CN109450816A - A kind of array dispatching method, device, the network equipment and storage medium - Google Patents
A kind of array dispatching method, device, the network equipment and storage medium Download PDFInfo
- Publication number
- CN109450816A CN109450816A CN201811388030.9A CN201811388030A CN109450816A CN 109450816 A CN109450816 A CN 109450816A CN 201811388030 A CN201811388030 A CN 201811388030A CN 109450816 A CN109450816 A CN 109450816A
- Authority
- CN
- China
- Prior art keywords
- core
- forwarding
- forwarding core
- buffer queue
- consumption value
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of array dispatching method, device, the network equipment and storage mediums.This method is applied to the network equipment, and the network equipment includes control core and multiple forwarding cores, and network equipments configuration has multiple buffer queues, and each forwarding verification in multiple forwarding cores should have at least one buffer queue, and a buffer queue only corresponds to a forwarding core;Method comprises determining that control core determines the second forwarding core of processing capacity from each forwarding core in multiple forwarding cores in addition to the corresponding first forwarding core of the first buffer queue there are when the first buffer queue of message accumulation in multiple buffer queues;The configuration relation that core deletes the first buffer queue and the first forwarding core is controlled, the configuration relation of the first buffer queue and the second forwarding core is established, the second forwarding core is made to handle the message in the first buffer queue.This method reduce the risks of forwarding core processing capacity overload, realize the data stream scheduling between forwarding core, avoid unnecessary data flow packet loss problem.
Description
Technical field
The invention belongs to fields of communication technology, and in particular to a kind of array dispatching method, device, the network equipment and storage are situated between
Matter.
Background technique
In the multi-core network processor framework for having Message processing demand, current mainstream technology scheme are as follows: with ether number
For, after the network interface of processor receives Ethernet message, different messages is sent into different queues according to configuration (can
The message queue to be entered is determined with the message five-tuple HASH value received according to interface, wherein five-tuple HASH value is will to mark
The five-tuple (source IP, destination IP, IP agreement value, source port number, destination slogan) of quasi- Ethernet message is calculated by HASH algorithm
HASH value out), forwarding core again from queue obtain message handled (can configure forwarding core and queue mapping relations,
Some or certain queues are such as assigned to some or certain forwarding cores are handled).
According to above-mentioned network processing unit to the process flow of Ethernet message it is found that processor provides flexible data processing
Mechanism can also not only be handled queue assignment to different forwarding core according to the queue of configuration selection data receiver, but
It is that this configuration cannot be carried out arbitrarily, needs to carry out different configurations according to the difference of application scenarios, by taking router as an example, due to
Router belongs to network intermediary device, realizes the routing forwarding of data, therefore is that cannot change data flow (to have under normal circumstances
The message of identical five-tuple HASH value) sequence, but if the data of same data stream are handled by multiple forwarding cores, just can not
Guarantee the sequence of data flow, therefore for identical data flow, the principle based on order-preserving can only be distributed data flow to same
Queue and by the processing of the same forwarding core, i.e., identical data flow can only enter identical queue, and identical queue can only be by
Identical forwarding core processing.
Under normal circumstances, the number of queue is more much more than forwarding the number of core, therefore certainly exists the more of multiple queues
Data stream is assigned to the case where the processing of the same forwarding core, is more than when handling forwarding nuclear resource consumed by these data flows
It will lead to data flow packet loss when the processing capacity of single forwarding core.
Summary of the invention
In consideration of it, the purpose of the present invention is to provide a kind of array dispatching method, device, the network equipment and storage medium,
To effectively improve the above problem.
The embodiment of the present invention is achieved in that
In a first aspect, the embodiment of the present application provides a kind of array dispatching method, it is applied to the network equipment, the network is set
Standby includes multi-core processor, and the multi-core processor includes that control core and multiple forwarding cores, the network equipments configuration have multiple
Buffer queue, the multiple each forwarding verification forwarded in core should have at least one buffer queue, and a buffer queue is only
A corresponding forwarding core;The described method includes: determining the first caching team in the multiple buffer queue there are message accumulation
When column, each turn for controlling core from the multiple forwarding core in addition to the corresponding first forwarding core of first buffer queue
The second forwarding core of processing capacity is determined in hair core;The control core deletes first buffer queue and described first turn
The configuration relation for sending out core establishes the configuration relation of first buffer queue and the second forwarding core, makes second forwarding
Core handles the message in first buffer queue.In the embodiment of the present application, by having checked whether that message accumulation occurs in queue,
When thering is queue message accumulation i.e. corresponding forwarding core (the first forwarding core) occur to overload, determined from remaining forward in core
There is the forwarding core (the second forwarding core) of processing capacity, and the queue for having message to accumulate is rescheduled at the second forwarding core
Reason thus reduces the risk of forwarding core processing capacity overload, improves system to the response speed of burst flow, realizes and turn
The data stream scheduling between core is sent out, unnecessary data flow packet loss problem is avoided.
A kind of optional embodiment with reference to first aspect, the control core remove described first from the multiple forwarding core
The second forwarding core of processing capacity is determined in each forwarding core outside the corresponding first forwarding core of buffer queue, comprising: institute
State control core by consumption value of the first forwarding core in preset duration respectively in the multiple forwarding core except described first
Consumption value of each forwarding core in the preset duration outside forwarding core is subtracted each other, and obtains multiple subtracting each other result;The control
Core processed chooses maximum subtract each other as a result, subtracting each other result as target from the multiple subtract each other in result;The control core determines institute
It states target and subtracts each other processing consumption value of the result greater than the first forwarding core that first buffer queue occupies;The control core
The target is subtracted each other into the corresponding forwarding core of result, as the second forwarding core.In the embodiment of the present application, by by first turn
Send out consumption value of consumption value of the core in preset duration respectively with each forwarding core in addition to the first forwarding core in preset duration
Subtracted each other, and choose maximum value in result as objective result from multiple subtract each other, and determines that objective result is greater than the first caching
When the processing consumption value for the first forwarding core that queue occupies, by the corresponding forwarding core of the objective result, core is forwarded as second, with
Guarantee that the second forwarding core there are enough processing capacity places to go when the queue scheduling that there will be message to accumulate is handled to the second forwarding core
The problem of managing the message in the queue, avoid because the second forwarding core processing capacity is insufficient, leading to data flow packet loss.
A kind of optional embodiment with reference to first aspect, in the control core by the first forwarding core in preset duration
Interior consumption value is respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core in the preset duration
Interior consumption value is subtracted each other, before obtaining multiple the step of subtracting each other result, the method also includes: the control core obtains institute
State consumption value of each forwarding core in multiple forwarding cores in the preset duration;The control core is getting each forwarding
When the consumption value of core, determine that there are the first buffer queues of message accumulation in the multiple buffer queue.The application is real
It applies in example, first obtains consumption value of each forwarding core in multiple forwarding cores in preset duration, then check whether exist again
The queue of message accumulation, compared to first checking whether with the presence of queue message accumulation, when determining the queue accumulated there are message,
For the scheme for obtaining the consumption value of each forwarding core again, the effective influence to the second forwarding core can be reduced, avoid because when
Difference is imitated, second finally determined is caused to forward core different.
A kind of optional embodiment with reference to first aspect, the control core is by the first forwarding core in preset duration
Consumption value respectively with it is the multiple forwarding core in except it is described first forwarding core in addition to each forwarding core in the preset duration
Consumption value subtracted each other, obtain multiple subtracting each other result, comprising: the control core obtains each of the multiple forwarding core and turns
Send out consumption value of the core in the preset duration;It is described control core by it is described first forwarding core consumption value respectively with it is the multiple
The consumption value of each forwarding core in forwarding core in addition to the first forwarding core is subtracted each other, and obtains multiple subtracting each other result.This Shen
Please be in embodiment, when determining has queue message accumulation occur, then remove to obtain each forwarding cores in multiple forwarding cores default
Consumption value in duration compared to first obtaining the consumption value of each forwarding core, then checks whether that there are the queues that message is accumulated
For scheme, it is possible to reduce workload avoids after having got the consumption value of each forwarding core, the case where but accumulation without queue
Occur.
A kind of optional embodiment with reference to first aspect, the control core obtain each of the multiple forwarding core and turn
Send out consumption value of the core in the preset duration, comprising: the control core obtains each forwarding core in the multiple forwarding core
The message data obtained from corresponding buffer queue in the preset duration respectively counted;The control core is based on the report
Literary data determine the consumption value of each forwarding core.In the embodiment of the present application, respectively counted by each forwarding core of acquisition
The message data obtained from corresponding buffer queue in the preset duration, and each consumption for forwarding core is calculated with this
Value, ensure that the reliability and accuracy of calculated result, and then ensure that the second forwarding core determined based on these consumption values
Reliability, the risk of single forwarding core processing capacity overload is reduced with this, avoids unnecessary data flow packet loss problem.
A kind of optional embodiment with reference to first aspect, the control core obtain each of the multiple forwarding core and turn
The message data obtained from corresponding buffer queue in the preset duration that hair core respectively counts, comprising: the control core
The state of periodically-varied global variable switch is at first state or the second state, wherein in the first state,
Each forwarding core in the multiple forwarding core respectively counts the report obtained from corresponding buffer queue in the preset duration
Literary data;In second state, the control core obtains the message of each forwarding core feedback in the multiple forwarding core
Data.During the application is implemented, control core is become when obtaining the message data of each forwarding nuclear statistics by the periodically-varied overall situation
The state of switch is measured, so that each forwarding core decides whether to count message data according to the state that global variable switchs,
Ensure that each consistency of the forwarding core in accounting message data in time, so ensure that calculated result accuracy and can
By property, while the state that need to be only switched by changing global variable, it can be realized to the message data of each forwarding nuclear statistics
It obtains, simplifies control flow.
Second aspect, the embodiment of the present application also provides a kind of network equipment, including multi-core processor, the multicore processing
Device includes control core and multiple forwarding cores, and the network equipments configuration has multiple buffer queues, every in the multiple forwarding core
A forwarding verification should have at least one buffer queue, and a buffer queue only corresponds to a forwarding core;The control core, is used for
It determines in the multiple buffer queue there are when the first buffer queue of message accumulation, is removed from the multiple forwarding core described
The second forwarding core of processing capacity is determined in each forwarding core outside the corresponding first forwarding core of first buffer queue;It is described
Core is controlled, is also used to delete the configuration relation of first buffer queue and the first forwarding core, establishes first caching
The configuration relation of queue and the second forwarding core makes the second forwarding core handle the message in first buffer queue;
Each forwarding core in the multiple forwarding core, for handling the message at least one corresponding buffer queue.
In conjunction with a kind of optional embodiment of second aspect, the control core is also used to the first forwarding core pre-
If the consumption value in duration is respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core described pre-
If the consumption value in duration is subtracted each other, obtain multiple subtracting each other result;The control core is also used to subtract each other result from the multiple
It is middle to choose maximum subtract each other as a result, subtracting each other result as target;It is big to be also used to determine that the target subtracts each other result for the control core
In the processing consumption value for the first forwarding core that first buffer queue occupies;The control core is also used to the mesh
Mark subtracts each other the corresponding forwarding core of result, as the second forwarding core.
In conjunction with a kind of optional embodiment of second aspect, the control core is also used to obtain in the multiple forwarding core
Consumption value of each forwarding core in the preset duration;The control core is also used in the institute for getting each forwarding core
When stating consumption value, determine that there are the first buffer queues of message accumulation in the multiple buffer queue.
In conjunction with a kind of optional embodiment of second aspect, the control core is also used to obtain in the multiple forwarding core
Consumption value of each forwarding core in the preset duration;The control core is also used to the consumption of the first forwarding core
Consumption value of the value respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core is subtracted each other, and is obtained more
It is a to subtract each other result.
In conjunction with a kind of optional embodiment of second aspect, the control core is also used to obtain in the multiple forwarding core
The message data obtained from corresponding buffer queue in the preset duration that respectively counts of each forwarding core;The control
Core is also used to determine the consumption value of each forwarding core based on the message data.
In conjunction with a kind of optional embodiment of second aspect, the control core is also used to periodically-varied global variable switch
State be at first state or the second state, wherein in the first state, each of the multiple forwarding core
Forwarding core respectively counts the message data obtained from corresponding buffer queue in the preset duration;In second state
When, the control core is also used to obtain the message data of each forwarding core feedback in the multiple forwarding core.
The third aspect, the embodiment of the present application also provides a kind of queue scheduling device, applied to including multi-core processor
The network equipment, the multi-core processor include multiple forwarding cores, and the network equipments configuration has multiple buffer queues, the multiple
Each forwarding verification in forwarding core should have at least one buffer queue, and a buffer queue only corresponds to a forwarding core;Institute
It states device and comprises determining that module and configuration module;Determining module, for determining, there are messages in the multiple buffer queue
When the first buffer queue of accumulation, from the multiple forwarding core in addition to the corresponding first forwarding core of first buffer queue
The second forwarding core of processing capacity is determined in each forwarding core;Configuration module, for delete first buffer queue with
The configuration relation of the first forwarding core establishes the configuration relation of first buffer queue and the second forwarding core, makes institute
It states the second forwarding core and handles message in first buffer queue.
In conjunction with a kind of optional embodiment of the third aspect, the determining module is also used to the control core for described
One consumption value of the forwarding core in preset duration respectively with each turn in the multiple forwarding core in addition to the first forwarding core
It sends out consumption value of the core in the preset duration to be subtracted each other, obtains multiple subtracting each other result;The control core is from the multiple phase
Subtract and chooses maximum subtract each other as a result, subtracting each other result as target in result;It is big that the control core determines that the target subtracts each other result
In the processing consumption value for the first forwarding core that first buffer queue occupies;The target is subtracted each other knot by the control core
The corresponding forwarding core of fruit, as the second forwarding core.
In conjunction with a kind of optional embodiment of the third aspect, described device further include: obtain module and the second determining mould
Block;Module is obtained, obtains each forwarding core in the multiple forwarding core in the preset duration for the control core
Consumption value;Second determining module determines described more for the control core when getting the consumption value of each forwarding core
There are the first buffer queues of message accumulation in a buffer queue.
In conjunction with a kind of optional embodiment of the third aspect, the determining module is also used to described in the control core acquisition
Consumption value of each forwarding core in the preset duration in multiple forwarding cores;The control core forwards core for described first
Consumption value of the consumption value respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core is subtracted each other, and is obtained
Subtract each other result to multiple.
In conjunction with a kind of optional embodiment of the third aspect, the acquisition module or the determining module are also used to described
What each forwarding core that control core obtains in the multiple forwarding core respectively counted caches team from corresponding in the preset duration
The message data obtained in column;The control core determines the consumption value of each forwarding core based on the message data.
In conjunction with a kind of optional embodiment of the third aspect, the acquisition module or the determining module are also used to described
The state that control nuclear periodicity changes global variable switch is at first state or the second state, wherein described first
When state, each forwarding core in the multiple forwarding core is respectively counted to be obtained from corresponding buffer queue in the preset duration
The message data taken;In second state, the control core obtains each forwarding core feedback in the multiple forwarding core
Message data.
Fourth aspect, the embodiment of the present application also provides a kind of storage mediums, are stored thereon with computer program, the meter
Calculation machine program executes first aspect and/or a kind of optional embodiment with reference to first aspect mentions when being run by processor
The method of confession.
Other features and advantages of the present invention will be illustrated in subsequent specification, also, partly be become from specification
It is clear that being understood by implementing the embodiment of the present invention.The objectives and other advantages of the invention can be by written
Specifically noted structure is achieved and obtained in specification, claims and attached drawing.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.By the way that shown in attached drawing, above and other purpose of the invention, feature and advantage will be more clear.In whole
Identical appended drawing reference indicates identical part in attached drawing.Attached drawing, emphasis deliberately are not drawn by actual size equal proportion scaling
It is to show the gist of the present invention.
Fig. 1 shows the structural schematic diagram of the multi-core processor framework in the network equipment provided in an embodiment of the present invention.
Fig. 2 shows a kind of flow diagrams of array dispatching method provided in an embodiment of the present invention.
Fig. 3 shows the flow diagram of the step S101 in Fig. 2 provided in an embodiment of the present invention.
Fig. 4 shows a kind of module diagram of queue scheduling device provided in an embodiment of the present invention.
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.The present invention being usually described and illustrated herein in the accompanying drawings is implemented
The component of example can be arranged and be designed with a variety of different configurations.
Therefore, the detailed description of the embodiment of the present invention provided in the accompanying drawings is not intended to limit below claimed
The scope of the present invention, but be merely representative of selected embodiment of the invention.Based on the embodiments of the present invention, this field is common
Technical staff's every other embodiment obtained without creative efforts belongs to the model that the present invention protects
It encloses.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.
In the description of the present invention, it should be noted that term " first ", " second ", " third " etc. are only used for distinguishing and retouch
It states, is not understood to indicate or imply relative importance.Furthermore term "and/or" in the application, only a kind of description is closed
Join the incidence relation of object, indicates may exist three kinds of relationships, for example, A and/or B, can indicate: individualism A is deposited simultaneously
In A and B, these three situations of individualism B.
What the network equipment used at present is substantially multi-core processor, and processor architecture is as shown in Figure 1.With ether data
For, after the network interface of processor receives Ethernet message, different messages is sent into different queues according to configuration (can be with
The message five-tuple HASH value received according to interface determines the message queue to be entered), forwarding core obtains message from queue again
Handled (can with dynamic configuration forward core and queue mapping relations, such as by some or certain queues be assigned to some or
The certain forwarding core processing of person).Wherein, five-tuple HASH value is by five-tuple (source IP, destination IP, the IP association of standard Ethernet message
View value, source port number, destination slogan) pass through the calculated HASH value of HASH algorithm.
According to above-mentioned network processing unit to the process flow of Ethernet message it is found that processor provides flexible data processing
Mechanism can also not only be handled queue assignment to different forwarding core according to the queue of configuration selection data receiver, but
It is that this configuration cannot be carried out arbitrarily, needs to carry out different configurations according to the difference of application scenarios.By taking router as an example, due to
Router belongs to intermediary network device, realizes the routing forwarding of data, therefore is that cannot change data flow (to have under normal circumstances
The message of identical five-tuple HASH value) sequence, but if the data of same data stream are handled by multiple forwarding cores, just can not
Guarantee the sequence of data flow, therefore for identical data flow, the principle based on order-preserving can only be distributed data flow to same
Queue and by the same forwarding core processing.
The order-preserving processing for realizing data flow needs to configure according to hardware meter first in multi-core network processor framework
The message five-tuple HASH value calculated carries out the distribution of queue, and identical five-tuple HASH value enters identical queue, in this way
One data stream is only possible to enter a queue, in addition, configuration same queue can only be handled by identical forwarding core, i.e., one
Queue can only be handled by a forwarding core, thus achieved the purpose that a data stream is handled by a forwarding core, realized
The order-preserving of data flow.
Under normal circumstances, the number of queue is more much more than forwarding the number of core, and has identical five-tuple HASH value
Message has very much, therefore a plurality of data flow for certainly existing multiple queues is assigned to the same forwarding core the case where handling, when
It will lead to data flow when handling the processing capacity for forwarding nuclear resource to be more than single forwarding core consumed by these data flows to lose
Packet.
Present inventor has found when studying the above problem: mainstream scheme has problems in that queue and forwarding core
Mapping relations be it is fixed, when distribute to forwarding core all queues in data traffic summation be more than forward core processing
Flow scheduling to other free time can not be forwarded core by way of dynamic dispatching under the premise of data flow order-preserving by ability.Its
In, it should be noted that the reason of for defect present in above scheme and leading to defect is inventor by real
Trample and obtained after carefully studying as a result, therefore, the discovery procedure of the above problem and the hereinafter embodiment of the present invention are for upper
The solution that the problem of stating is proposed all should be the contribution that inventor makes the present invention in process of the present invention.
In consideration of it, as shown in Fig. 2, the embodiment of the present application provides a kind of array dispatching method, applied to having at message
The network equipment for having multi-core processor of reason demand, such as router, firewall.Wherein, multi-core processor includes control core
With multiple forwarding cores.The network equipments configuration has multiple buffer queues, and each forwarding verification in the multiple forwarding core should have
At least one buffer queue, and a buffer queue only corresponds to a forwarding core.It is said below in conjunction with step shown in Fig. 2
It is bright.
Step S101: determine in the multiple buffer queue there are message accumulation the first buffer queue when, the control
Core processed is determined from each forwarding core in the multiple forwarding core in addition to the corresponding first forwarding core of first buffer queue
There is the second forwarding core of processing capacity out.
All buffer queues of nuclear periodicity poll are controlled, have checked whether message accumulation.Determining that it is tired that there are messages
When long-pending first buffer queue, namely show that the corresponding first forwarding core of the first buffer queue is currently at overload state, this
When, control core just determines second turn of processing capacity from each forwarding core in multiple forwarding cores in addition to the first forwarding core
Core is sent out, is handled to reschedule the queue for having message to accumulate to the second forwarding core, forwarding core processing is thus reduced
The risk of ability overload improves system to the response speed of burst flow, realizes the data stream scheduling between forwarding core, keep away
Exempt from unnecessary data flow packet loss problem.
Second turn of processing capacity is being determined from each forwarding cores in multiple forwarding cores in addition to the first forwarding core
When sending out core, it can be according to the consumption value within a preset time of each forwarding core and determine, be also possible to according to each forwarding
The memory accounting of core and/or memory size determine.As an alternative embodiment, below with reference to step shown in Fig. 3
To illustrate that control core determines the detailed process of the second forwarding core.
Step S201: the control core by consumption value of the first forwarding core in preset duration respectively with it is the multiple
Consumption value of each forwarding core in the preset duration in forwarding core in addition to the first forwarding core is subtracted each other, and is obtained more
It is a to subtract each other result.
For example, control core by the first consumption value of the forwarding core in preset duration respectively in N number of forwarding core except first turn
Consumption value of each forwarding core in preset duration outside hair core is subtracted each other, and is obtained N-1 and is subtracted each other result.Assuming that first turn
Sending out consumption value of the core in preset duration is E1, remaining consumption value of each forwarding core in preset duration is respectively E2,
E3... ... En, then the N-1 obtained is a to subtract each other result are as follows: (E1- E2), (E1- E3), (E1- E4) ... ... (E1- EN-1),
(E1- En).Wherein, (E1- E2), (E1- E3), (E1- E4) ... ... (E1- EN-1), (E1- En) these differences have reacted turn
Send out the free processing capacity of core.Wherein, N is the integer more than or equal to 2.
Step S202: the control core chooses maximum subtract each other as a result, as target phase from the multiple subtract each other in result
Subtract result.
For example, the control core chooses maximum phase from a subtract each other of N-1 after obtaining N-1 and subtracting each other result in result
Subtract as a result, subtracting each other result as target.Core is controlled from (E1- E2), (E1- E3), (E1- E4) ... ... (E1- EN-1), (E1?
En) choose maximum difference in this N-1 difference and subtract each other as target as a result, being assumed to be (E1- E2)。
Step S203: it is described greater than what first buffer queue occupied that the control core determines that the target subtracts each other result
The processing consumption value of first forwarding core.
(it is assumed to be (E obtaining target and subtract each other result1- E2)) after, control core judges that the target subtracts each other whether result is greater than
The processing consumption value for the first forwarding core that first buffer queue occupies is determining that target subtracts each other result and account for greater than the first buffer queue
When the processing consumption value of the first forwarding core, step S204 is executed, it is on the contrary then do not handle, wait next polling cycle arrival
The consumption value for recalculating each forwarding core again, is then judged again.Due to (E1- E2), (E1- E3), (E1- E4) ... ...
(E1- EN-1), (E1- En) these differences have reacted the free processing capacity of corresponding forwarding core, when maximum value therein is greater than the
When the processing consumption value for the first forwarding core that one buffer queue occupies, show accordingly to forward core (such as E2Corresponding forwarding core) there is foot
Enough processing capacities remove the message of the first buffer queue of processing.
Step S204: the target is subtracted each other the corresponding forwarding core of result by the control core, as the second forwarding core.
After obtaining target and subtracting each other result, target is subtracted each other into the corresponding forwarding core of result, as the second forwarding core, it is assumed that mesh
It is (E that mark, which subtracts each other result,1- E2), then the second forwarding core is E2Corresponding forwarding core.
Wherein, how control core obtains the consumption value of each forwarding core within a preset time, due to forwarding the consumption of core
Value is closely bound up with the ability for handling data flow, and the ability for handling data flow is both related with the uninterrupted of data flow, also with
The type of data flow is related.(different types of data flow, process flow is different, and the processor resource of consumption is also different).Therefore
Quantitative analysis is carried out firstly the need of to forwarding nuclear resource consumed by different type message, how that obtains processing different type number
The resource of consumption needed for forwarding core according to stream? single can be played respectively for different types of data and cross current-carrying capacity, i.e. data flow
Flow will be more than the processing capacity of single core also only can be by one since a data stream can only enter a queue
Forward core processing, can be obtained by the maximum capacity of single forwarding core processing respective type data in this way, here with IPv4,
IPv6, wide area network POS three types message for, (pass through buffer queue by beating current-carrying capacity to these three messages respectively
Accumulation state judge whether to overload), that is, the single data traffic for corresponding to message need be more than single forwarding core processing capacity,
Single message single-turn hair core processor of three kinds of different type messages available in this way consumes resource, here labeled as Δ tipv4,
Δtipv6, Δ tpos.For example, if individually the processing capacity of forwarding core is 800Kpps, each message needs for IPv4
The resource of consumption forwarding core is 1/800000s, and marking the consumption resource here is Δ tipv4, similarly for IPv6 and wide area network
POS message obtains corresponding single message and individually forwards the relative resource consumption of core to be Δ tipv6With Δ tpos。
Wherein, it should be noted that the relative resource consumption needs of the single message of every kind of message are calculated in advance.This
Sample is when calculating consumption value of each forwarding core in preset duration, it is only necessary to count each forwarding core in preset duration from right
The message data obtained in buffer queue is answered, for example, calculating consumption value E of the forwarding core 1 in preset duration1When, it obtains
The message data for taking forwarding core 1 to obtain from corresponding buffer queue in preset duration, it is assumed that the forwarding corresponding buffer queue of core 1
For queue 1 and queue 2, if IPv4, IPv6, POS message that forwarding core 1 is obtained from queue 1 are calculated as respectively in preset duration
Q1ipv4、Q1ipv6、Q1pos, similarly, IPv4, IPv6, POS message that forwarding core 1 is obtained from queue 2 are calculated as Q2 respectivelyipv4、
Q2ipv6、Q2pos, then forwarding the consumption value of core 1 within a preset time is (Q1ipv4+Q2ipv4) * Δ tipv4+(Q1ipv6+Q2ipv6) *
Δtipv6+(Q1pos+Q2pos) * Δ tpos.Similarly, consumption value E of the forwarding core 2 in preset duration is being calculated2When, obtain forwarding
The message data that core 2 is obtained from corresponding buffer queue in preset duration, it is assumed that the forwarding corresponding buffer queue of core 2 is queue
3 and queue 4, if IPv4, IPv6, POS message that forwarding core 2 is obtained from queue 3 are calculated as Q3 respectively in preset durationipv4、
Q3ipv6、Q3pos, similarly, IPv4, IPv6, POS message that forwarding core 2 is obtained from queue 4 are calculated as Q4 respectivelyipv4、Q4ipv6、
Q4pos, then forwarding the consumption value of core 2 within a preset time is (Q3ipv4+Q4ipv4) * Δ tipv4+(Q3ipv6+Q4ipv6) * Δ tipv6
+(Q3pos+Q4pos) * Δ tpos.And so on, available all forwarding cores processor consumption value E1 within a preset time,
E2…En.Wherein, if message accumulation occurs in queue 1, i.e. queue 1 is the first buffer queue, is turned since queue 1 is assigned to
It sends out core 1 to handle, then forwarding core 1 is the first forwarding core, then the processing consumption value for the first forwarding core that the first buffer queue occupies
For Q1ipv4* Δ tipv4+Q1ipv6* Δ tipv6+Q1pos* Δ tpos。
Wherein, how control core obtains each forwarding core in multiple forwarding cores in preset duration from corresponding buffer queue
The message data of middle acquisition? as an alternative embodiment, can be by changing the state of global variable switch come real
Existing, optionally, the state that control nuclear periodicity changes global variable switch is at first state or the second state;Wherein,
In the first state, each forwarding core in the multiple forwarding core is respectively counted in the preset duration from corresponding slow
Deposit the message data obtained in queue;In second state, the control core obtains each of the multiple forwarding core
Forward the message data of core feedback.For example, the state that control nuclear periodicity changes global variable switch is at and plays on or off
Closed state, forwarding core decide whether to count data according to the state that global variable switchs, for example, switching in global variable
In the open state, each forwarding core respectively counts the message data obtained from corresponding buffer queue in preset duration;
When global variable switch is in close state, each forwarding core stops statistics, and the message data respectively counted is fed back to
Control core.Wherein, control core can terminate front opening global variable switch in upper primary polling tasks, open in epicycle polling tasks
Global variable switch is closed after beginning.In addition, above-mentioned statistics is also possible to the statistics when global variable is switched in closing, complete
Office's variable, which is in, stops statistics when opening.
Wherein, consumption value of each forwarding cores in preset duration that control core obtains in multiple forwarding cores, which can be, to be controlled
All buffer queues of nuclear periodicity poll processed, before the step of having checked whether message accumulation, namely control core is first periodically
The multiple forwarding cores of acquisition in consumption value of each forwarding cores in preset duration, it is to be obtained to each forwarding core when default
After consumption value in length, check whether there are message accumulate queue then weighed if not provided, just waiting until next cycle
After newly obtaining consumption value of each forwarding core in multiple forwarding cores in preset duration, then check whether that there are message accumulations
Queue;If so, then needing the whole processor consumption according to the occupied processor resource of accumulation queue and each forwarding core
It decides whether to carry out pyrene scheduling to accumulation queue.Compared to first checking whether with the presence of queue message accumulation, in determination
There are message accumulation queue when, then obtain it is each forwarding core consumption value scheme for, can reduce actual effect to second
The influence for forwarding core avoids causing second finally determined to forward core different because of timeliness difference.
Wherein, consumption value of each forwarding cores in preset duration that control core obtains in multiple forwarding cores, which can be, to be controlled
All buffer queues of nuclear periodicity poll processed after the step of having checked whether message accumulation, namely are determining multiple cachings
There are when the first buffer queue of message accumulation in queue, control core obtains each forwarding core in multiple forwarding cores pre- again
If the consumption value in duration.If there is not the queue of message accumulation, each forwarding core not obtained in multiple forwarding cores exists
Consumption value in preset duration.Compared to first obtain it is each forwarding core consumption value, then check whether there are message accumulate team
For the scheme of column, it is possible to reduce workload avoids after having got the consumption value of each forwarding core, but without queue accumulation
It happens.
Wherein, it should be noted that preset time is setting in advance, can be arranged according to different forwarding tasks different
Time, can be with flexible setting.
Step S102: the control core deletes the configuration relation of first buffer queue and the first forwarding core, builds
The configuration relation for founding first buffer queue and the second forwarding core makes the second forwarding core processing first caching
Message in queue.
Control core is determining processing energy from each forwarding core in the multiple forwarding core in addition to the first forwarding core
After second forwarding core of power, control core deletes the configuration relation of the first buffer queue and the first forwarding core, establishes the first caching team
The configuration relation of column and the second forwarding core, makes the second forwarding core handle the message in the first buffer queue.This queue dynamic point
Core scheduling mechanism not only can quickly identify network burst flow and make corresponding processing, can more reduce at single forwarding core
The risk of reason ability overload, avoids unnecessary data flow packet loss problem.
Wherein, it should be noted that " queue " and " buffer queue " in the application can be exchanged mutually.
Second embodiment
The embodiment of the present application provides a kind of network equipment, including multi-core processor, and the multi-core processor includes control
Core and multiple forwarding cores, the network equipments configuration have multiple buffer queues, each forwarding verification in the multiple forwarding core
There should be at least one buffer queue, and a buffer queue only corresponds to a forwarding core, each of the multiple forwarding core turns
Core is sent out, for handling the message at least one corresponding buffer queue.Wherein, the framework of the multi-core processor can be refering to figure
1.It should be noted that control core can be and be randomly assigned from multiple kernels, for example, specify some kernels for control core,
Remaining all kernels are forwarding core.
Wherein, the control core, for determining the first caching team in the multiple buffer queue there are message accumulation
When column, determined from each forwarding core in the multiple forwarding core in addition to the corresponding first forwarding core of first buffer queue
There is the second forwarding core of processing capacity out;The control core is also used to delete first buffer queue and first forwarding
The configuration relation of core establishes the configuration relation of first buffer queue and the second forwarding core, makes the second forwarding core
Handle the message in first buffer queue.
Optionally, the control core, be also used to by consumption value of the first forwarding core in preset duration respectively with institute
Consumption value of each forwarding core in multiple forwarding cores in addition to the first forwarding core in the preset duration is stated to be subtracted each other,
It obtains multiple subtracting each other result;The control core is also used to choose maximum subtract each other as a result, conduct in result from the multiple subtract each other
Target subtracts each other result;The control core is also used to determine that the target subtracts each other what result was occupied greater than first buffer queue
The processing consumption value of the first forwarding core;The control core is also used to subtract each other the target the corresponding forwarding core of result, makees
For the second forwarding core.
Optionally, the control core is also used to obtain each forwarding core in the multiple forwarding core when described default
Consumption value in length;The control core is also used to determine the multiple slow when getting the consumption value of each forwarding core
Deposit in queue there are message accumulation the first buffer queue.
Optionally, the control core is also used to obtain each forwarding core in the multiple forwarding core when described default
Consumption value in length;The control core is also used to the consumption value by the first forwarding core respectively and in the multiple forwarding core
The consumption value of each forwarding core in addition to the first forwarding core is subtracted each other, and obtains multiple subtracting each other result.
Optionally, the control core, be also used to obtain it is the multiple forwarding core in each forwarding core respectively count
The message data obtained from corresponding buffer queue in the preset duration;The control core is also used to based on the message number
According to the consumption value of each forwarding core of determination.
Optionally, the control core, the state for being also used to periodically-varied global variable switch are at first state
Or second state, wherein in the first state, each forwarding core in the multiple forwarding core is respectively counted described pre-
If the message data obtained from corresponding buffer queue in duration;In second state, the control core is also used to obtain
The message data of each forwarding core feedback in the multiple forwarding core.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
3rd embodiment
The embodiment of the present application also provides a kind of queue scheduling dresses applied in the network equipment including multi-core processor
100 are set, as shown in Figure 4.Wherein, the multi-core processor further includes multiple forwarding cores, and the network equipments configuration has multiple slow
Queue is deposited, each forwarding verification in the multiple forwarding core there should be at least one buffer queue, and a buffer queue is only right
Answer a forwarding core.The queue scheduling device 100 comprises determining that module 110 and configuration module 120.
Determining module 110, for determine in the multiple buffer queue there are message accumulation the first buffer queue when,
It has been determined from each forwarding core in the multiple forwarding core in addition to the corresponding first forwarding core of first buffer queue
Second forwarding core of processing capacity.Optionally, the determining module 110 is also used to the control core for the first forwarding core
Consumption value in preset duration is respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core in institute
The consumption value stated in preset duration is subtracted each other, and obtains multiple subtracting each other result;The control core subtracts each other in result from the multiple
Choose maximum subtract each other as a result, subtracting each other result as target;The control core determines that the target subtracts each other result and is greater than described the
The processing consumption value for the first forwarding core that one buffer queue occupies;It is corresponding that the target is subtracted each other result by the control core
Core is forwarded, as the second forwarding core.Optionally, it is the multiple to be also used to the control core acquisition for the determining module 110
Forward consumption value of each forwarding core in core in the preset duration;The control core is corresponding by first buffer queue
The consumption value of the first forwarding core disappear respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core
Consumption value is subtracted each other, and obtains multiple subtracting each other result.Optionally, the determining module 110 is also used to described in the control core acquisition
The message obtained from corresponding buffer queue in the preset duration that each forwarding core in multiple forwarding cores respectively counts
Data;The control core determines the consumption value of each forwarding core based on the message data.Optionally, the determining module
110, it is also used to the state that the control nuclear periodicity changes global variable switch and is at first state or the second state,
In, in the first state, each forwarding core in the multiple forwarding core is respectively counted in the preset duration from right
Answer the message data obtained in buffer queue;In second state, the control core is obtained in the multiple forwarding core
The message data of each forwarding core feedback.
Configuration module 120 establishes institute for deleting the configuration relation of first buffer queue and the first forwarding core
The configuration relation for stating the first buffer queue and the second forwarding core makes the second forwarding core handle first buffer queue
In message.
Optionally, queue scheduling device 100 further include: obtain module and the second determining module.
Module is obtained, obtains each forwarding core in the multiple forwarding core in the preset duration for the control core
Interior consumption value.Optionally, the acquisition module, being also used to the state that the control nuclear periodicity changes global variable switch makes
It is in first state or the second state, wherein in the first state, each forwarding core in the multiple forwarding core is each
The message data obtained from corresponding buffer queue in the preset duration from statistics;In second state, the control
Core processed obtains the message data of each forwarding core feedback in the multiple forwarding core.Optionally, the acquisition module, is also used to
The state that the control nuclear periodicity changes global variable switch is at first state or the second state, wherein described
When first state, each forwarding core in the multiple forwarding core is respectively counted in the preset duration from corresponding buffer queue
The message data of middle acquisition;In second state, the control core obtains each forwarding core in the multiple forwarding core
The message data of feedback.
Second determining module, for the control core when getting the consumption value of each forwarding core, described in determination
There are the first buffer queues of message accumulation in multiple buffer queues.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
The technical effect of queue scheduling device 100 provided by the embodiment of the present invention, realization principle and generation and aforementioned
Embodiment of the method is identical, and to briefly describe, Installation practice part does not refer to place, can refer to corresponding in preceding method embodiment
Content.
Fourth embodiment
The embodiment of the present application also provides a kind of computer readable storage medium, deposited on the computer readable storage medium
Computer program is contained, the computer program executes the step of method described in above-mentioned second embodiment when being run by processor
Suddenly.Specific implementation can be found in embodiment of the method, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and block diagram in attached drawing
Show the device of multiple embodiments according to the present invention, the architectural framework in the cards of method and computer program product,
Function and operation.In this regard, each box in flowchart or block diagram can represent the one of a module, section or code
Part, a part of the module, section or code, which includes that one or more is for implementing the specified logical function, to be held
Row instruction.It should also be noted that function marked in the box can also be to be different from some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that every in block diagram and or flow chart
The combination of box in a box and block diagram and or flow chart can use the dedicated base for executing defined function or movement
It realizes, or can realize using a combination of dedicated hardware and computer instructions in the system of hardware.
In addition, each functional module in each embodiment of the present invention can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, laptop, server or network equipment etc.) execute the whole of each embodiment the method for the present invention
Or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.It should be noted that, in this document, relational terms such as first and second and the like are used merely to one
A entity or operation with another entity or operate distinguish, without necessarily requiring or implying these entities or operation it
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant are intended to
Cover non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or setting
Standby intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in the process, method, article or apparatus that includes the element.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of array dispatching method, which is characterized in that it is applied to the network equipment, the network equipment includes multi-core processor,
The multi-core processor includes control core and multiple forwarding cores, and the network equipments configuration has multiple buffer queues, the multiple
Each forwarding verification in forwarding core should have at least one buffer queue, and a buffer queue only corresponds to a forwarding core;Institute
The method of stating includes:
Determine that the control core is from the multiple there are when the first buffer queue of message accumulation in the multiple buffer queue
Processing capacity is determined in each forwarding core in forwarding core in addition to the corresponding first forwarding core of first buffer queue
Second forwarding core;
The control core deletes the configuration relation of first buffer queue and the first forwarding core, establishes first caching
The configuration relation of queue and the second forwarding core makes the second forwarding core handle the message in first buffer queue.
2. the method according to claim 1, wherein the control core is from the multiple forwarding core except described the
The second forwarding core of processing capacity is determined in each forwarding core outside the corresponding first forwarding core of one buffer queue, comprising:
The control core by consumption value of the first forwarding core in preset duration respectively in the multiple forwarding core except institute
Consumption value of each forwarding core in the preset duration stated outside the first forwarding core is subtracted each other, and obtains multiple subtracting each other result;
The control core chooses maximum subtract each other as a result, subtracting each other result as target from the multiple subtract each other in result;
The control core determines that the target subtracts each other result greater than the first forwarding core that first buffer queue occupies
Handle consumption value;
The target is subtracted each other the corresponding forwarding core of result by the control core, as the second forwarding core.
3. according to the method described in claim 2, it is characterized in that, forwarding core when default for described first in the control core
Consumption value in length is respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core when described default
Consumption value in length is subtracted each other, before obtaining multiple the step of subtracting each other result, the method also includes:
The control core obtains consumption value of each forwarding core in the preset duration in the multiple forwarding core;
The control core determines that there are reports in the multiple buffer queue when getting the consumption value of each forwarding core
First buffer queue of text accumulation.
4. according to the method described in claim 2, it is characterized in that, the control core is by the first forwarding core in preset duration
Interior consumption value is respectively with each forwarding core in the multiple forwarding core in addition to the first forwarding core in the preset duration
Interior consumption value is subtracted each other, and obtains multiple subtracting each other result, comprising:
The control core obtains consumption value of each forwarding core in the preset duration in the multiple forwarding core;
It is described control core by it is described first forwarding core consumption value respectively in the multiple forwarding core except it is described first forward core
The consumption value of outer each forwarding core is subtracted each other, and obtains multiple subtracting each other result.
5. the method according to claim 3 or 4, which is characterized in that the control core obtains in the multiple forwarding core
Consumption value of each forwarding core in the preset duration, comprising:
It is described control core obtain it is the multiple forwarding core in each forwarding core respectively count in the preset duration from right
Answer the message data obtained in buffer queue;
The control core determines the consumption value of each forwarding core based on the message data.
6. according to the method described in claim 5, it is characterized in that, the control core obtains each of the multiple forwarding core
The message data obtained from corresponding buffer queue in the preset duration that forwarding core respectively counts, comprising:
The state that the control nuclear periodicity changes global variable switch is at first state or the second state, wherein
When the first state, each forwarding core in the multiple forwarding core, which is respectively counted, to be cached in the preset duration from corresponding
The message data obtained in queue;In second state, the control core obtains each of the multiple forwarding core and turns
Send out the message data of core feedback.
7. a kind of network equipment, which is characterized in that including multi-core processor, the multi-core processor includes control core and multiple turns
Core is sent out, the network equipments configuration there are multiple buffer queues, and each forwarding verification in the multiple forwarding core should have at least one
A buffer queue, and a buffer queue only corresponds to a forwarding core;
The control core, for determine in the multiple buffer queue there are message accumulation the first buffer queue when, from institute
It states in each forwarding core in multiple forwarding cores in addition to the corresponding first forwarding core of first buffer queue and has determined processing
Second forwarding core of ability;
The control core is also used to delete the configuration relation of first buffer queue and the first forwarding core, described in foundation
The configuration relation of first buffer queue and the second forwarding core handles the second forwarding core in first buffer queue
Message;
Each forwarding core in the multiple forwarding core, for handling the message at least one corresponding buffer queue.
8. the network equipment according to claim 7, it is characterised in that:
The control core, be also used to by consumption value of the first forwarding core in preset duration respectively with the multiple forwarding core
In consumption value of each forwarding core in the preset duration in addition to the first forwarding core subtracted each other, obtain multiple subtract each other
As a result;
The control core is also used to choose maximum subtract each other as a result, subtracting each other result as target in result from the multiple subtract each other;
The control core is also used to determine that the target subtracts each other result and is greater than described first turn that first buffer queue occupies
Send out the processing consumption value of core;
The control core is also used to subtracting each other the target into the corresponding forwarding core of result, as the second forwarding core.
9. a kind of queue scheduling device, which is characterized in that applied to the network equipment including multi-core processor, the multicore processing
Device further includes multiple forwarding cores, and the network equipments configuration has multiple buffer queues, each forwarding in the multiple forwarding core
Verification should have at least one buffer queue, and a buffer queue only corresponds to a forwarding core, and described device includes:
Determining module, for determining in the multiple buffer queue there are when the first buffer queue of message accumulation, from described
Processing energy has been determined in each forwarding core in multiple forwarding cores in addition to the corresponding first forwarding core of first buffer queue
Second forwarding core of power;
Configuration module establishes described first for deleting the configuration relation of first buffer queue and the first forwarding core
The configuration relation of buffer queue and the second forwarding core makes the second forwarding core handle the report in first buffer queue
Text.
10. a kind of storage medium, which is characterized in that be stored thereon with computer program, the computer program is by processor
When operation, such as method of any of claims 1-6 is executed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811388030.9A CN109450816B (en) | 2018-11-19 | 2018-11-19 | Queue scheduling method, device, network equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811388030.9A CN109450816B (en) | 2018-11-19 | 2018-11-19 | Queue scheduling method, device, network equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109450816A true CN109450816A (en) | 2019-03-08 |
CN109450816B CN109450816B (en) | 2022-08-12 |
Family
ID=65552804
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811388030.9A Active CN109450816B (en) | 2018-11-19 | 2018-11-19 | Queue scheduling method, device, network equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109450816B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112068965A (en) * | 2020-09-23 | 2020-12-11 | Oppo广东移动通信有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN112073332A (en) * | 2020-08-10 | 2020-12-11 | 烽火通信科技股份有限公司 | Message distribution method, multi-core processor and readable storage medium |
CN113176940A (en) * | 2021-03-29 | 2021-07-27 | 新华三信息安全技术有限公司 | Data flow splitting method and device and network equipment |
CN113328960A (en) * | 2020-02-28 | 2021-08-31 | 华为技术有限公司 | Queue cache management method and device, storage medium and equipment |
CN113992589A (en) * | 2021-10-21 | 2022-01-28 | 绿盟科技集团股份有限公司 | Message distribution method and device and electronic equipment |
CN114024915A (en) * | 2021-10-28 | 2022-02-08 | 北京锐安科技有限公司 | Traffic migration method, device and system, electronic equipment and storage medium |
WO2023093805A1 (en) * | 2021-11-26 | 2023-06-01 | 中兴通讯股份有限公司 | Storage control method, storage controller, storage chip, network card, and readable medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101616083A (en) * | 2009-08-06 | 2009-12-30 | 杭州华三通信技术有限公司 | A kind of message forwarding method and device |
CN103338157A (en) * | 2013-07-01 | 2013-10-02 | 杭州华三通信技术有限公司 | Internuclear data message caching method and equipment of multinuclear system |
CN105159779A (en) * | 2015-08-17 | 2015-12-16 | 深圳中兴网信科技有限公司 | Method and system for improving data processing performance of multi-core CPU |
CN105634958A (en) * | 2015-12-24 | 2016-06-01 | 东软集团股份有限公司 | Packet forwarding method and device based on multi-core system |
CN106713185A (en) * | 2016-12-06 | 2017-05-24 | 瑞斯康达科技发展股份有限公司 | Load balancing method and apparatus of multi-core CPU |
CN108259369A (en) * | 2018-01-26 | 2018-07-06 | 迈普通信技术股份有限公司 | The retransmission method and device of a kind of data message |
CN108694087A (en) * | 2017-03-31 | 2018-10-23 | 英特尔公司 | For the dynamic load leveling in the network interface card of optimal system grade performance |
CN108777662A (en) * | 2018-06-20 | 2018-11-09 | 迈普通信技术股份有限公司 | Entry management method and device |
US11012507B2 (en) * | 2016-08-29 | 2021-05-18 | Vmware, Inc. | High throughput layer 2 extension leveraging CPU flow affinity |
-
2018
- 2018-11-19 CN CN201811388030.9A patent/CN109450816B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101616083A (en) * | 2009-08-06 | 2009-12-30 | 杭州华三通信技术有限公司 | A kind of message forwarding method and device |
CN103338157A (en) * | 2013-07-01 | 2013-10-02 | 杭州华三通信技术有限公司 | Internuclear data message caching method and equipment of multinuclear system |
CN105159779A (en) * | 2015-08-17 | 2015-12-16 | 深圳中兴网信科技有限公司 | Method and system for improving data processing performance of multi-core CPU |
CN105634958A (en) * | 2015-12-24 | 2016-06-01 | 东软集团股份有限公司 | Packet forwarding method and device based on multi-core system |
US11012507B2 (en) * | 2016-08-29 | 2021-05-18 | Vmware, Inc. | High throughput layer 2 extension leveraging CPU flow affinity |
CN106713185A (en) * | 2016-12-06 | 2017-05-24 | 瑞斯康达科技发展股份有限公司 | Load balancing method and apparatus of multi-core CPU |
CN108694087A (en) * | 2017-03-31 | 2018-10-23 | 英特尔公司 | For the dynamic load leveling in the network interface card of optimal system grade performance |
CN108259369A (en) * | 2018-01-26 | 2018-07-06 | 迈普通信技术股份有限公司 | The retransmission method and device of a kind of data message |
CN108777662A (en) * | 2018-06-20 | 2018-11-09 | 迈普通信技术股份有限公司 | Entry management method and device |
Non-Patent Citations (1)
Title |
---|
黄亮: "多核处理器中任务调度与负载均衡的研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113328960A (en) * | 2020-02-28 | 2021-08-31 | 华为技术有限公司 | Queue cache management method and device, storage medium and equipment |
CN113328960B (en) * | 2020-02-28 | 2023-11-17 | 华为技术有限公司 | Queue cache management method, device, storage medium and equipment |
CN112073332A (en) * | 2020-08-10 | 2020-12-11 | 烽火通信科技股份有限公司 | Message distribution method, multi-core processor and readable storage medium |
CN112068965A (en) * | 2020-09-23 | 2020-12-11 | Oppo广东移动通信有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN113176940A (en) * | 2021-03-29 | 2021-07-27 | 新华三信息安全技术有限公司 | Data flow splitting method and device and network equipment |
CN113992589A (en) * | 2021-10-21 | 2022-01-28 | 绿盟科技集团股份有限公司 | Message distribution method and device and electronic equipment |
CN113992589B (en) * | 2021-10-21 | 2023-05-26 | 绿盟科技集团股份有限公司 | Message distribution method and device and electronic equipment |
CN114024915A (en) * | 2021-10-28 | 2022-02-08 | 北京锐安科技有限公司 | Traffic migration method, device and system, electronic equipment and storage medium |
WO2023071272A1 (en) * | 2021-10-28 | 2023-05-04 | 北京锐安科技有限公司 | Traffic migration method, apparatus and system, and electronic device and storage medium |
WO2023093805A1 (en) * | 2021-11-26 | 2023-06-01 | 中兴通讯股份有限公司 | Storage control method, storage controller, storage chip, network card, and readable medium |
Also Published As
Publication number | Publication date |
---|---|
CN109450816B (en) | 2022-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109450816A (en) | A kind of array dispatching method, device, the network equipment and storage medium | |
CN107196869B (en) | The adaptive load balancing method, apparatus and system of Intrusion Detection based on host actual loading | |
CN104539440B (en) | Traffic management with in-let dimple | |
CN102918499B (en) | For application strategy with the method and apparatus of dispatch network bandwidth between virtual machine | |
CN110022269B (en) | Communication data processing method, device and equipment | |
Noormohammadpour et al. | {DCCast}: Efficient point to multipoint transfers across datacenters | |
CN105323099B (en) | Business network flowmeter factor method, network resource scheduling method and network element | |
CN106471773A (en) | The distribution of integrated form global resource and load balance | |
CN110209492A (en) | A kind of data processing method and device | |
Lin et al. | DTE-SDN: A dynamic traffic engineering engine for delay-sensitive transfer | |
Nazar et al. | Modified shortest job first for load balancing in cloud-fog computing | |
CN109831524A (en) | A kind of load balance process method and device | |
US9516169B2 (en) | Automatic contact center expansion and contraction | |
Simoncelli et al. | Stream-monitoring with blockmon: convergence of network measurements and data analytics platforms | |
Alwasel et al. | Iotsim-sdwan: A simulation framework for interconnecting distributed datacenters over software-defined wide area network (sd-wan) | |
CN108376103A (en) | A kind of the equilibrium of stock control method and server of cloud platform | |
CN107623632A (en) | The connection method of procotol based on DPDK and system | |
Liang et al. | Low-latency service function chain migration in edge-core networks based on open Jackson networks | |
US20100198971A1 (en) | Dynamically provisioning clusters of middleware appliances | |
CN104780118A (en) | Fluid control method and device based on tokens | |
Mori et al. | Queueing theoretic approach to job assignment strategy considering various inter-arrival of job in fog computing | |
WO2017198007A1 (en) | Method and apparatus for determining routing policy | |
Suen et al. | Affordable avionic networks with Gigabit Ethernet assessing the suitability of commercial components for airborne use | |
Tan et al. | Path‐flow matching: Two‐sided matching and multiobjective evolutionary algorithm for traffic scheduling in cloud data* center network | |
Jung et al. | Distributed multipath routing algorithm for data center networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |