CN104699632B - The management method and device of operation exchange - Google Patents
The management method and device of operation exchange Download PDFInfo
- Publication number
- CN104699632B CN104699632B CN201510144310.5A CN201510144310A CN104699632B CN 104699632 B CN104699632 B CN 104699632B CN 201510144310 A CN201510144310 A CN 201510144310A CN 104699632 B CN104699632 B CN 104699632B
- Authority
- CN
- China
- Prior art keywords
- cell
- order buffer
- address space
- business
- address
- 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.)
- Active
Links
Abstract
The embodiment of the present invention provides the management method and device of a kind of operation exchange, this method comprises: receiving the cell for carrying independent numbering that core switch is sent, and by the cell according in number and type of service the write-in order buffer of the cell;It detects in the order buffer with the presence or absence of the first cell;The number of first cell is next adjacent number of the number of the cell taken out last time;First cell is then taken out from the order buffer, and removes the memory space for storing first cell in the order buffer by the first cell if it exists;The cell taken out is reconstructed into Business Stream form, by configuring dedicated order buffer in outlet device, the cell that will be numbered in inlet device cutting, it is successively buffered in order buffer according to number, and it is taken out according to number, complicated synchronization mechanism is not needed, the complexity of cell switching is effectively reduced, and improves steady degree.
Description
Technical field
The present embodiments relate to computer networking technology more particularly to the management methods and device of a kind of operation exchange.
Background technique
The core of cell switching system is that the data of Business Stream are cut into a lesser cell, is existed according to cell granularity
It is swapped on switching network, realizes the exchange between any two port of data in a network.
By taking time division multiplexing (English: Time Division Multiplexing, abbreviation: TDM) Business Stream as an example, existing skill
Management of the common business Jing Guo cell switching system in art is to be handed over by the way of sorting step by step in every level-one of switching network
Exchange device carries out synchronization/sequence of whole net rank, by the synchronization signal of whole net, restores TDM from fixed time slot point sampling
Business Stream, specific steps are as follows: at inlet device, by the Business Stream of each fixed-bandwidth according to fixed length slicing mode be sliced for
Several cells, respective cell are divided the core switch that cell switching system is sent to after time slot, due to data from
Inlet device has delay to outlet device, therefore nucleus module needs are synchronized with after sampling processing, realize switch
It is synchronous with inlet device, according to user configuration, the cell received " being put into " is exported into each port after multistage exchange
Each time slot, finally go to the outlet device of each port of receiving side, outlet device is sampled in correct sampling instant to be believed
Cell flow is reverted to the Business Stream of constant rate of speed further according to user configuration by member.
However, currently existing scheme needs to have unpredictable delay in cell switching system, it is both needed in the exchange of every level-one
Complicated synchronization mechanism is wanted to synchronize processing, and additions and deletions change business is required to do global consideration, causes to realize cell switching
Complexity it is high and steady degree is poor.
Summary of the invention
The embodiment of the present invention provides the management method and device of a kind of operation exchange, for solving in the prior art, every
The cell switching of level-one is required to complicated synchronization mechanism and synchronizes processing, and additions and deletions change business and are required to do the overall situation and examine
Consider, cause the complexity for realizing cell switching high and steadily and surely spends the problem of difference.
First aspect of the embodiment of the present invention provides a kind of management method of operation exchange, comprising:
Step 1 receives the cell for carrying independent numbering that core switch is sent, and by the cell according to described
In number and type of service the write-in order buffer of cell;Wherein, the order buffer is the reserved free time for being exclusively used in sequence
Space;
It whether there is the first cell in step 2, the detection order buffer;The number of first cell is same service class
Next adjacent number of the number for the cell that the last time of type takes out;First cell if it exists, then by first cell from institute
It states in order buffer and takes out, and remove the memory space for storing first cell in the order buffer;If it does not exist first
Cell repeats step 1 and step 2;
The cell taken out is reconstructed into Business Stream form, and repeats step 1 to step 3 by step 3.
With reference to first aspect, in the first possible embodiment of first aspect, the reception core switch
Before the cell for carrying independent numbering sent, the method also includes:
One section of idle continuation address space is pre-configured with as the order buffer according to business demand.
With reference to first aspect or the first possible embodiment of first aspect, second in first aspect are possible
In embodiment, the number of the cell is N, then described that the cell is written according to the number and type of service of the cell
In order buffer, comprising:
Whether the number N for detecting the cell is greater than the number of last time taking-up, identical with cell type cell
M;
If N is greater than M, the order buffer is written into the cell;Otherwise, the cell is abandoned, step 1 is repeated.
The possible embodiment of second with reference to first aspect, in the third possible embodiment of first aspect
In, it include at least two pieces of continuous address spaces in the order buffer, then it is described that the order buffer is written into the cell,
Include:
The cell is written in the first address in the first address space;Wherein, first address space is described
Address space identical with the cell type of service in order buffer;First address be first address space in
The adjacent address of second cell address;The number of second cell is adjacent with the number of the cell;
Alternatively,
The second address space is written into the cell, the second address space is address completely idle in the order buffer
Space.
With reference to first aspect, the first of first aspect is to any possible embodiment in the third, first
In the third possible embodiment of aspect, if receiving business modification instruction, the method also includes:
The first Business Stream is deleted in the business modification instruction instruction, then by the first Business Stream pair described in the order buffer
The cell answered all is deleted;
Free address space in the order buffer is organized into continuous address space;
Alternatively,
The business modification instruction instruction increases by the second Business Stream, then according to the bandwidth of second Business Stream in the row
Continuous third address space is distributed in sequence caching, the third address space is used to cache the cell of second Business Stream;
The business modification instruction includes the bandwidth of second Business Stream;
Free address space in the order buffer is organized into continuous address space.
Second aspect of the embodiment of the present invention provides a kind of management method of operation exchange, comprising:
Business Stream to be exchanged is subjected to fixed length slice, and each cell after slice according to class of service and is sliced suitable
Sequence carries out the independent multiple cells for carrying independent numbering numbered and obtained;
The multiple cell for carrying independent numbering is sent to core switch, so that core switch is by institute
It states multiple cells for carrying independent numbering and is transmitted to outlet device.
The third aspect of the embodiment of the present invention provides a kind of management method of operation exchange, comprising:
Receive multiple cells for carrying independent numbering that inlet device is sent;Wherein, the multiple to carry independent volume
Number cell be that Business Stream to be exchanged is carried out fixed length slice by the inlet device, and to each cell after slice according to industry
Business classification and slice sequence carry out what independent number obtained;
The multiple cell for carrying independent numbering is directly forwarded to outlet device.
