novel FDL configuration structure and control method, the present invention shares output buffers and the electronic memory advantage of buffer memory continuously at random in conjunction with FDL, a kind of buffer control method is proposed, the competition resolution ability and the light grouping cache probability of success of finite population FDL are improved, reduce the volume of optical switch and improve the optical switch performance.
Summary of the invention
At existing light exchange configuration fibre delay line FDL buffer memory solve grouping competition exist buffer memory granularity bulky, that FDL is discrete make its utilance not high, arrive load in business and exist the competition resolution ability to descend and the big defective of grouping packet loss when heavier, the present invention is based on output and share fibre delay line, design a kind of photoelectric hybrid cache device and buffer control method and solve optical switching grouping competition in conjunction with a little electrons memory.
The technical scheme that the present invention solves the problems of the technologies described above is, design and a kind ofly share the photoelectric hybrid cache device of fibre delay line in conjunction with electronic memory based on output, this buffer storage comprises: output shared buffer memory module, feedback type electric quantum memory RAM, optical branching device, optical combiner, light/electric transducer, electric to optic converter, SOA optical gate switch, optical signal amplifier and conflict control and buffer scheduling module.Wherein, the output port of optical switch connects SOA optical gate switch by optical branching device respectively, the output of SOA optical gate switch connects output shared buffer memory module, output shared buffer memory module is that N output port of optical switch shared, form by B root fibre delay line, provide D, 2D,, BD kind time delay is exported by optical combiner.Be responsible for the fiber delay line cache that the conflict light of N output port of optical switch divides group selection to use with N optical branching device, the light packetization delay time limit arrives the back and outputs to corresponding output port by optical combiner.
Conflict control and buffer scheduling module are used for sharing the fibre delay line failure light that produces that conflicts and dividing into groups to dispatch exporting, converting electric packet signal to by light/electricity is stored in the feedback type electric quantum memory, the output port for the treatment of switch is during the free time, to electricity grouping carrying out electricity/light conversion, the also amplification back scheduling output of storage.The feedback type electric quantum memory of sharing has m to input/output end port, is connected by electrical to optical converter and optical-electrical converter between the input and output port of optical switch, and the buffer memory of continuous time is provided for the light grouping of cache failure.
Conflict control and buffer scheduling module are to each the electric classified statistics time started of buffer memory in the electronic memory and the limit time value of buffer memory, but the fibre delay line time spent of or output idle with the buffer scheduling module monitors to output port when conflict control, the electricity grouping that is stored in the electronic memory is converted to the light grouping by electric to optic converter, after amplifying, the traveling optical signal of going forward side by side is scheduled on the output port or available FDL delay line of optical switch free time, if light grouping cache limit time also is not converted output when arriving, then abandon this light grouping, limit value when in electronic memory, buffer memory being set, when surpassing buffer memory behind the limit value, abandon the light grouping of limit value when surpassing buffer memory in the electronic memory, to avoid grouping to be stored in for a long time among the RAM, help improving the service efficiency of finite population electronic memory.
Asynchronous packet can adopt the FDL with " seizing every opportunity " function first-selected promptly in-dispatching algorithm of filling a vacancy (FF-VF) is sought available FDL line, because channel 0 is at t
a+ D has the space constantly, newly to the light grouping by first-selection promptly in-dispatching algorithm of filling a vacancy is to the t of delivery channel 0
a+ D exports constantly, like this, does not produce new space, and originally bigger space (void) is split into two little space sheet void 1 and void 2 on the channel 0, and the FDL utilance is improved.
The present invention is directed to the characteristics that the fiber delay line cache utilance is low, bulky and cost is higher of existing optical switch, proposed with fibre delay line as main shared output buffers, realize that N output port share limited B root fibre delay line, with the photoelectric hybrid cache device of electronic memory as auxiliary cache, with improved first-selection promptly in-delivery channel of (FF-VF) algorithmic dispatching of filling a vacancy conflict light grouping.
When the light packet scheduling that arrives the optical switch input port when output port is failed, earlier with first-selected promptly in-dispatching algorithm of filling a vacancy seeks the shortest available FDL to the light buffer memory that divides into groups to carry out, if do not find available FDL, by FDL buffer memory light grouping failure, to be buffered among the feedback type electric quantum memory RAM after light grouping carrying out light/electricity conversion, before coming in the maximum time limit of storage,, electricity grouping carrying out electricity/light conversion and light signal are amplified the back be dispatched to output port output by the input port of optical switch if but light packet collisions control is idle to output port with the buffer scheduling module monitors or the FDL line time spent of output.
The present invention adopts above-mentioned photoelectric hybrid cache device to realize the light packet scheduling is controlled, specifically comprise: if the light packet scheduling that arrive to close the switch inlet from input port runs into conflict during to output port, conflict control is at first sought available shared output optical fibre delay line with the buffer scheduling module and is solved the light grouping that conflicts that produces, (the B root fibre delay line as buffer memory all uses up if the output optical fibre delay line is all unavailable, or block length is oversize, the buffer memory capacity of fibre delay line is not enough), conflict control is carried out light/electricity by light/electric transducer to the grouping that conflicts with the buffer scheduling module schedules and is converted into the signal of telecommunication, the signal of telecommunication is stored among the feedback electronic memory RAM waits for the output port free time.The request service time of each output port grouping of conflict control and buffer scheduling module monitors switch and grouping time departure, do you judge whether output port idle? in case the output port free time, the signal stored among the RAM gone forward side by side through electricity/light conversion is transported to the input port of optical switch after traveling optical signal amplifies, be dispatched to the appointed output terminal mouth by switch.If conflict divides group selection to share output buffers and the feedback electronic memory is all failed, then abandon this grouping.
It is less to have a volume based on the cache lines of N ports share B root fibre delay line of output, the characteristics that the cache lines service efficiency is high, but limited FDL makes the probability of contention packet FDL cache failure also bigger.In conjunction with a spot of electronic memory RAM, and first-selected buffer scheduling control method in promptly, can further improve the grouping cache probability of successful, reduce the probability that divides into groups to be dropped, and reduce light Packet Switch Node FDL buffer memory total demand and the competition resolution ability that improves node, help the realization of optical switch and the raising of performance.
Description of drawings
The photoelectric hybrid cache device structural representation of Fig. 1 optical switch configuration
The scheduling schematic diagram of filling a vacancy among Fig. 2 FDL
FDL in Fig. 3 optical and electrical mixed cache scheduling schematic diagram of filling a vacancy
The improved first-selection of Fig. 4 promptly in-scheduling flow figure fills a vacancy
Embodiment
The present invention is based on output and share buffer storage and the scheduling controlling strategy of fibre delay line, design a kind of photoelectric hybrid cache device and buffer control method and solve optical switching grouping competition in conjunction with a little electrons memory.Photoelectric hybrid cache device comprises, output shares fibre delay line, with organic electronic buffer, light/electric conversion and electric to optic converter, optical gate SOA switch, optical multiplexer and optical branching device, optical signal amplifier, realize that the light that arrives the switch inlet divides into groups to divide into groups because of the light that competition output is blocked, and solves the collision problem of grouping in time.
Conflict control and of the conflict grouping of buffer scheduling part to arrival optical switch inlet, conflict grouping adopt first-selected promptly in-algorithm of filling a vacancy, seek conventional FDL line, if available FDL output line is not found in the light grouping, it is converted into electric grouping cache in electronic memory RAM at random.When conflict control and buffer scheduling partly monitor output port when available channel or buffer memory space (gap lengths should greater than the length of grouping) be arranged, the electricity grouping is converted to the light grouping again and outputs to this channel or be filled in this space and export.This light caching method has been inherited the advantage of FDL optical fiber extended line and electronics buffer, has overcome defective separately to a certain extent, and less electronics input/output interface is used in the optical and electrical mixed cache scheduling, significantly reduces the number with fibre delay line FDL.As when the time, only there be PLR value and the optical and electrical mixed cache of OPS when B=26 of FDL buffer memory to get B=10 at the FDL number, m=4 (η=400T at load ρ=0.9
OnBe every pair of port provide 400 the scheduling durations the buffer memory abilities) time the PLR value identical, the FDL of expression OPS node configuration can be reduced to 10 from 26, add 4 electronic memory input/output end ports in addition, saved 16 fibre delay lines (11D, 12D, 26D), can reduce the FDL total length and reach 236.8km approximately in speed is the OPS network of 10Gb/s, this is of great benefit to reducing OPS node volume.
Below in conjunction with accompanying drawing technical scheme of the present invention is described further.
Accompanying drawing 1 is the photoelectric hybrid cache device of optical switch configuration, comprises the cache module of the shared formula fibre delay line formation of output and the electronics RAM memory that m constitutes electronic memory.This photoelectric hybrid cache device specifically comprises: formula cache module (output shared buffer), electronic memory (electronic RAM), optical branching device, optical combiner, light/electric transducer, electric to optic converter, SOA optical gate switch, optical signal amplifier and conflict control and buffer scheduling module are shared in output.
Wherein, output is shared the formula cache module and is made up of B root fibre delay line FDL, D is provided, 2D, BD kind time delay, for N output port of Optical Switch Node shared, share in the formula cache module in output, D is the base unit time delay of fibre delay line, and the formula buffer memory is shared in FDL output can provide the discrete time of the D integral multiple of routine to postpone, the buffer memory degree of depth is B, the total length of fibre delay line is 0.5B (B+1) D like this, and fibre delay line 0.5NB (B+1) D that needs than the FDL output buffers of routine has lacked N-1 doubly, helps reducing of Optical Switch Node volume.The enforcement of N the shared B root fibre delay line of output port can be adopted following scheme: each output port of Optical Switch Node connects the optical branching device input separately, and be connected with the buffer scheduling module with conflict control, be used to inquire about Optical Switch Node output port state, whether conflict control is idle with buffer scheduling module inquiry Optical Switch Node output port, if the output port free time, then the 1st of optical branching device the branch switch opened, and optical packet signal directly passes out on the fibre delay line of outlet line; If output port is busy, then to seek available short fiber delay line and control the SOA optical gate switch opens of this branch road, signal outputs on short fibre delay line time of delay; If all B root fibre delay lines are all unavailable, whether the length that conflict control and buffer scheduling module are then judged the light grouping is less than the buffer memory limiting value of electronic memory, if buffer memory limiting value less than electronic memory, satisfy the requirement of time period value, Optical Switch Node after changing stores light grouping carrying out light/electricity in the electronic memory into.
The electronic memory of sharing has and many (m to) input/output end port is connected by electrical to optical converter and optical-electrical converter between the input and output port of Optical Switch Node, and the buffer memory of continuous time is provided for the grouping of light cache failure.Conflict control need be to the electronic packets statistics time started of each buffer memory and the limit time value of buffer memory with the buffer scheduling module, but the fibre delay line time spent of or output idle with the buffer scheduling module monitors to output port when conflict control, the electricity grouping that is stored in the electronic memory is converted to the light grouping by electric to optic converter, after amplifying, the traveling optical signal of going forward side by side is scheduled on idle output port of optical switch output or the available FDL delay line, if light grouping cache limit time also is not converted output when arriving, then abandon this light grouping, limit value when in electronic memory, buffer memory being set, when surpassing buffer memory behind the limit value, abandon the light grouping of limit value when surpassing buffer memory in the electronic memory, to avoid grouping to be stored in for a long time among the RAM, help improving the service efficiency of finite population electronic memory.
It is the base unit delay of fibre delay line FDL that conflict control is selected suitable D value with the buffer scheduling module, the B root fibre delay line of sharing output provides the grouping cache time to be respectively D, 2D, ..., BD, the 1st FDL that shares output FDL provides D unit delay, and the 2nd FDL provides 2D unit delay, the rest may be inferred, and B root FDL provides the BD unit interval to postpone.The above-mentioned discrete caching delay value that FDL provides inevitably can produce space (void) on each FDL output line, be illustrated in figure 2 as the scheduling schematic diagram of filling a vacancy among the FDL
In Fig. 2, suppose that light grouping (arrival packe) is at t
aArrive constantly, be desirably in t
fThe service of constantly obtaining is packet-switching on the delivery channel 0 light (at t
fChannel 0 is also idle constantly), and 0＜t
f-t
a＜D, the time of delay that the light that then arrives grouping needs
Though channel t has been arrived in the grouping of delivery channel 0 current cache
a+ 2D the moment, but because the FDL buffer memory only can provide the discrete time of integral multiple D to postpone, newly arrived light grouping can not be at t
fConstantly obtain service.Can adopt the FDL with " seizing every opportunity " function first-selected promptly in-dispatching algorithm of filling a vacancy (FF-VF) because channel 0 is at t
a+ D has the space constantly, newly to the light grouping by first-selection promptly in-algorithmic dispatching of filling a vacancy is to the t of delivery channel 0
a+ D exports constantly, like this, does not produce new space, and originally bigger space (void) is split into two little space sheet void1 and void 2 on the channel 0, and the FDL utilance is improved.
Be illustrated in figure 3 as FDL in the optical and electrical mixed cache scheduling schematic diagram of filling a vacancy.Suppose that light is grouped in t
aArrive constantly, be desirably in t
fThe service of constantly obtaining exchanges on the delivery channel 0 (at t
fThe moment and t
a+ D channel 0 is constantly taken by other grouping), if optical switch does not have electronic memory, this grouping will can be with being dropped owing to no FDL, the grouping of network Loss Rate is increased, if the buffer storage that the photoelectricity that has adopted the present invention to propose mixes, then this light will be changed by light/electricity after being grouped in the FDL cache failure, and be stored at random in the RAM electronic memory, but the wait output port FDL line time spent idle or that share is scheduled on the delivery channel 0, has reduced the Loss Rate of grouping.In photoelectric hybrid cache device, FDL provides the discrete time of integral multiple D to postpone, and electronic memory provides continuous time delay, is arranged during the space the suitable time on the output line, and the light grouping in the electronic memory just can be exported in electricity/light conversion back.At the characteristics of optical and electrical mixed cache structure of the present invention, adopt improved first-selection promptly in-dispatching algorithm of filling a vacancy scheduling light is grouped in the moment output of channel idle.
In at first with first-selection being to the asynchronous packet that arrives-and the conventional FDL line of the algorithm of filling a vacancy searching, if asynchronous packet does not find suitable desirable output line, light/electric transducer is converted to electric grouping cache in electronic memory with asynchronous packet.When grouping found to have available channel or FDL space (gap lengths should greater than the length of grouping), the electricity grouping was converted to the light grouping again and outputs to this channel or be filled in this FDL space and export.In conflict control and buffer scheduling module, establish matrix Q[i, j] be illustrated in t constantly
aInformation among the buffer queue FDL will be removed the channel j of output port i, and matrix V [i, j] expression buffer queue FDL is at moment t
fSpace on the channel j of output port i.Matrix Q[i, j]=a, the element value of b represents that the time started that grouping information takies port i channel j is a and holding time length b, matrix V [i, j]=c, the element value of d is illustrated in port i and channel j and goes up the time started c in the space that produces and time span d (being the gap lengths of FDL) (consider when gap lengths d is also littler than minimum block length, just do not write down and upgrade V[i, j] element value).Matrix E[i] represent will go in the electronic memory electricity of port i to divide into groups E[i]=e, element value f represents the length value and the electric unit interval number of waiting in the memory that is grouped in of electricity grouping.
Be illustrated in figure 4 as adopt improved optical and electrical mixed cache first-selected promptly in-scheduling flow figure fills a vacancy.In the dispatching cycle of each optical switch, conflict control and buffer scheduling module adopt first-selected promptly in-algorithm of filling a vacancy seeks the shortest available FDL line that satisfies the output time requirement, the conflict light grouping to arriving reduces the space on the FDL.Improved first-selection promptly in-the algorithm buffer scheduling implementation of filling a vacancy is:
The 1st step: will go to port i if detect the light grouping, conflict control postpones the available delivery channel j that FDL is sought in scheduling with the buffer scheduling module by FDL, if find, then exporting light is grouped into channel j and calculates block length, and upgrade matrix Q[i, j] and V[i, j] element value; Do not find available delivery channel if the light grouping postpones scheduling by FDL, then whether detected electrons memory receiving terminal has conflict, if do not have, light grouping carrying out light/electricity conversion, store in the electronic memory after sticking time tag, upgrade matrix E[i] value, and put f=0.Receive conflict is arranged if detect electronic memory, oversize as the length of grouping, greater than the maximum length of electronic memory, or the electronic memory memory space uses up, and then this grouping is dropped.
The 2nd step: check matrix V [i, j] the middle element a that represents space time started and time ascending order, does b numerical value have grouping to go to port i in the search electronic memory? if have, relatively whether the gap lengths d of FDL is greater than the length e of electricity grouping, if d＞e then is converted to the light grouping with the grouping of the electricity in the electronic memory by electric to optic converter, seeks available FDL and output to channel j, deletion matrix E[i] value, and upgrade matrix Q[i, j] and V[i, j] element value.
The 3rd step: check whether surpass maximum wait allowed time value the time of delay of dividing into groups in the electronic memory, usually establish η=kT
On, k=1,2 ..., T
OnThe unit interval of the packetization delay in the expression electronic memory.The stand-by period surpasses feasible value in the electrical storage if be grouped in, and promptly f＞η then will divide into groups to delete from electronic memory; Otherwise, with formation E[i] in represent that stand-by period element value f adds 1.
Claims (8)
1. photoelectric hybrid cache device that solves optical switching grouping competition, comprise, the fiber delay line cache module is shared in output, the electronics buffer, SOA optical gate switch, optical branching device, optical combiner, light/electric transducer, electric to optic converter, optical signal amplifier and conflict control and buffer scheduling module, it is characterized in that, the output port of optical switch connects output by optical branching device and SOA optical gate switch respectively and shares the fiber delay line cache module, output shared buffer memory module is made up of fibre delay line, for all output ports of switch are shared, its output provides time delay to export by optical combiner, electronic memory is connected between the input and output port of optical switch by electrical to optical converter and optical-electrical converter, FDL provides the discrete time of the integral multiple of D to postpone, electronic memory provides continuous time delay, arranged during the space the suitable time on the output line, light grouping in the electronic memory is by electricity/light conversion back output, produce the grouping of failure light when share the fibre delay line conflict by output, conflicting is stored in the electronics buffer after this failure light grouping of control and buffer scheduling module schedules is changed by light/electricity, treats that the output port of switch carries out electricity/light conversion during the free time, and carry out signal and amplify back scheduling output.
2. photoelectric hybrid cache device according to claim 1 is characterized in that, fibre delay line employing output shared buffer memory mode in the fiber delay line cache module is shared in output, and N output port shared the buffer memory ability that B root fibre delay line provides; The electronics buffer has many to input/output end port, the electronics buffer memory of continuous time is provided for light conflict grouping, limit value when in the electronics buffer, buffer memory being set, conflict control and each electric classified statistics time started and the buffer memory time limit value of buffer scheduling module to buffer memory in the electronic memory, when surpassing buffer memory, behind the limit value, abandon the light grouping of limit value when surpassing buffer memory in the electronic memory.
3. photoelectric hybrid cache device according to claim 1, it is characterized in that, conflict control and buffer scheduling module adopt first-selected promptly in-algorithm of filling a vacancy seeks the shortest available FDL line that satisfies the output time requirement, the conflict light grouping to arriving reduces the space on the FDL.
4. adopt photoelectric hybrid cache device to solve the method for optical switching grouping competition, it is characterized in that, produce the grouping of failure light when share the fibre delay line conflict by output, conflict control is at first sought available output optical fibre delay line with the buffer scheduling module and is solved the failure light grouping that produces in the shared fiber delay line cache module of the output that is connected the optical switch output port by optical branching device and SOA optical gate switch, if the output optical fibre delay line is all unavailable, to the fail light grouping of light/electric transducer is carried out light/electricity and is converted the signal of telecommunication to, be stored in the electronics buffer and wait for the optical switch output port free time, FDL provides the discrete time of the integral multiple of D to postpone, electronic memory provides continuous time delay, arranged during the space the suitable time on the output line, light grouping in the electronic memory is by electricity/light conversion back output, but when optical switch output port free time or output optical fibre delay line time spent, the signal process electric to optic converter of storing in the electronics buffer is changed the input port of delivering to optical switch after the traveling optical signal of going forward side by side amplifies, be dispatched to optical switch appointed output terminal mouth, all can not receive above-mentioned failure light grouping if share output optical fibre delay line and electronics buffer, then abandon this failure light grouping.
5. method according to claim 4 is characterized in that, the request service time of each output port grouping of conflict control and buffer scheduling module monitors switch and grouping time departure judge whether output port is idle.
6. method according to claim 4, it is characterized in that, be responsible for the fiber delay line cache that the conflict light of a plurality of output ports of optical switch divides group selection to use with a plurality of optical branching devices, the light packetization delay time limit arrives the back and outputs to corresponding output port by optical combiner.
7. method according to claim 4, it is characterized in that, in seeking available output optical fibre delay line process, to the conflict light grouping that arrives with first-selected promptly in-algorithm of filling a vacancy searching satisfies the shortest available FDL line that output time requires, the space on the FDL reduced.
8. method according to claim 7, it is characterized in that, described first-selection promptly in-algorithm of filling a vacancy specifically comprises, conflict control and buffer scheduling module are as finding the available delivery channel of FDL, then export light and be grouped into this channel, if do not find available delivery channel, and the not conflict of electronic memory receiving terminal, the conversion of light/electricity is carried out in light grouping, stores in the electronic memory after sticking time tag; Relatively whether the gap lengths d of FDL greater than the length e of electricity grouping, if d＞e then is converted to the light grouping with the grouping of the electricity in the electronic memory by electric to optic converter, seeks available FDL it is outputed to channel; The stand-by period surpasses feasible value in the electronic memory if electricity is grouped in, and then the electricity grouping is deleted from electronic memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101079277A CN101778047B (en) | 2010-02-09 | 2010-02-09 | Method for resolving optical switching grouping competition and photoelectric hybrid cache device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101079277A CN101778047B (en) | 2010-02-09 | 2010-02-09 | Method for resolving optical switching grouping competition and photoelectric hybrid cache device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101778047A true CN101778047A (en) | 2010-07-14 |
| CN101778047B CN101778047B (en) | 2013-01-23 |
Family
ID=42514378
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101079277A Active CN101778047B (en) | 2010-02-09 | 2010-02-09 | Method for resolving optical switching grouping competition and photoelectric hybrid cache device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101778047B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104780122A (en) * | 2015-03-23 | 2015-07-15 | 中国人民解放军信息工程大学 | Control method for hierarchical network-on-chip router based on cache redistribution |
| CN105165024A (en) * | 2013-05-10 | 2015-12-16 | 华为技术有限公司 | System and method for photonic switching |
| WO2016008145A1 (en) * | 2014-07-18 | 2016-01-21 | 华为技术有限公司 | Routing node, optical switching network and optical signal transmission method |
| WO2016095111A1 (en) * | 2014-12-16 | 2016-06-23 | 华为技术有限公司 | Optical switching signal processing method and device |
| CN105959241A (en) * | 2016-04-27 | 2016-09-21 | 西安空间无线电技术研究所 | Unified scheduling method for space-borne photovoltaic hybrid exchange system |
| CN106416110A (en) * | 2014-05-28 | 2017-02-15 | 华为技术有限公司 | Scalable silicon photonic switching architectures for optical networks |
| CN111615022A (en) * | 2020-05-20 | 2020-09-01 | 重庆邮电大学 | Conflict resolution method for all-optical nodes of space division multiplexing elastic optical network |
| CN114827047A (en) * | 2022-06-24 | 2022-07-29 | 北京国科天迅科技有限公司 | Data transmission method and device, computer equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1717113A (en) * | 2004-07-02 | 2006-01-04 | 电子科技大学 | Photoelectric mixed exchanging struture supporting light burst/group exchanging |
| KR100715523B1 (en) * | 2006-02-13 | 2007-05-07 | 한국정보통신대학교 산학협력단 | Apparatus for data burst transmission in optical burst switching networks and a method thereof |
-
2010
- 2010-02-09 CN CN2010101079277A patent/CN101778047B/en active Active
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105165024A (en) * | 2013-05-10 | 2015-12-16 | 华为技术有限公司 | System and method for photonic switching |
| CN105165024B (en) * | 2013-05-10 | 2019-02-05 | 华为技术有限公司 | System and method for photon count statistic |
| US9955243B2 (en) | 2014-05-28 | 2018-04-24 | Huawei Technologies Co., Ltd | Scalable silicon photonic switching architectures for optical networks |
| CN106416110A (en) * | 2014-05-28 | 2017-02-15 | 华为技术有限公司 | Scalable silicon photonic switching architectures for optical networks |
| WO2016008145A1 (en) * | 2014-07-18 | 2016-01-21 | 华为技术有限公司 | Routing node, optical switching network and optical signal transmission method |
| CN105453495A (en) * | 2014-07-18 | 2016-03-30 | 华为技术有限公司 | Routing node, optical switching network and optical signal transmission method |
| US10117007B2 (en) | 2014-07-18 | 2018-10-30 | Huawei Technologies Co.., Ltd. | Routing node, optical switching network, and optical signal transmission method |
| WO2016095111A1 (en) * | 2014-12-16 | 2016-06-23 | 华为技术有限公司 | Optical switching signal processing method and device |
| CN106797272A (en) * | 2014-12-16 | 2017-05-31 | 华为技术有限公司 | A kind of light exchanges signal processing method and device |
| CN106797272B (en) * | 2014-12-16 | 2019-11-15 | 华为技术有限公司 | A kind of light exchange signal processing method and device |
| CN104780122A (en) * | 2015-03-23 | 2015-07-15 | 中国人民解放军信息工程大学 | Control method for hierarchical network-on-chip router based on cache redistribution |
| CN105959241A (en) * | 2016-04-27 | 2016-09-21 | 西安空间无线电技术研究所 | Unified scheduling method for space-borne photovoltaic hybrid exchange system |
| CN105959241B (en) * | 2016-04-27 | 2019-03-12 | 西安空间无线电技术研究所 | A kind of spaceborne photoelectricity hybrid switching United Dispatching method |
| CN111615022A (en) * | 2020-05-20 | 2020-09-01 | 重庆邮电大学 | Conflict resolution method for all-optical nodes of space division multiplexing elastic optical network |
| CN111615022B (en) * | 2020-05-20 | 2022-04-19 | 重庆邮电大学 | Conflict resolution method for all-optical nodes of space division multiplexing elastic optical network |
| CN114827047A (en) * | 2022-06-24 | 2022-07-29 | 北京国科天迅科技有限公司 | Data transmission method and device, computer equipment and storage medium |
| CN114827047B (en) * | 2022-06-24 | 2022-10-04 | 北京国科天迅科技有限公司 | Data transmission method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101778047B (en) | 2013-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101778047B (en) | Method for resolving optical switching grouping competition and photoelectric hybrid cache device | |
| Danielsen et al. | Optical packet switched network layer without optical buffers | |
| Kantarci et al. | Bandwidth distribution solutions for performance enhancement in long-reach passive optical networks | |
| Pattavina | Architectures and performance of optical packet switching nodes for IP networks | |
| CN101370315A (en) | Sharing type optical packet switching node structure based on inside wavelength | |
| CN1141817C (en) | Multi-grain optical router based on optical burst switch | |
| CN102843294A (en) | Buffer device and buffer method for feeding back shared light based on FDL (fiber delay line) loop | |
| CN100428660C (en) | Optical burst exchange node with internal acceleration | |
| CN101193049B (en) | A core node switching structure for integrated edge node | |
| CN1235357C (en) | Full optical packet switching node structure for supporting burst or non-burst businesses | |
| CN101035390B (en) | Parallel channel scheduler in the optical burst switching | |
| CN102916890A (en) | Shared wavelength conversion device and method for solving multicast competition | |
| CN102111692A (en) | Optical burst switching (OBS) channel scheduling method based on optical buffer (OB) | |
| CN101193050B (en) | A method for data receiving and transmitting of core node switching device in optical sudden network | |
| Shaddad et al. | Contention resolution of optical burst switching for data center | |
| CN1320789C (en) | Method for resoving light burst resource conflict | |
| CN101656900B (en) | OPS node optical buffering device and method based on load selection | |
| CN1316799C (en) | Method for supporting real-time service in optical burst exchange | |
| CN101754058A (en) | Wavelength routing method and system for optical network | |
| Shukla et al. | Optical switching in next-generation data centers: Architectures based on optical switching | |
| CN100346619C (en) | Light burst/grouping switching structure without team head block | |
| Orphanoudakis et al. | A hybrid optical switch architecture with shared electronic buffers | |
| Wang et al. | Study on contention resolution with fiber delay lines in OBS network | |
| Bjomstad et al. | A scalable optical packet switch for variable length packets employing shared electronic buffering | |
| CN116346745A (en) | Optical packet switch and optical packet switching method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |