CN107483355B - Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme - Google Patents

Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme Download PDF

Info

Publication number
CN107483355B
CN107483355B CN201710805436.1A CN201710805436A CN107483355B CN 107483355 B CN107483355 B CN 107483355B CN 201710805436 A CN201710805436 A CN 201710805436A CN 107483355 B CN107483355 B CN 107483355B
Authority
CN
China
Prior art keywords
bandwidth
data
edge
data center
node
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
Application number
CN201710805436.1A
Other languages
Chinese (zh)
Other versions
CN107483355A (en
Inventor
崔勇
李楚鸣
杨振杰
肖诗汉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Original Assignee
Tsinghua University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tsinghua University filed Critical Tsinghua University
Priority to CN201710805436.1A priority Critical patent/CN107483355B/en
Publication of CN107483355A publication Critical patent/CN107483355A/en
Application granted granted Critical
Publication of CN107483355B publication Critical patent/CN107483355B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • 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

Abstract

A data center-oriented online scene low-bandwidth overhead traffic scheduling scheme is an Internet transmission control technology. The method aims to minimize bandwidth renting cost of a data center owner by reasonably scheduling each large transport stream and ensure that all the large transport streams are finished on time. The execution flow is 1) an operation controller obtains source and destination nodes, data volume, arrival and deadline time of all large transport streams in a lease period in real time; 2) when each large transport stream arrives, the minimization of the bandwidth expenditure is achieved on the premise of ensuring that all large transport streams can be completed by taking the minimization of the extra bandwidth leasing expenditure as a target. The invention can effectively reduce the expense of renting bandwidth from the internet service provider by the data center owner and reduce the operation cost of the data center on the premise of ensuring the service quality.

Description

Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme
Technical Field
The invention belongs to the technical field of internet, relates to a traffic scheduling technology, and particularly relates to a data center-oriented low-bandwidth overhead traffic scheduling scheme in an online scene.
Background
Many network service providers and cloud services maintain multiple data centers to support their businesses, such as microsoft, google. The data centers run various globally distributed applications and are distributed in different geographical areas, which determines that the data centers have the requirement of mutual communication across the geographical areas, and the wide area network plays a key role in ensuring that the data centers can mutually communicate between different geographical positions. The large amount of data transport streams between data centers results in high bandwidth overhead, and data center owners rent wide area network bandwidth to internet service providers every year, at a cost of hundreds of millions. More critically, unreasonable traffic scheduling results in low bandwidth utilization across data centers, with the majority of links having bandwidth utilization of no more than 60%, which means a large percentage of waste in the high bandwidth overhead. How to reasonably and effectively carry out flow scheduling, reduce bandwidth overhead and ensure that data flow is completed on time becomes an important problem in the field of flow scheduling among data centers.
A bulk transport stream is defined as a type of stream that is significant, large in data volume, and long in duration in inter-data center wide area network traffic. Typically, a large block transport stream accounts for 85% to 95% of the inter-data center traffic, with data volumes of several TBs to several PB, lasting for several hours. Two typical examples of this are: the financial institution backs up transaction records at the remote end of the transaction day, and the search engine periodically synchronizes the index items among the data centers, and the like. Another type of stream between data centers is interactive streamlets, which are short in duration and highly sensitive to latency. Compared with a large transmission stream, the requirement on the time delay is not high, and the time delay caused by scheduling by using the centralized controller can be tolerated. In conclusion, it is of great significance to reasonably schedule a large transport stream. In some scenarios, the parameters of all chunk transport streams are unpredictable over a period of time, the arrival time, the deadline, and the amount of data of a chunk transport stream are known only after it is generated, and these scenarios are collectively referred to as online scenarios. The reasonable scheduling of the large data flow in the online scene is not only an important guarantee of the network service quality, but also an effective way of saving a large amount of bandwidth renting expenses.
Much research work has emerged in recent years around the rational scheduling of large transport streams. One of the main ideas is to add a storage device to a data center, and to select whether to store or forward data when the data arrives, i.e. a store-and-forward strategy. The first work proposes that the arriving data is temporarily stored when the link is busy, and the data is transmitted when the link is idle, so that the utilization rate of the bandwidth is finally improved in the time dimension. Another work balances bandwidth utilization on each link through store-and-forward strategies, thereby achieving load balancing. Because the traffic passing through each data center needs to be temporarily stored, the device deployment under the idea needs to add an additional storage device to each data center, which not only additionally increases the storage overhead, but also makes traffic scheduling more complex. Therefore, it is desirable to find a more reasonable scheduling scheme to optimize the bandwidth overhead, and simultaneously ensure that each large transport stream is completed on time without increasing the additional storage overhead.
Disclosure of Invention
In order to overcome the disadvantages of the prior art, the present invention aims to provide a data center-oriented online scene low-bandwidth overhead traffic scheduling scheme, which minimizes the extra bandwidth lease overhead caused by each large transport stream through reasonable scheduling on the premise of ensuring that all large transport streams can be completed on time, thereby minimizing the total bandwidth overhead; the invention reasonably distributes the bandwidth for each large transmission stream in each transmission time slot, sets the transmission path, and minimizes the bandwidth renting cost on the basis of ensuring that all the large transmission streams are completed on time.
In order to achieve the purpose, the invention adopts the technical scheme that:
a data center-oriented online scene low-bandwidth overhead traffic scheduling scheme is realized under a wide area network among data centers according to the following steps:
step (1), dividing a lease period into a plurality of transmission time slots, namely 1, …, T, representing links between the data centers by a directed graph G ═ V, E, wherein V is a node set of the directed graph and represents a set of all the data centers, E is an edge set of the directed graph and represents a set of all the links, and a quintuple r is used for representing the links between the data centersi＝(si,ti,di,ai,τi) To represent a large transport stream, where si，ti，di，ai，τiRespectively representing a source node, a destination node, data volume, arrival time and deadline of the ith chunk transport stream; the operation integrated controller dynamically obtains a source node, a destination node, data volume, arrival time and deadline of each large transport stream;
step (2), running OPPG algorithm, calculating its scheduling scheme when each large block transport stream arrives, and making all bandwidth values c when algorithm is initializede＝0；
And (3) when each large transport stream arrives, calculating the minimum extra bandwidth overhead through an OPPG algorithm, and specifically comprising the following steps:
step (3a) of traversing all feasible paths of the stream by a depth-first search algorithm to find the duration [ a ] of the streami,ti]Inner bandwidth value cePaths that are not fully utilized, and the paths with residual bandwidth are used for transmitting the data of the stream; if r isiIf the transmission is finished, the algorithm is terminated, and the step (3e) is carried out, otherwise, the step (3b) is carried out;
step (3b), first order all Ve＝0，VeThe rental value for link e;
and (3c) traversing all feasible paths of the flow through a depth-first search algorithm, wherein all feasible paths have a congestion edge at any time because all residual bandwidth is utilized in the step 2, namely the bandwidth value used on the congestion edge at the time is ce(ii) a In the traversing process, the congestion edges of all the paths at all the moments are found out, and for a congestion edge e of a certain path at a certain moment, the bandwidth value c of the edge is calculatedeAfter increasing one unit, the amount d of data that can be additionally transmitted on the path at that moment is as small as possible
Figure BDA0001402581110000031
Wherein u iseA price per bandwidth that represents edge e;
step (3d), after which V is selectedeMaximum edge e, order ce＝ceStep (3a) of going to step (3a) in which new surplus bandwidth is introduced, and surplus traffic is transmitted using the surplus bandwidth;
step (3e), the algorithm is terminated, and the scheduling scheme of the flow is obtained;
and (4) generating a flow scheduling scheme according to the algorithm result, and performing flow scheduling.
P for the present invention0Expressing the objective function under three constraints of flow constraint, capacity constraint and integer constraint:
Figure BDA0001402581110000032
a minimized optimization problem;
wherein, there are two flow constraints, the first flow constraint is:
Figure BDA0001402581110000041
and v ≠ si,v≠ti,
Figure BDA0001402581110000042
t∈[ai,τi]And t ∈ N+
δ+(v) Representing the set of all directed edges, δ, starting from node v-(v) Representing the set of all directed edges, x, ending with node vi,e(t) represents the amount of data transmitted at the t-th instant on connection e for the ith request, N+Represents a set of positive integers;
another flow constraint is:
Figure BDA0001402581110000043
wherein, delta+(si) Represented by node siSet of all directed edges as starting points, δ-(si) Represented by node siA set of all directed edges that are end points;
the capacity constraint is:
Figure BDA0001402581110000044
wherein, ceFor the value of bandwidth leased on edge e, represents the number of units of bandwidth leased on edge e by the data center owner, δcRepresenting the size of a unit bandwidth, δtRepresenting the size of each time slice;
the integer constraint is:
Figure BDA0001402581110000045
where N represents a set of natural numbers.
Compared with the prior art, the invention has the beneficial effects that:
1) the overhead of bandwidth leasing is minimized on the premise of ensuring that all large transport streams can be transmitted within a specified time.
2) The scheme provided by the invention considers that the ISP charges according to a certain granularity, and the practicability is stronger.
3) The scheme provided by the invention does not need to introduce additional storage equipment, so that the total scheduling overhead is saved.
Drawings
Fig. 1 is a schematic diagram of an online scene oriented to a data center.
Fig. 2 is a specific flowchart of a data center-oriented online scenario low-bandwidth overhead traffic scheduling scheme. Wherein c iseFor the value of the bandwidth leased on edge e, VeThe rental value for edge e.
Detailed Description
The embodiments of the present invention will be described in detail below with reference to the drawings and examples.
As shown in fig. 1, the present invention considers the scheduling problem of a large transport stream within a lease period, and divides the period into several transmission slots, i.e., 1, …, T. The data centers and the links between the data centers are represented by a directed graph G ═ V, E, where V is the set of nodes of the directed graph and represents the set of all the data centers, and E is the set of edges of the directed graph and represents the set of all the links. By five members ri＝(si,ti,di,ai,τi) To represent a large transport stream, where si，ti，di，ai，τiRespectively representing the source node, destination node, data volume, arrival time, and deadline of the ith chunk transport stream.
For a large transport stream riThe time for transmitting data is limited to the time interval [ a ]i,τi]Within the time slice. In addition, a source destination node s of a requestiAnd tiCan store inIn a plurality of possible paths, each path being formed by one or more edges E in E in series, using x according to the above descriptioni,e(t) to represent the amount of data transmitted at the t-th instant on edge e for the ith request, a traffic constraint can be derived:
Figure BDA0001402581110000051
and v ≠ si,v≠ti,
Figure BDA0001402581110000052
t∈[ai,τi]And t ∈ N+
The implication of this constraint is that at all times any one bulk transport stream must satisfy traffic conservation at nodes other than all its source-destination nodes, i.e., the sum of the flows belonging to the bulk transport stream that flow out of the node must equal the sum of the flows belonging to the bulk transport stream that flow into the node. Wherein, delta+(v) Representing the set of all directed edges, δ, starting from node v-(v) Representing the set of all directed edges that end at node v.
Another flow constraint is:
Figure BDA0001402581110000061
the constraint ensures that the sum of the flows belonging to any one large transport stream from the source node minus the sum of the flows belonging to the large transport stream into the source node is equal to the total data transmission volume of the large transport stream at all times, and the function of the constraint is to ensure that all large transport streams can be completed within a specified time.
To ensure that the total rate of transmission traffic of any link in any transmission time slot does not exceed the leased link bandwidth size, xi,e(t) the capacity constraint must be met:
Figure BDA0001402581110000062
wherein, c iseNumber of units, δ, representing the bandwidth leased on edge e by the data center ownercRepresenting the size of a unit bandwidth, δtThe size of each time slice is represented.
Since the data center owner must lease an integer unit of bandwidth when leasing bandwidth, ceIs an integer variable, therefore ceInteger constraints need to be satisfied:
Figure BDA0001402581110000063
in order to achieve the goal of minimizing bandwidth lease overhead of the present invention, the scheme uses P0Expressing the objective function under three constraints of flow constraint, capacity constraint and integer constraint:
Figure BDA0001402581110000064
a minimized optimization problem. Wherein u iseIndicating the price per bandwidth of this edge.
Because of P0Is generated time-divisionally rather than predicted in advance, P0Rather than a standard integer linear programming problem, the present solution includes an online rental scheme generation algorithm (OPPG) that solves the problem. The idea of the algorithm is that when each chunk transport stream arrives, the added overhead for scheduling the newly arrived chunk transport stream is minimized, and the flow of the algorithm is as follows:
1. let all bandwidth values ceEqual to zero.
2. Whenever a new chunk transport stream r is transmittediUpon arrival, all feasible paths of the stream are traversed by a depth-first search algorithm to find the duration [ a ] of the streami,ti]Inner bandwidth value cePaths that are not fully utilized use these paths with the remaining bandwidth to transmit the data of the stream. If r isiThe transmission is complete and the algorithm ends.
3. If the data of the stream still hasAnd if the link is not available, the new bandwidth needs to be leased additionally according to the value of the link. Definition VeFor the rental value of e this edge, VeThe calculation method of (c) is as follows:
first order all VeEqual to zero. Thereafter, all feasible paths of the flow are traversed through the depth-first search algorithm, and since all remaining bandwidth has been utilized in step 2, all feasible paths must have a congestion edge at any time, i.e., the bandwidth value used on the congestion edge at that time is ce. In the traversing process, the congestion edges of all the paths at all the moments are found out, and for a congestion edge e of a certain path at a certain moment, the bandwidth value c of the edge is calculatedeAfter increasing one unit, the amount d of data that can be additionally transmitted on the path at that moment is as small as possible
Figure BDA0001402581110000071
Then, V is selectedeMaximum edge, order ce←ce+1, so as to introduce a new residual bandwidth, and then jump to step 2, and transmit the residual traffic by using the residual bandwidth.
Therefore, referring to fig. 2, the data center-oriented online scene low-bandwidth overhead traffic scheduling scheme of the present invention is implemented under the wide area network between the data centers according to the following steps:
and (1) operating the integrated controller to dynamically obtain a source node, a destination node, data volume, arrival time and deadline of each large transport stream. While running the OPPG algorithm, its scheduling scheme is calculated as each chunk transport stream arrives.
When the algorithm in the step (2) is initialized, all c are enablede＝0。
And (3) when each large transport stream arrives, calculating the minimum extra bandwidth overhead by using an OPPG algorithm, and specifically comprising the following steps:
step (3a) traverses all feasible paths of the stream by a depth-first search algorithm to find the duration [ a ] of the streami,ti]Inner bandwidth value cePaths not fully utilized, these paths with surplus bandwidth being used to transmit the data of the stream. If r isiAnd (4) finishing the transmission, terminating the algorithm, and turning to the step (3e), otherwise, turning to the step (3 b).
Step (3b) ordering all Ve＝0。
Step (3c) traverses all feasible paths of the flow through a depth-first search algorithm, and because all remaining bandwidth has been utilized in step 2, all feasible paths must have a congestion edge at any time, i.e., the bandwidth value used on the congestion edge at that time is ce. In the traversing process, the congestion edges of all the paths at all the moments are found out, and for a congestion edge e of a certain path at a certain moment, the bandwidth value c of the edge is calculatedeAfter increasing one unit, the amount d of data that can be additionally transmitted on the path at that moment is as small as possible
Figure BDA0001402581110000081
After step (3d), selecting VeMaximum edge e, order ce←ce+1, so as to introduce a new residual bandwidth, and then jumping to step (3a) to transmit the residual traffic by using the residual bandwidth.
And (3e) terminating the algorithm to obtain the scheduling scheme of the flow.
And (4) generating a flow scheduling scheme according to the algorithm result, and performing flow scheduling.
In summary, the invention provides a data center-oriented online scene low-bandwidth overhead traffic scheduling scheme. The scheme can ensure that all large transport streams are completed on time, and meanwhile, extra storage overhead is not introduced. Under the premise, the scheme greatly improves the link utilization rate, and minimizes the extra bandwidth leasing overhead brought by each flow, thereby saving the operation cost of the data center.

Claims (1)

1. A data center-oriented online scene low-bandwidth overhead traffic scheduling scheme is characterized by being implemented under a wide area network among data centers according to the following steps:
step (1), dividing a lease period into a plurality of transmission time slots, namely 1And (V, E) representing links between the data centers, wherein V is a node set of the directed graph and represents a set of all the data centers, E is an edge set of the directed graph and represents a set of all the links, and a quintuple r is usedi＝(si，ti，di，ai，τi) To represent a large transport stream, where si，ti，di，ai，τiRespectively representing a source node, a destination node, data volume, arrival time and deadline of the ith chunk transport stream; the operation integrated controller dynamically obtains a source node, a destination node, data volume, arrival time and deadline of each large transport stream;
step (2), running an online renting scheme generation algorithm, calculating a scheduling scheme of each large block transmission stream when the transmission stream arrives, and enabling all bandwidth values c to be initialized when the algorithm is initializede＝0；
And (3) when each large transport stream arrives, calculating the minimum extra bandwidth overhead by using an online leasing scheme generation algorithm, and specifically comprising the following steps:
step (3a) of traversing all feasible paths of the stream by a depth-first search algorithm to find the duration [ a ] of the streami，ti]Inner bandwidth value cePaths that are not fully utilized, and the paths with residual bandwidth are used for transmitting the data of the stream;
by P0Expressing the objective function under three constraints of flow constraint, capacity constraint and integer constraint:
Figure FDA0002466099780000011
a minimized optimization problem;
wherein, there are two flow constraints, the first flow constraint is:
Figure FDA0002466099780000012
and v ≠ si，v≠ti，
Figure FDA0002466099780000013
And t ∈ N+
δ+(v) Representing the set of all directed edges, δ, starting from node v-(v) Representing the set of all directed edges, x, ending with node vi，e(t) represents the amount of data transmitted at the t-th instant on connection e for the ith request, N+Represents a set of positive integers;
another flow constraint is:
Figure FDA0002466099780000021
wherein, delta+(si) Represented by node siSet of all directed edges as starting points, δ-(si) Represented by node siA set of all directed edges that are end points;
the capacity constraint is:
Figure FDA0002466099780000022
wherein, ceFor the value of bandwidth leased on edge e, represents the number of units of bandwidth leased on link e by the data center owner, δcRepresenting the size of a unit bandwidth, δtRepresenting the size of each time slice;
the integer constraint is:
Figure FDA0002466099780000023
wherein N represents a set of natural numbers;
if r isiIf the transmission is finished, the algorithm is terminated, and the step (3e) is carried out, otherwise, the step (3b) is carried out;
step (3b), first order all Ve＝0，VeThe rental value for edge e;
a step (3c) of,all feasible paths of the flow are traversed by the depth-first search algorithm, and because all the remaining bandwidth is already utilized in step 2, all feasible paths must have a congestion edge at any time, that is, the bandwidth value used on the congestion edge at that time is ce(ii) a In the traversing process, the congestion edges of all the paths at all the moments are found out, and for a congestion edge e of a certain path at a certain moment, the bandwidth value c of the edge is calculatedeAfter increasing one unit, the amount d of data that can be additionally transmitted on the path at that moment is as small as possible
Figure FDA0002466099780000024
Wherein u iseRepresents the price per bandwidth of link e;
step (3d), after which V is selectedeMaximum edge e, order ce←ce+1, so as to introduce a new residual bandwidth, and then skipping to the step (3a) to transmit the residual traffic by using the residual bandwidth;
step (3e), the algorithm is terminated, and the scheduling scheme of the flow is obtained;
and (4) generating a flow scheduling scheme according to the algorithm result, and performing flow scheduling.
CN201710805436.1A 2017-09-08 2017-09-08 Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme Active CN107483355B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710805436.1A CN107483355B (en) 2017-09-08 2017-09-08 Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710805436.1A CN107483355B (en) 2017-09-08 2017-09-08 Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme

Publications (2)

Publication Number Publication Date
CN107483355A CN107483355A (en) 2017-12-15
CN107483355B true CN107483355B (en) 2020-06-16

Family

ID=60585181

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710805436.1A Active CN107483355B (en) 2017-09-08 2017-09-08 Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme

Country Status (1)

Country Link
CN (1) CN107483355B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108833294B (en) * 2018-08-08 2020-10-30 清华大学 Low-bandwidth-overhead flow scheduling method for data center wide area network
CN109150597A (en) * 2018-08-08 2019-01-04 清华大学 The bandwidth cost of cloud service-oriented provider reduces method
CN109617710B (en) * 2018-11-09 2020-07-07 西北大学 Large data transmission bandwidth scheduling method with deadline constraint between data centers
CN115442313B (en) * 2022-07-20 2023-09-19 中通服咨询设计研究院有限公司 Online scheduling system for wide area deterministic service flow

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106209683A (en) * 2016-06-20 2016-12-07 中国科学院上海高等研究院 Data transmission method based on data center's wide area network and system
CN106911567A (en) * 2017-01-13 2017-06-30 西北大学 The a plurality of fixed route bandwidth scheduling method towards in the high performance network of big data

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9806987B2 (en) * 2014-07-30 2017-10-31 Futurewei Technologies, Inc. Method and apparatus for reducing response time in information-centric networks

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106209683A (en) * 2016-06-20 2016-12-07 中国科学院上海高等研究院 Data transmission method based on data center's wide area network and system
CN106911567A (en) * 2017-01-13 2017-06-30 西北大学 The a plurality of fixed route bandwidth scheduling method towards in the high performance network of big data

Also Published As

Publication number Publication date
CN107483355A (en) 2017-12-15

Similar Documents

Publication Publication Date Title
CN107483355B (en) Data center-oriented online scene low-bandwidth overhead traffic scheduling scheme
CN108260169B (en) QoS guarantee-based dynamic service function chain deployment method
CN110297699B (en) Scheduling method, scheduler, storage medium and system
CN108566659B (en) 5G network slice online mapping method based on reliability
CN103309738B (en) User job dispatching method and device
Ordónez et al. Robust capacity expansion of network flows
CN103036792B (en) Transmitting and scheduling method for maximizing minimal equity multiple data streams
CN111030835B (en) Task scheduling model of TTFC network and message scheduling table generation method
CN110502328B (en) Offshore edge computing trusted cooperative task migration method
CN108965014B (en) QoS-aware service chain backup method and system
Zhao et al. A unified modeling framework for distributed resource allocation of general fork and join processing networks
WO2023040022A1 (en) Computing and network collaboration-based distributed computation offloading method in random network
CN107454009B (en) Data center-oriented offline scene low-bandwidth overhead traffic scheduling scheme
CN104683488A (en) Flow-type calculation system as well as dispatching method and dispatching device of flow-type calculation system
CN108833294B (en) Low-bandwidth-overhead flow scheduling method for data center wide area network
CN109412963B (en) Service function chain deployment method based on stream splitting
CN114071582A (en) Service chain deployment method and device for cloud-edge collaborative Internet of things
CN109379281A (en) A kind of traffic scheduling method and system based on time window
CN109873772A (en) Stream scheduling method, device, computer equipment and storage medium based on stable matching
CN101013955A (en) Fast simulated annealing for traffic matrix estimation
CN115907038A (en) Multivariate control decision-making method based on federated split learning framework
CN110048966B (en) Coflow scheduling method for minimizing system overhead based on deadline
Li et al. Redundancy-guaranteed and receiving-constrained disaster backup in cloud data center network
CN108833297A (en) Priority classification method, dispatching method and the device of big data flow
CN103617083A (en) Storage scheduling method and system, job scheduling method and system and management node

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