Fourth aspect of the embodiment of the present invention provides a kind of outlet device, comprising:
Receiving module, for receiving the cell for carrying independent numbering of core switch transmission,
Processing module, for the cell to be written in order buffer according to the number and type of service of the cell;Its
In, the order buffer is the reserved free space for being exclusively used in sequence;
First detection module, for detecting in the order buffer with the presence or absence of the first cell;The volume of first cell
Number for type of service last time take out cell number next adjacent number;If the first detection module detects
There are the first cells, then the processing module takes out first cell from the order buffer, and removes the sequence
The memory space of first cell is stored in caching;If the first detection module detects that there is no the first cells, weigh
The function of the multiple receiving module, the processing module and the first detection module;
The cell that the processing module is also used to have taken out is reconstructed into Business Stream form.
In conjunction with fourth aspect, in the first possible embodiment of fourth aspect, the processing module is also used to root
One section of idle continuation address space is pre-configured with as the order buffer according to business demand.
In conjunction with the possible embodiment of the first of fourth aspect or fourth aspect, second in fourth aspect is possible
In embodiment, it is assumed that the number of the cell is N, then the processing module is specifically used for:
Whether the number N for detecting the cell is greater than the number of last time taking-up, identical with cell type cell
M;
If N is greater than M, the order buffer is written into the cell;Otherwise, the cell is abandoned, the reception is repeated
The function of module.
In conjunction with second of possible embodiment of fourth aspect, in the third possible embodiment of fourth aspect
In, include at least two pieces of continuous address spaces in the order buffer, then the processing module is specifically used for:
The cell is written in the first address in the first address space;Wherein, first address space is described
Address space identical with the cell type of service in order buffer;First address be first address space in
The adjacent address of second cell address;The number of second cell is adjacent with the number of the cell;
Alternatively,
The second address space is written into the cell, the second address space is address completely idle in the order buffer
Space.
In conjunction with fourth aspect, fourth aspect the first to any possible embodiment in the third, the 4th
In the third possible embodiment of aspect, if the receiving module receives business modification instruction, the processing module
It is also used to:
If the first Business Stream is deleted in the business modification instruction instruction, by the first Business Stream described in the order buffer
Corresponding cell is all deleted;
Free address space in the order buffer is organized into continuous address space;
Alternatively,
If the business modification instruction instruction increases by the second Business Stream, according to the bandwidth of second Business Stream described
Continuous third address space is distributed in order buffer, the third address space is used to cache the letter of second Business Stream
Member;The business modification instruction includes the bandwidth of second Business Stream;
Free address space in the order buffer is organized into continuous address space.
The 5th aspect of the embodiment of the present invention provides a kind of inlet device, comprising:
Processing module, for Business Stream to be exchanged to be carried out fixed length slice, and to each cell after slice according to industry
Business classification and slice sequence carry out the independent multiple cells for carrying independent numbering numbered and obtained;
Sending module, for the multiple cell for carrying independent numbering to be sent to core switch, so that core
The multiple cell for carrying independent numbering is transmitted to outlet device by heart switch.
The 6th aspect of the embodiment of the present invention provides a kind of core switch, comprising:
Receiving module, for receiving multiple cells for carrying independent numbering of inlet device transmission;Wherein, the multiple
The cell for carrying independent numbering is the inlet device by Business Stream progress fixed length slice to be exchanged, and to every after slice
A cell carries out what independent number obtained according to class of service and slice sequence;
Sending module, for the multiple cell for carrying independent numbering to be directly forwarded to outlet device.
The present invention provides the management method and device of a kind of operation exchange, configures dedicated order buffer in outlet device,
Corresponding inlet device only needs to carry out fixed length slice and number to the Business Stream of tape swap, is forwarded to by core switch
Outlet device, outlet device successively carry out being buffered in sorting and delay to the cell numbered in inlet device cutting according to number
It deposits, and the number order according to cell is successively taken out, then by the cell of taking-up, Business Stream form is reconstructed into, and repeat
Above-mentioned steps are completed up to all cell switchings, do not need complicated synchronization mechanism, the complexity of cell switching is effectively reduced,
And improve steady degree.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to do one simply to introduce, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow chart of the management method embodiment one of operation exchange of the present invention;
Fig. 2 is the flow chart of the management method embodiment two of operation exchange of the present invention;
Fig. 3 is the flow chart of the management method embodiment three of operation exchange of the present invention;
Fig. 4 is that order buffer free address arranges schematic diagram in the management method of operation exchange of the present invention;
Fig. 5 is the flow chart of the management method example IV of operation exchange of the present invention;
Fig. 6 is the flow chart of the management method embodiment five of operation exchange of the present invention;
Fig. 7 a is the transmission schematic diagram of one example of management method of operation exchange of the present invention;
Fig. 7 b is the sequence schematic diagram of one example of management method of operation exchange of the present invention;
Fig. 8 is the structural schematic diagram of outlet device embodiment one of the invention;
Fig. 9 is the structural schematic diagram of inlet device embodiment one of the invention;
Figure 10 is the structural schematic diagram of core of the invention switch embodiment one;
Figure 11 is the structural schematic diagram of outlet device embodiment one of the invention;
Figure 12 is the structural schematic diagram of access device embodiment one of the invention;
Figure 13 is the structural schematic diagram of core of the invention switching equipment embodiment one.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Technical solution of the present invention is applied to exchange system, realizes the packet switch between any two ports in a network, main
It will sequence for realizing TDM Business Stream and cache management.
Fig. 1 is the flow chart of the management method embodiment one of operation exchange of the present invention, as shown in Figure 1, the present embodiment is held
Row main body is outlet device, the specific steps of the management method of the operation exchange are as follows:
S101: the cell for carrying independent numbering that core switch is sent is received, and by the cell according to described
In number and type of service the write-in order buffer of cell;Wherein, the order buffer is the reserved free time for being exclusively used in sequence
Space.
In the present embodiment, the cell for carrying independent numbering is that inlet device cuts Business Stream progress fixed length to be exchanged
Piece, and independent number is carried out according to class of service and slice sequence to each cell after slice and is turned by switch
Hair.
Receive core switch send carry the cell of independent numbering before, the method also includes: according to
Business demand is pre-configured with one section of idle continuation address space as the order buffer.Order buffer is one piece of continuation address
Caching, the maximum delay difference of the size of the order buffer switching path different from current exchange network and present port
Service bandwidth is related, is exclusively used in being ranked up the cell received previously according to port bandwidth configuration when the order buffer
Spatial cache.
When outlet device receives the cell for carrying independent numbering, the number of the cell is detected in order buffer
Caching situation is written to corresponding spatial cache according to number and type of service.It is continuous that the order buffer can be divided into muti-piece
The cell of address space, different types of service can be assigned different address spaces, in the present solution, cell number and
Although buffer address be not it is completely the same, have fixed offset relationship between it.
S102: it detects in the order buffer with the presence or absence of the first cell;The number of first cell is same service class
Next adjacent number of the number for the cell that the last time of type takes out;First cell if it exists, then by first cell from institute
It states in order buffer and takes out, and remove the memory space for storing first cell in the order buffer;If it does not exist first
Cell repeats S101 and S102.
In the present embodiment, it after a cell is written to order buffer, detects and was taken in order buffer either with or without last time
The corresponding cell of next number of cell reference numeral out, if in the presence of the first cell for meeting above-mentioned condition, then by this
One cell takes out, and continues to test the cell of next number either with or without the first cell number, can continue to take out if it exists.
It i.e. during executing S102, is not limited to once take out a cell, if cell present in order buffer meets
Number order can successively take out multiple cells, until number is discontinuous, then continue to test after waiting write-in next time cell.
If there is no such cell, then outlet device continues to receive cell, according to number and type of service continue into
Row temporary cache then proceedes to cell all in detection order buffer, if the letter taken out there are any type of service last time
First cell of the same type of service of next adjacent number of the number of member, exists, takes out and empty corresponding space, weight
The process of multiple the preceding paragraph.
In the implementation procedure of S102, detecting the first cell, the first cell of taking-up and reception and cell is written be can be simultaneously
Capable treatment process is also possible to after receiving cell every time and being written into order buffer, carries out the first cell of detection, and take
Out, but not limit be only detect it is primary, can with repeated detection by the cell for meeting coded sequence cached in order buffer according to
Secondary taking-up, detailed process please refer to subsequent instance.
S103: the cell taken out is reconstructed into Business Stream form, and repeats S101 to S103;
In the present embodiment, after receiving new cell and caching every time, the cell taken out will be reconstructed, it will
The form that the cell that can be reconstructed is reconstructed into Business Stream is further processed, and then proceedes to repeat S101 to S103 to continue
The exchange of cell is handled.
The management method of operation exchange provided in this embodiment configures dedicated order buffer in outlet device, corresponding
Inlet device only needs to carry out fixed length slice and number to the Business Stream of tape swap, is forwarded to outlet dress by core switch
It sets, outlet device successively carries out being buffered in order buffer to the cell numbered in inlet device cutting, according to numbering, and according to
It is believed that the number order of member is successively taken out, then by the cell of taking-up, Business Stream form is reconstructed into, and repeat the above steps
Until all cell switchings are completed, complicated synchronization mechanism is not needed, the complexity of cell switching is effectively reduced, and is improved steady
Strong degree.
Fig. 2 is the flow chart of the management method embodiment two of operation exchange of the present invention, as shown in Fig. 2, in above-described embodiment
On the basis of, it is assumed that inlet module successively numbers the cell of cutting according to ascending order, the number of the received cell of outlet device is
N, the cell number that, last time identical as the type of service of cell that the number is N takes out from order buffer is M, the business
The specific implementation step of the management method of exchange are as follows:
S201: the cell for carrying independent numbering N that core switch is sent is received.
S202: whether the number N for detecting the cell is greater than last time taking-up, identical with cell type cell
Number M.
Wherein, the type of service for the cell that the type of service for the cell that number is N and number are M is identical, S202 be for
Detection exception, under the premise of inlet module successively numbers the cell of cutting according to ascending order, when the volume of the cell newly received
Number be less than with type of service last time take out cell number when, it is believed that be abnormal conditions, i.e. the corresponding cell of M small number is
Exported, what is received is mistake or duplicate cell certainly, if therefore N be greater than M, then follow the steps S203, otherwise execute step
Rapid S204.
S203: the order buffer then is written into the cell.S205 is executed after write-in order buffer.
It under normal circumstances, include at least two pieces of continuous address spaces in the order buffer, not limiting specifically can root
According to the size of port bandwidth configuration order buffer, order buffer can also be carried out to subregion, form multiple continuous address spaces,
It on this basis, include at least two modes by a cell write-in order buffer are as follows:
The first, will the cell be written the first address space in the first address in;Wherein, first address space
For address space identical with the cell type of service in the order buffer;First address is that first address is empty
Between in the address adjacent with the second cell address;The number of second cell is adjacent with the number of the cell;
First address space is the space for distributing to the storage of all cells of Business Stream corresponding with the cell, the
One address is then in the first address space, the address that the address of second cell adjacent with the cell number is adjacent.
If there are the second cells in the first address space, the number of the second cell is adjacent with the cell, then by institute
Cell is stated to be sequentially written in the address adjacent with second cell, if there is no then search in the first address space with institute
The number of the nearest third cell of cell number is stated, the number and the number deviant of the cell of third cell are then calculated,
Then select the address for deviating the deviant with the storage address of third cell that the cell is written.
Second, the second address space is written into the cell, the second address space is completely empty in the order buffer
Not busy address space.This kind of mode is to select one section of second new address space, and second address is written in the cell
Any address in space.
S204: the cell is abandoned.After the cell for having abandoned mistake, cell is continued to, i.e., repeatedly S201.
S205: it detects in the order buffer with the presence or absence of the first cell;The number of first cell is same service class
Next adjacent number of the number for the cell that the last time of type takes out.
First cell if it exists thens follow the steps S206, and the first cell thens follow the steps S201 if it does not exist.
S206: first cell is taken out from the order buffer, and is removed in the order buffer described in storage
The memory space of first cell.It then proceedes to receive cell, i.e. execution step S201.
S207: detect whether that there are the corresponding cells of Business Stream all to have taken out.
In the present embodiment, S206 and S207 can be it is parallel, be also possible to successively to execute, it is without limitation.
S208: there are the cells of the first Business Stream all to have taken out, then reconstructs multiple cells of first Business Stream
At Business Stream form.
It is fully received if there is the cell of the first Business Stream of some type of service, then by first Business Stream
Corresponding all cells are reconstructed, and to carry out subsequent processing, then proceed to receive cell, that is, repeat step S201 extremely
S208。
The management method of operation exchange provided in this embodiment configures dedicated order buffer in outlet device, corresponding
Inlet device only needs to carry out fixed length slice and number to the Business Stream of tape swap, is forwarded to outlet dress by core switch
It sets, outlet device successively carries out being buffered in order buffer to the cell numbered in inlet device cutting, according to numbering, and according to
It is believed that the number order of member is successively taken out, then the cell of taking-up is reconstructed, and continue to new cell, is not needed multiple
Miscellaneous synchronization mechanism, existing business sequencer procedure are not influenced by the sequence of other business, and answering for cell switching is effectively reduced
Miscellaneous degree, and improve steady degree.
Fig. 3 is the flow chart of the management method embodiment three of operation exchange of the present invention, as shown in figure 3, in above-described embodiment
On the basis of, if user needs the additions and deletions for carrying out business and changes to match, the method also includes:
S301: business modification instruction is received.
User is according to business needs, when needing to carry out the additions and deletions of business, sends business modification instruction by user port, should
Business modification instruction includes at least: the type of service of deletion;Alternatively, increased type of service and service bandwidth.It further include executing
Movement execute S302 and S303 when certain business is deleted in the instruction instruction of specific business modification.Business modification instruction instruction increases
When certain business, S304 and S305 is executed.
S302: the first Business Stream is deleted in the business modification instruction instruction, then by the first industry described in the order buffer
Business is flowed corresponding cell and is all deleted.
S303: the free address space in the order buffer is organized into continuous address space.
Business modification instruction, which is explicitly indicated, deletes the first Business Stream, all the first Business Streams pair cached in order buffer
The cell answered all is deleted, and address fragment will be occurred in original order buffer, and be cannot be used directly for biggish new industry of providing a loan
Business, it is therefore desirable to the address space of whole order buffer.
S304: the business modification instruction instruction increases by the second Business Stream, then is existed according to the bandwidth of second Business Stream
Continuous third address space is distributed in the order buffer, the third address space is for caching second Business Stream
Cell;The business modification instruction includes the bandwidth of second Business Stream.
S305: the free address space in the order buffer is organized into continuous address space.
On the basis of the above, order buffer is divided into main order buffer and standby order buffer, taken an example for illustration below pair
The arrangement method in the free address space in order buffer, Fig. 4 are order buffer in the management method of operation exchange of the present invention
Free address arranges schematic diagram, as shown in figure 4, being to change in whole process by left and right, each stage use of order buffer is shown
It is intended to, to the specific arrangement process of caching are as follows:
Firstly, such as the raw cache state of the leftmost side in Fig. 4, it is assumed that three business time-slots (former time slot 0, former time slot 1 and originals
Time slot 2) main order buffer is used, 3 block sequencing spaces are occupied respectively, are cached without fragment.
Assuming that user needs to delete former time slot 0 and former time slot 2, and new time slot 0 is configured, and the bandwidth of new time slot 0 is
Twice of former time slot 0, then by after the business deletion of former time slot 0 and former time slot 2, the buffer address of original principal space becomes fragment.
It cannot be used directly for the sequence of the business of bandwidth hair.New time slot 0 is then configured using standby Sorting space, continuous address is divided and uses
In sequence.The deletion and increase of business are had been completed at this time.New time slot 0 and 1 business of unadjusted former time slot is normal
Work.
Although business recovery, still need to arrange the address space of master cache.Otherwise, next batch service occurs to increase
When deleting, may or no continuation address block can occur can use problem.Therefore it needs in the interval of batch service additions and deletions twice, base
The fragment of spatial cache is arranged in following mechanism.
Firstly, inquiring the distribution condition of current cache.The occupancy of former time slot 1 is cached toward following and moves (this by transmitting order to lower levels
When guarantee service nondestructive).New base_addr is configured, top_addr (distributes the starting/knot in continuous order buffer space
Beam address), configure start order.New base_addr, top_addr do not come into force, therefore business sequence at this time will not receive
To any influence.
After Start order starts, the current_addr pointer that former time slot 1 sorts is detected.When pointer points to pre_
When base addr, new_top addr is enabled to come into force.It is the reason of service nondestructive at this time, since order buffer size is greater than most
Big sequence delay.Therefore when current_addr=pre_base addr, the caching used (is written with cell, wait arrange
Sequence output) it one is positioned between pre_base addr~pre_top addr, expand top_addr at this time, for coming into
The cell of sorting unit is invisible.
Since Sorting space is by temporary expansion, so sequence current_addr pointer is cumulative.Order buffer is occupied
Address field is also gradually from the section addr~pre_top addr pre_base toward the area new_base addr~new_top addr
Between move.When current_addr=new_base addr, new_base addr is enabled to come into force.Service nondestructive reason at this time
With it is similar to the above, in the case where spatial cache is sufficiently large, order buffer occupy absolutely in new_base addr~new_top
In the section addr, original order buffer section must be empty.
Further, after new_base addr and new_top addr come into force, it is slow that representative completes the former sequence of time slot 1
That deposits moves.Sorting module feedback done signal gives central processing unit (English: Central Processing Unit, letter at this time
Claim: CPU), then CPU carries out moving next time in the manner described above.After repeatedly moving, several caching fragments are removed
Caching top is moved on to, the continuously available section such as rightmost side Fig. 4 is formed and illustrates.
Based on mechanism as above, business increase/deletion, and corresponding caching defragmentation operation be it is online, it is lossless,
If demand adjusts service bandwidth, increases the bandwidth (increase of buffer sorting spatial cache) of some time slot, then first pass through caching and remove
Walk is rapid, and the caching that the caching being located at below this time slot is removed/is located at downwards above time slot is removed upwards, after the completion of moving, is pressed
Realize that Sorting space increases according to above step.
Other than above-mentioned free space Collator Mode, it is slow that business additions and deletions/adjustment can not need active and standby two block sequencing
It deposits, it is only necessary to after directly deleting some business, move caching section " squeezing out bubble " downwards.
Increase operation flow compared with aforesaid way, first carry out caching defragmentation so that caching upper interval be it is continuous/
The not non-space matched.Then new business is added into so far partial section again.It is compared with aforesaid way, reducing caching needs
It asks, but increases the delay that business needs and increase (must first wait caching to move completion).
Particularly, in practical application scene, the bandwidth of different TDM business is integral multiple relation, therefore according to continuous
Address block carries out cache management, and then the caching in block realizes the cache management of continuation address, since bandwidth is that integer several times are closed
System, therefore the order buffer size needed according to minimum after whole cache management, theoretically can reduce address and remove as one piece
The time (only being moved in block) of shifting.Still direct take-back strategy is used between block.Fragmentation problem between block is solved by chained list.
Under the support of the Collator Mode of above-mentioned free time free time to order buffer, operation exchange management method of the invention
In, order buffer has address space that is continuous, meeting bandwidth requirement, effectively accelerates business additions and deletions and changes the processing effect matched
Rate.
Fig. 5 is the flow chart of the management method example IV of operation exchange of the present invention, as shown in figure 5, the present embodiment is held
Row main body is inlet device, then the management method of the operation exchange, specific steps are as follows:
S401: carrying out fixed length slice for Business Stream to be exchanged, and to each cell after slice according to class of service and
Slice sequence carries out the independent multiple cells for carrying independent numbering numbered and obtained.
In the present embodiment, such as: the cell that TDM Business Stream is sliced as fixed length by inlet module, since TDM Business Stream is
Without burst, relative constant bandwidth, it always exists, therefore fixed length cell is also constant rate of speed, is always existed, at this point, each
Business independently carries out serial number 1,2,3 ..., using type of service mark difference between different business, is opened by header
Pin takes core switch to.Unique type of service mark is also carried in each header simultaneously.
S402: being sent to core switch for the multiple cell for carrying independent numbering, so that core exchange dress
It sets and the multiple cell for carrying independent numbering is transmitted to outlet device.
Core switch in the present embodiment does not need to do any synchronization process, directly forwards the cell received
It is handled to outlet module.
The management method of operation exchange provided in this embodiment configures dedicated order buffer, entrance dress in outlet device
It sets and only needs to carry out fixed length slice and number to the Business Stream of tape swap, outlet device is forwarded to by core switch, with
Make corresponding outlet device to the cell numbered in inlet device cutting, successively carries out being buffered in sorting according to number and delay
It deposits, and the number order according to cell is successively taken out, and Business Stream form is reconstructed into the cell of taking-up, do not need complexity
Synchronization mechanism, be effectively reduced the complexity of cell switching, and improve steady degree.
Fig. 6 is the flow chart of the management method embodiment five of operation exchange of the present invention, as shown in fig. 6, the present embodiment is held
Row main body is core switch, then the management method of the operation exchange, specific steps are as follows:
S501: multiple cells for carrying independent numbering that inlet device is sent are received.
Wherein, the multiple cell for carrying independent numbering is that the inlet device is determined Business Stream to be exchanged
Long slice, and what independent number obtained is carried out according to class of service and slice sequence to each cell after slice.
S502: the multiple cell for carrying independent numbering is directly forwarded to outlet device.
It in the present embodiment, may be due to the difference of transmission path, different business after core switch receives cell
It is out-of-order between multiple cells of type.Core switch does not need global synchronization mechanisms at this time, directly according in header
Type of service mark carry out port between forward.That is, core switch is merely ensuring that cell exit port just
True property does not need to carry out any sorting operation.
The management method of operation exchange provided in this embodiment configures dedicated order buffer in outlet device, passes through core
Heart switch receives the cell that the Business Stream to tape swap that inlet device is sent carries out fixed length slice and number obtains, and turns
Be sent to outlet device so that corresponding outlet device to inlet device cutting number cell, according to number successively into
Row is buffered in order buffer, and the number order according to cell is successively taken out, and by the cell of taking-up, business manifold is reconstructed into
Formula does not need complicated synchronization mechanism, and the complexity of cell switching is effectively reduced, and improves steady degree.
On the basis of above-mentioned Fig. 1 is to embodiment shown in fig. 6, operation exchange of the invention is taken an example for illustration below
The specific implementation of management method:
Fig. 7 a is the transmission schematic diagram of one example of management method of operation exchange of the present invention, as shown in Figure 7a, inlet device
The Business Stream by tetra- seed type of A, B, C, D is needed to be sent to outlet device, inlet device carries out fixed length slice to each Business Stream,
And individual sequential is carried out for each Business Stream and numbers to obtain:
The cell of A Business Stream: A1, A2, A3 and A4;The cell of B Business Stream: B1, B2, B3 and B4;The cell of C Business Stream:
C1, C2, C3 and C4;The cell of D Business Stream: D1, D2, D3 and D4.
Multiple cells after the Business Stream cutting of tetra- seed type of A, B, C, D are sent to core switch by inlet device,
Influence of the cell that core switch receives due to factors such as transmission paths, out-of-order between cell, core exchange dress
It sets and does not need to do any processing out-of-order cell is directly forwarded to outlet device.
The outlet device includes one piece of preconfigured order buffer, and outlet device random ordering receives cell, outlet module
Type of service mark and number based on the cell received, cell is written in order buffer.Concrete implementation mode are as follows:
Firstly, being ranked up caching distribution: order buffer may be selected to be divided into the continuous address space of muti-piece.According to different
Type of service, actual bandwidth and reorder delay demand are distributed different address spaces and are ranked up respectively, do not interfere with each other.Cell
There is fixed offset relationship, by the specific time slot in one piece of traffic assignments of each type between number and order buffer address
It number indicates, such as the slot0/slot1 in the following figure.
Secondly, start-up operation: when system has just started or business is just online, i.e., outlet device receives first cell
When, the processor of outlet device can not know the type identification and number of the cell of last output, only need to be arranged one at this time
An a upper serial number (English: Serial Number, referred to as: SN) effective status position, starting when, are defaulted as to be invalid,
When invalid, first cell of receiving is considered valid cell, then exports the cell, while by first cell SN
It records (being recorded as last SN), initial operation phase may detect that abnormal (since first cell may be disorderly
Sequence cell), but when one section after, being capable of self- recoverage.
Fig. 7 b is the sequence schematic diagram of one example of management method of operation exchange of the present invention, is row in figure as shown in Figure 7b
One section of spatial cache in sequence caching receives the processing of first cell wherein only caching the cell of a type of Business Stream
Mode is according to above-mentioned processing, and recording the cell number taken out after upper minor sort is N (current_sn), receives same service class
Next cell of type, the number for detecting the cell is N+1, if it is directly exports the cell, or detection will
The cell of the N+1 is buffered in corresponding position, then exports, if the number of the cell is not N+1, and the number of the cell is less than
N, then it is assumed that be abnormal conditions, which is abandoned;If the number of the cell is not N+1, and the number of the cell is greater than N+1
The cell is then buffered in the address of current_sn+2 of the spatial cache by (being assumed to be N+2).
Cell is continued to, if it is identical as the type of service of above-mentioned cell, and detect that the number of the cell is N+1,
Then the cell is buffered in, in current_sn+1, detects the type of service for existing in order buffer and taking out with last time at this time
It is identical, and the cell of the N+1 adjacent with number N, then the cell taking-up for N+1 will be numbered, and remove the current_sn+1's
Data, it is identical to continue to test in order buffer that there is also types of service, and the cell of the N+2 adjacent with number N+1, then will number
It is taken out for the cell of N+2, continues to test the cell that N+2 is not present in discovery, then continue to cell, and arrange according to above-mentioned caching
Sequential mode is handled.
Subsequent cell is then buffered in other spatial caches if it is other types of service, and processing mode is consistent, is pressed
According to this processing mode, no matter by how many grades of exchanges, no matter out-of-order What gives, the cell of next output is N+1 forever
Number cell, i.e., if receiving N+M cell, temporary cache, after N+M-1 cell to be output, then its is defeated by N+M cell
Out.
In above-mentioned this mode, sequencing numbers are not directly equal to the address of order buffer, but have certain entirety
The cycle period of offset, sequencing numbers and order buffer can be different or identical.
Number and downlink order buffer when inlet device is sliced are not infinitely great, therefore are actually all made of circulation
The mode of number, as long as circulating cycle time value is theoretically greater than 2 times of maximum delay difference.As shown above, when number value is super
Circulation maximum value is crossed, then overturning sequence/number since 0.
In above-mentioned scheme, when carrying out abnormality processing, need to consider the influence of numbering cycle when size judges, when
When difference is greater than the 1/2 of numbering cycle period, it is believed that counted and turned over circle, needed to recalculate difference.
The management method of operation exchange provided by the invention simplifies the complexity of cell switching system, does not need global same
Step mechanism realizes cell random ordering problem, and steady degree is high.Monolithic port has no effect on other port traffics extremely.Order buffer demand
It is only related to the bandwidth of the delay situation of current system and single port.When System Expansion business increases, not needing increase is
Has the order buffer of port in system.Reduce complexity and cost of implementation.In situation without exception, end-to-end delay prolongs for ideal
When, not since treatment mechanism introduces the additional overhead time, improving performance, and the add drop that business can also be supported online/same
The Bandwidth adjustment of one business, and support lossless.
Fig. 8 is the structural schematic diagram of outlet device embodiment one of the invention, as shown in figure 8, the outlet device 10 includes:
Receiving module 11, for receiving the cell for carrying independent numbering of core switch transmission,
Processing module 12, for the cell to be written in order buffer according to the number and type of service of the cell;
Wherein, the order buffer is the reserved free space for being exclusively used in sequence;
First detection module 13, for detecting in the order buffer with the presence or absence of the first cell;First cell
Number is next adjacent number of the number of the cell taken out the last time of same type of service;If the first detection module 13 is examined
Measure that there are the first cells, then the processing module 12 takes out first cell from the order buffer, and removes institute
State the memory space that first cell is stored in order buffer;If the first detection module 13 detects that there is no the first letters
Member then repeats the function of the receiving module 11, the processing module 12 and the first detection module 13;
The cell that the processing module 12 is also used to have taken out is reconstructed into Business Stream form.
The processing module 12 is also used to be pre-configured with one section of idle continuation address space according to business demand as institute
State order buffer.
Outlet device provided in this embodiment, for execute Fig. 1 to Fig. 4, Fig. 7 a- Fig. 7 b it is any shown in method technology
Scheme, it is similar that the realization principle and technical effect are similar, and processing module configures dedicated order buffer, corresponding entrance in outlet device
Device only needs to carry out fixed length slice and number to the Business Stream of tape swap, is forwarded to outlet device by core switch,
Processing module successively carries out being buffered in order buffer, and foundation to the cell numbered in inlet device cutting according to number
The number order of cell is successively taken out, and by the cell of taking-up, Business Stream form is reconstructed into and carries out subsequent processing, does not need
The complexity of cell switching is effectively reduced in complicated synchronization mechanism, and improves steady degree.
In the embodiment two of outlet device 10 of the present invention, it is assumed that the number of the cell is N, then the processing module 12
It is specifically used for:
Whether the number N for detecting the cell is greater than the number of last time taking-up, identical with cell type cell
M;
If N is greater than M, the order buffer is written into the cell;Otherwise, the cell is abandoned, the reception is repeated
The function of module 11.
Specifically, including at least two pieces of continuous address spaces in the order buffer, then the processing module 12 is specific
For:
The cell is written in the first address in the first address space;Wherein, first address space is described
Address space identical with the cell type of service in order buffer;First address be first address space in
The adjacent address of second cell address;The number of second cell is adjacent with the number of the cell;
Alternatively,
The second address space is written into the cell, the second address space is address completely idle in the order buffer
Space.
Optionally, if the receiving module 11 receives business modification instruction, the processing module 12 is also used to:
If the first Business Stream is deleted in the business modification instruction instruction, by the first Business Stream described in the order buffer
Corresponding cell is all deleted;
Free address space in the order buffer is organized into continuous address space;
Alternatively,
If the business modification instruction instruction increases by the second Business Stream, according to the bandwidth of second Business Stream described
Continuous third address space is distributed in order buffer, the third address space is used to cache the letter of second Business Stream
Member;The business modification instruction includes the bandwidth of second Business Stream;
Free address space in the order buffer is organized into continuous address space.
Outlet device provided in this embodiment, for execute Fig. 1 to Fig. 4, Fig. 7 a- Fig. 7 b it is any shown in method technology
Scheme, it is similar that the realization principle and technical effect are similar, is not repeating again.
Fig. 9 is the structural schematic diagram of inlet device embodiment one of the invention, as shown in figure 9, the inlet device 20, packet
It includes:
Processing module 21, for Business Stream to be exchanged to be carried out fixed length slice, and to each cell after slice according to
Class of service and slice sequence carry out the independent multiple cells for carrying independent numbering numbered and obtained;
Sending module 22, for the multiple cell for carrying independent numbering to be sent to core switch, so that
The multiple cell for carrying independent numbering is transmitted to outlet device by core switch.
Inlet device provided in this embodiment, for executing the technical solution of method shown in fig. 5, realization principle and skill
Art effect is similar, and details are not described herein.
Figure 10 is the structural schematic diagram of core of the invention switch embodiment one, as shown in Figure 10, core exchange
Device 30, comprising:
Receiving module 31, for receiving multiple cells for carrying independent numbering of inlet device transmission;Wherein, described more
A cell for carrying independent numbering is that Business Stream to be exchanged is carried out fixed length slice by the inlet device, and to slice after
Each cell carries out what independent number obtained according to class of service and slice sequence;
Sending module 32, for the multiple cell for carrying independent numbering to be directly forwarded to outlet device
Core switch provided in this embodiment, for executing the technical solution for the method that Fig. 6 shows, realization principle and
Technical effect is similar, and details are not described herein.
Figure 11 is the structural schematic diagram of outlet device embodiment one of the invention, as shown in figure 11, the outlet device 40 packet
It includes:
Receiver 41, for receiving the cell for carrying independent numbering of core switching device transmission,
Processor 42, for the cell to be written in order buffer according to the number and type of service of the cell;Its
In, the order buffer is the reserved free space for being exclusively used in sequence;
The processor 42 is also used to detect in the order buffer with the presence or absence of the first cell;The volume of first cell
Number for type of service last time take out cell number next adjacent number;If it is detected that there are the first cell,
First cell is taken out from the order buffer, and removes the storage for storing first cell in the order buffer
Space;If it is detected that the first cell is not present, the implementation procedure of the receiver 41 and processor 42 is repeated;
The cell that the processor 42 is also used to have taken out is reconstructed into Business Stream form.
The processor 42 is also used to be pre-configured with described in one section of idle continuation address space is used as according to business demand
Order buffer.
Outlet device provided in this embodiment, for execute Fig. 1 to Fig. 4, Fig. 7 a- Fig. 7 b it is any shown in method technology
Scheme, it is similar that the realization principle and technical effect are similar, and processor configures dedicated order buffer in outlet device, and corresponding entrance is set
It is standby only to need to carry out fixed length slice and number to the Business Stream of tape swap, outlet device is forwarded to by core switching device, is located
It manages device the cell numbered in access device cutting is successively carried out being buffered in order buffer according to number, and according to cell
Number order successively take out, the cell that will have been taken out is reconstructed, and does not need complicated synchronization mechanism, cell is effectively reduced
The complexity of exchange, and improve steady degree.
In the embodiment two of outlet device 40 of the present invention, it is assumed that the number of the cell is N, then the processor 42 has
Body is used for:
Whether the number N for detecting the cell is greater than the number of last time taking-up, identical with cell type cell
M;
If N is greater than M, the order buffer is written into the cell;Otherwise, the cell is abandoned, the reception is repeated
The function of device 41.
Specifically, including at least two pieces of continuous address spaces in the order buffer, then the processor 42 is specifically used
In:
The cell is written in the first address in the first address space;Wherein, first address space is described
Address space identical with the cell type of service in order buffer;First address be first address space in
The adjacent address of second cell address;The number of second cell is adjacent with the number of the cell;
Alternatively,
The second address space is written into the cell, the second address space is address completely idle in the order buffer
Space.
Optionally, if the receiver 41 receives business modification instruction, the processor 42 is also used to:
If the first Business Stream is deleted in the business modification instruction instruction, by the first Business Stream described in the order buffer
Corresponding cell is all deleted;
Free address space in the order buffer is organized into continuous address space;
Alternatively,
If the business modification instruction instruction increases by the second Business Stream, according to the bandwidth of second Business Stream described
Continuous third address space is distributed in order buffer, the third address space is used to cache the letter of second Business Stream
Member;The business modification instruction includes the bandwidth of second Business Stream;
Free address space in the order buffer is organized into continuous address space.
Outlet device provided in this embodiment, for execute Fig. 1 to Fig. 4, Fig. 7 a- Fig. 7 b it is any shown in method technology
Scheme, it is similar that the realization principle and technical effect are similar, is not repeating again.
Figure 12 is the structural schematic diagram of access device embodiment one of the invention, as shown in figure 12, the access device 50, packet
It includes:
Processor 51, for Business Stream to be exchanged to be carried out fixed length slice, and to each cell after slice according to industry
Business classification and slice sequence carry out the independent multiple cells for carrying independent numbering numbered and obtained;
Transmitter 52, for the multiple cell for carrying independent numbering to be sent to core switching device, so that core
The multiple cell for carrying independent numbering is transmitted to outlet device by heart switching equipment.
Access device provided in this embodiment, for executing the technical solution of method shown in fig. 5, realization principle and skill
Art effect is similar, and details are not described herein.
Figure 13 is the structural schematic diagram of core of the invention switching equipment embodiment one, as shown in figure 13, core exchange
Equipment 60, comprising:
Receiver 61, for receiving multiple cells for carrying independent numbering of access device transmission;Wherein, the multiple
The cell for carrying independent numbering is the access device by Business Stream progress fixed length slice to be exchanged, and to every after slice
A cell carries out what independent number obtained according to class of service and slice sequence;
Transmitter 62, for the multiple cell for carrying independent numbering to be directly forwarded to outlet device
Core switching device provided in this embodiment, for executing the technical solution for the method that Fig. 6 shows, realization principle and
Technical effect is similar, and details are not described herein.
Further, in the embodiment of above-mentioned access device and outlet device, it should be appreciated that the processor can be center
Processing unit (English: Central Processing Unit, referred to as: CPU), it can also be other general processors, number letter
Number processor (English: Digital Signal Processor, referred to as: DSP), specific integrated circuit (English:
Application Specific Integrated Circuit, referred to as: ASIC), ready-made programmable gate array (English:
Field-Programmable Gate Array, referred to as: FPGA) either other programmable logic device, discrete gate or crystalline substance
Body pipe logical device, discrete hardware components etc..General processor can be microprocessor or the processor be also possible to it is any
Conventional processor etc..The step of method in conjunction with disclosed in the embodiment of the present invention, can be embodied directly in hardware processor execution
Complete, or in processor hardware and software module combine execute completion.Software module can be located at random access memory, dodge
It deposits, read-only memory, this fields such as programmable read only memory or electrically erasable programmable memory, register are mature to deposit
In storage media.The storage medium is located at memory, and processor reads the information in memory, completes the above method in conjunction with its hardware
The step of.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence;And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (10)
1. a kind of management method of operation exchange characterized by comprising
Step 1 receives the cell for carrying independent numbering that core switch is sent, and by the cell according to the cell
Number and type of service write-in order buffer in;Wherein, the order buffer is the reserved free space for being exclusively used in sequence;
It whether there is the first cell in step 2, the detection order buffer;The number of first cell is same type of service
Next adjacent number of the number for the cell that last time takes out;First cell if it exists, then by first cell from the row
It is taken out in sequence caching, and removes the memory space for storing first cell in the order buffer;First cell if it does not exist,
Repeat step 1 and step 2;
The cell taken out is reconstructed into Business Stream form, and repeats step 1 to step 3 by step 3.
2. the method according to claim 1, wherein the reception core switch transmission carries independence
Before the cell of number, the method also includes:
One section of idle continuation address space is pre-configured with as the order buffer according to business demand.
3. method according to claim 1 or 2, which is characterized in that the number of the cell is N, then described by the letter
Member is according in number and type of service the write-in order buffer of the cell, comprising:
Whether the number N for detecting the cell is greater than the number M of last time taking-up, identical with cell type cell;
If N is greater than M, the order buffer is written into the cell;Otherwise, the cell is abandoned, step 1 is repeated.
4. according to the method described in claim 3, it is characterized in that, including at least two pieces of continuous addresses in the order buffer
Space, then it is described that the order buffer is written into the cell, comprising:
The cell is written in the first address in the first address space;Wherein, first address space is the sequence
Address space identical with the cell type of service in caching;First address is in first address space with second
The adjacent address of cell address;The number of second cell is adjacent with the number of the cell;
Alternatively,
The second address space is written into the cell, the second address space is that address completely idle in the order buffer is empty
Between.
5. method according to claim 1 or 2, which is characterized in that if receiving business modification instruction, the method is also
Include:
The first Business Stream is deleted in the business modification instruction instruction, then the first Business Stream described in the order buffer is corresponding
Cell is all deleted;
Free address space in the order buffer is organized into continuous address space;
Alternatively,
The business modification instruction instruction increases by the second Business Stream, then slow in the sequence according to the bandwidth of second Business Stream
The middle continuous third address space of distribution is deposited, the third address space is used to cache the cell of second Business Stream;It is described
Business modification instruction includes the bandwidth of second Business Stream;
Free address space in the order buffer is organized into continuous address space.
6. a kind of outlet device characterized by comprising
Receiving module, for receiving the cell for carrying independent numbering of core switch transmission,
Processing module, for the cell to be written in order buffer according to the number and type of service of the cell;Wherein, institute
Stating order buffer is the reserved free space for being exclusively used in sequence;
First detection module, for detecting in the order buffer with the presence or absence of the first cell;The number of first cell is
With next adjacent number of the number of the cell of the last time taking-up of type of service;If the first detection module detects exist
First cell, then the processing module takes out first cell from the order buffer, and removes the order buffer
The memory space of middle storage first cell;If the first detection module detects that the first cell is not present, repeatedly institute
State the function of receiving module, the processing module and the first detection module;
The cell that the processing module is also used to have taken out is reconstructed into Business Stream form.
7. device according to claim 6, which is characterized in that the processing module is also used to be matched in advance according to business demand
One section of idle continuation address space is set as the order buffer.
8. device according to claim 6 or 7, which is characterized in that assuming that the number of the cell is N, then the processing
Module is specifically used for:
Whether the number N for detecting the cell is greater than the number M of last time taking-up, identical with cell type cell;
If N is greater than M, the order buffer is written into the cell;Otherwise, the cell is abandoned, the receiving module is repeated
Function.
9. device according to claim 8, which is characterized in that include at least two pieces of continuous addresses in the order buffer
Space, then the processing module is specifically used for:
The cell is written in the first address in the first address space;Wherein, first address space is the sequence
Address space identical with the cell type of service in caching;First address is in first address space with second
The adjacent address of cell address;The number of second cell is adjacent with the number of the cell;
Alternatively,
The second address space is written into the cell, the second address space is that address completely idle in the order buffer is empty
Between.
10. device according to claim 6 or 7, which is characterized in that refer to if the receiving module receives business modification
It enables, then the processing module is also used to:
If the first Business Stream is deleted in the business modification instruction instruction, and the first Business Stream described in the order buffer is corresponding
Cell all delete;
Free address space in the order buffer is organized into continuous address space;
Alternatively,
If the business modification instruction instruction increases by the second Business Stream, according to the bandwidth of second Business Stream in the sequence
Continuous third address space is distributed in caching, the third address space is used to cache the cell of second Business Stream;Institute
State the bandwidth that business modification instruction includes second Business Stream;
Free address space in the order buffer is organized into continuous address space.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510144310.5A CN104699632B (en) | 2015-03-30 | 2015-03-30 | The management method and device of operation exchange |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510144310.5A CN104699632B (en) | 2015-03-30 | 2015-03-30 | The management method and device of operation exchange |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104699632A CN104699632A (en) | 2015-06-10 |
CN104699632B true CN104699632B (en) | 2019-02-26 |
Family
ID=53346781
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510144310.5A Active CN104699632B (en) | 2015-03-30 | 2015-03-30 | The management method and device of operation exchange |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104699632B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101146029A (en) * | 2006-09-13 | 2008-03-19 | 华为技术有限公司 | A packet resorting method and system |
CN101164367A (en) * | 2005-11-16 | 2008-04-16 | 华为技术有限公司 | Cell reordering method, and cell processing method and apparatus the same |
CN102447607A (en) * | 2010-10-08 | 2012-05-09 | 中兴通讯股份有限公司 | Method, device and system adopting address redundancy technique to realize packet regrouping |
CN103297350A (en) * | 2013-05-24 | 2013-09-11 | 华为技术有限公司 | Implementing method and switching equipment of cell switching system |
CN103428039A (en) * | 2012-05-18 | 2013-12-04 | 中兴通讯股份有限公司 | Method and device for detecting IMA delay difference |
-
2015
- 2015-03-30 CN CN201510144310.5A patent/CN104699632B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101164367A (en) * | 2005-11-16 | 2008-04-16 | 华为技术有限公司 | Cell reordering method, and cell processing method and apparatus the same |
CN101146029A (en) * | 2006-09-13 | 2008-03-19 | 华为技术有限公司 | A packet resorting method and system |
CN102447607A (en) * | 2010-10-08 | 2012-05-09 | 中兴通讯股份有限公司 | Method, device and system adopting address redundancy technique to realize packet regrouping |
CN103428039A (en) * | 2012-05-18 | 2013-12-04 | 中兴通讯股份有限公司 | Method and device for detecting IMA delay difference |
CN103297350A (en) * | 2013-05-24 | 2013-09-11 | 华为技术有限公司 | Implementing method and switching equipment of cell switching system |
Also Published As
Publication number | Publication date |
---|---|
CN104699632A (en) | 2015-06-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101607180B1 (en) | Method, apparatus and system for packet reassembly and reordering | |
CA2015514C (en) | Packet switching system having bus matrix switch | |
EP2613479A1 (en) | Relay device | |
US6907002B2 (en) | Burst switching in a high capacity network | |
EP0471344B1 (en) | Traffic shaping method and circuit | |
JP3783731B2 (en) | Fast switching network structure | |
US7940666B2 (en) | Communication node architecture in a globally asynchronous network on chip system | |
US20130250792A1 (en) | Router | |
JP4763405B2 (en) | Network-on-chip semi-automatic communication architecture for data flow applications | |
JP4480845B2 (en) | TDM switch system with very wide memory width | |
US20050259672A1 (en) | Method to improve forwarding information base lookup performance | |
JP6978596B2 (en) | Data transmission | |
TW201108668A (en) | Buffered crossbar switch system | |
US6430180B1 (en) | Method and apparatus for switching data between bitstreams of a time division multiplexed network | |
CN105573711B (en) | A kind of data cache method and device | |
WO2005093591A1 (en) | Integrated circuit and method for transaction abortion | |
JP2003037572A (en) | Scheduling system | |
CN104699632B (en) | The management method and device of operation exchange | |
JP2003069630A (en) | Data transfer method, data transfer device, and program | |
JP2000324115A (en) | Output buffer atm exchange and multiple address control method | |
CN113434455B (en) | Optical fiber interface data cache management method based on FPGA | |
Baiocchi et al. | The ACCI access protocol for a twin bus ATM metropolitan area network | |
CN113490084B (en) | FC-AE exchanger ultra-bandwidth transmission method supporting priority scheduling | |
CN103401812B (en) | Message processing method and device based on 2 groups of Message processing engines | |
JP4764837B2 (en) | BAND CONTROL SYSTEM, BAND CONTROL PROGRAM, AND BAND CONTROL 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |