WO2018054369A1 - Procédé et appareil de sélection d'un nœud de transmission de données de diffusion en continu - Google Patents

Procédé et appareil de sélection d'un nœud de transmission de données de diffusion en continu Download PDF

Info

Publication number
WO2018054369A1
WO2018054369A1 PCT/CN2017/103067 CN2017103067W WO2018054369A1 WO 2018054369 A1 WO2018054369 A1 WO 2018054369A1 CN 2017103067 W CN2017103067 W CN 2017103067W WO 2018054369 A1 WO2018054369 A1 WO 2018054369A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
load
transmission
transmission node
current
Prior art date
Application number
PCT/CN2017/103067
Other languages
English (en)
Chinese (zh)
Inventor
董文新
刘东东
Original Assignee
北京奇虎科技有限公司
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 北京奇虎科技有限公司 filed Critical 北京奇虎科技有限公司
Publication of WO2018054369A1 publication Critical patent/WO2018054369A1/fr

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
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • 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
    • H04L47/60—Queue scheduling implementing hierarchical scheduling

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a method for selecting a stream data transmission node, and a device for selecting a stream data transmission node.
  • Streaming data is data that can be transmitted over the network by means of streaming.
  • Large website systems often rely on the Content Delivery Network (CDN) to transmit streaming data.
  • CDN Content Delivery Network
  • the content distribution network is added to the existing network.
  • a new layer of network structure allows users to transfer data to or from the corresponding node.
  • a transmission node is deployed for buffering stream data, and provides services to users nearby. Since the number of users is large and the performance of the transmission node is limited, a group of transmission nodes generally provide services to users in an area, using one.
  • the scheduling node selects the transmission node for the transmission of the stream data. Because of the bottleneck of the processing capability of one scheduling node, the efficiency of selecting the transmission node is reduced.
  • the present invention has been made in order to provide a method and apparatus for selecting a stream data transmission node that overcomes the above problems or at least partially solves the above problems.
  • a method for selecting a stream data transmission node includes: selecting at least one transport node for current stream data from a current transport node cluster; and pre-acquiring according to a pre-acquired current load of the selected transport node Estimating the actual load of the transmitting node to cope with the cluster of scheduling nodes; and correcting the selected transport node according to the actual load.
  • a device for selecting a stream data transmission node comprising: a transmission node selection module, configured to select at least a current stream data from a current transmission node cluster a transmission node; a load estimation module, configured to estimate, according to a pre-acquired current load of the selected transmission node, an actual load of the transmission node to the cluster of the scheduling node; and a transmission node correction module, configured to be based on the actual load Correct the selected transport node.
  • a computer program comprising computer readable code, when said computer readable code is executed on an electronic device, causing said electronic device to perform selection according to said stream data transfer node method.
  • the method and apparatus for selecting a stream data transmission node can select at least one transport node for the current stream data from the current transport node cluster, thereby realizing the selection of the transport node for the transmission of the stream data through the cluster of the dispatch node, thereby solving the problem.
  • a single scheduling node selects the processing capability bottleneck problem of the transmission node for the transmission of the stream data, and improves the efficiency of selecting the transmission node.
  • each scheduling node can estimate the actual load of the transmission node, and correct the transmission node with excessive actual load to the load. Too large a transmission node avoids the problem that the scheduling node in the cluster of scheduling nodes simultaneously selects the transmission node for the transmission of the stream data and the load of the single transmission node is too large.
  • FIG. 1 is a flow chart showing the steps of a method for selecting a stream data transmission node according to Embodiment 1 of the present invention
  • FIG. 2 is a flow chart showing the steps of a method for selecting a stream data transmission node according to Embodiment 2 of the present invention
  • FIG. 3 is a schematic flow chart showing the steps of a method for selecting a stream data transmission node according to Embodiment 3 of the present invention
  • FIG. 4 is a flow chart showing the steps of a method for selecting a stream data transmission node according to Embodiment 4 of the present invention.
  • FIG. 5 is a structural block diagram of a device for selecting a stream data transmission node according to Embodiment 5 of the present invention.
  • Figure 6 shows schematically a block diagram of an electronic device for performing the method according to the invention
  • Fig. 7 schematically shows a storage unit for holding or carrying program code implementing the method according to the invention.
  • FIG. 1 is a schematic flowchart of the steps of a method for selecting a stream data transmission node according to the first embodiment of the present invention, which may specifically include the following steps:
  • Step 101 Select at least one transport node for the current stream data from the current transport node cluster.
  • the stream data is data that can be transmitted on the network by means of streaming, and the stream data can be transmitted from one node to at least one transport node.
  • a plurality of stream data transmissions can exist simultaneously on one transmission node, and the stream data that can be carried by one transmission node is determined by the load capacity of the transmission node.
  • the scheduling node may select a transmission node for the stream data, and the plurality of scheduling nodes may form a cluster of the scheduling node.
  • the transmission of stream data mostly exists in the content distribution network, because one transmission node cannot bear a high amount of concurrency, and there are usually multiple transmission nodes in the content distribution network, and the transmission node that distributes the stream data transmission by the scheduling node To achieve load balancing of the transit nodes.
  • the scheduling node is configured to select at least one transit node for the transmission of the stream data according to the stream data transmission request, and may select according to the network location where the originating end of the stream data transmission request is located and/or the load condition of all the transmitting nodes that the scheduling node can schedule, You can also select the transit node in the history that is selected for the current stream data.
  • the scheduling node can select at least one transmission for the transmission of the stream data in multiple application scenarios.
  • the node specifically, in the live video application scenario, the scheduling node selects at least one transport node for the streaming data transmission of the live video client, and the video live client transmits the recorded video to at least one transport node by using a streaming method.
  • the video stream data can be transmitted to each node on the content distribution network, so that other clients can obtain; in the remote camera application scenario, the scheduling node selects at least one transport node for the streaming data transmission of the remote camera client, and the remote camera client
  • the recorded video is streamed to at least one of the transport nodes so that the video stream data can be transmitted to various nodes on the content distribution network so that other clients can obtain it.
  • Step 102 Estimate, according to the current load of the selected transmission node that is obtained in advance, the actual load of the transmission node to the cluster of the scheduling node.
  • the scheduling node may obtain the current load of all the transmitting nodes that can be scheduled by the scheduling node, and the current load reflects the load of the transmitting node, which may include the current active connection number, the current load, and the bandwidth usage. Wait.
  • the scheduling node in the scheduling node cluster selects at least one transmission node for the transmission of the stream data according to the stream data transmission request, and the scheduling nodes have no connection with each other, that is, one scheduling node selects one transmission node for the transmission of one stream data, and the other
  • the scheduling node cannot know in real time the load that the transmission node increases due to the new transmission.
  • the scheduling node selects a transmission node for the transmission of a stream data, it estimates the load added to the transmission node due to the transmission of the stream data, and estimates the cluster of the scheduling node according to the number of scheduling nodes in the cluster of the scheduling node. All the dispatching nodes add the load to the transit node, and the estimated dispatch node cluster is used as the estimated actual load for the sum of the new load and the current load of the transport node.
  • the actual load is the load of the transmission node estimated by the scheduling node, and is not necessarily the exact load of the transmission node at this time. Specifically, the sum of the load added to the transmission node corresponding to each scheduling node set in advance, or the product of the load added to the transmission node and the set load multiplier corresponding to the current stream data is used as the The cluster of scheduling nodes is added to the load of the transmitting node.
  • the transit node in the content distribution network may send the current load to a management node, and the management node sends the current load of each transport node to a storage node, and the scheduling node periodically acquires the transport nodes from the storage node. Current load.
  • Step 103 Correct the selected transmission node according to the actual load.
  • the scheduling node determines whether it is necessary to correct the selected transmission node by comparing the actual load of the transmission node with the load performance parameter of the transmission node, where the load performance is Parameters are parameters that describe the maximum load capacity of the transit node, such as the maximum number of active connections, the maximum load, and the maximum network bandwidth. Specifically, if the actual load of the selected transit node exceeds the load performance parameter of the transit node, the scheduling node selects other transit nodes that do not exceed the load performance parameter for the current stream data.
  • An implementation manner of modifying the selected transmission node may be that the scheduling node may select among other transmission nodes randomly or in a set order, and determine whether the new modification needs to be corrected by comparing the actual load and the maximum load of the newly selected transmission node.
  • the selected transport node may be that the scheduling node may select among other transmission nodes randomly or in a set order, and determine whether the new modification needs to be corrected by comparing the actual load and the maximum load of the newly selected transmission node.
  • the scheduling node maintains a pre-acquired current load of the selected transport node, and each time the scheduling node selects the transport node, the estimated load node should update the current load of the dispatch node cluster actual load. Until the scheduling node acquires the current load of the transit node again, the current load recorded in the dispatch node is updated with the latest acquired current load.
  • the transmission node by selecting at least one transmission node for the current stream data from the current transmission node cluster, the transmission node is selected by the scheduling node cluster for the transmission of the stream data, and the single scheduling node is solved as the stream.
  • the transmission of data selects the processing capability bottleneck of the transmission node, which improves the efficiency of selecting the transmission node.
  • each scheduling node can estimate the actual load of the transmission node, and correct the transmission node with excessive actual load to the load. Too large a transmission node avoids the problem that the scheduling node in the cluster of scheduling nodes simultaneously selects the transmission node for the transmission of the stream data and the load of the single transmission node is too large.
  • the implementation of selecting at least one transit node for the current stream data from the current transport node cluster may be the current flow from the current transport node cluster according to a history record.
  • the data selects at least one transport node; wherein the history record indicates a transport node that has been used by the corresponding stream data.
  • the history record indicates a transmission node that is used by the corresponding stream data, where the corresponding stream data includes other stream data that is required by the same mobile client as the current stream data, and stream data that is the same as the current stream data, and the record is recorded in the scheduling node.
  • the node corresponds to the history record of the transport node selected by the stream data, specifically, at least one transport node corresponding to other stream data required by the same mobile client in the history record, or at least one transport node previously selected for the same stream data, scheduling After receiving the scheduling request, the node first searches the history record for the transmission node selected for the corresponding stream data. If yes, the transport node selected in the history for the corresponding stream data is used as the transport node selected by the scheduling node for the current stream data.
  • the method further includes: for the transmission node actually selected after the modification, adopting a pre-prepared transmission node for the actual selection
  • the estimated actual load replaces the recorded current load.
  • the actually selected transmission node after the modification is the transmission node whose actual load does not exceed the maximum load; if the actual load does not exceed the selected
  • the load performance parameter of the transmission node is corrected, and the actually selected transmission node is still the original selected transmission node.
  • the actual load of the actually selected transmission node after the correction is replaced with the current load recorded by the scheduling node, so that the scheduling node can use the estimated actual load as the scheduling center record before the scheduling node obtains the updated current load from the transmission node.
  • the current load of the transit node is used by the scheduling node to estimate the actual load of the next stream data transmission to avoid the problem that the load of the single transmission node is too large.
  • FIG. 2 it is a schematic flowchart of the steps of the method for selecting a stream data transmission node according to the second embodiment of the present invention, which may specifically include the following steps:
  • Step 201 Regularly obtain the current load of all the transit nodes in the current transport node cluster.
  • each scheduling node of the scheduling node cluster since the real-time acquisition of the current load from all the transmission nodes will generate a large number of input and output requirements, for the overall efficiency consideration, each scheduling node of the scheduling node cluster periodically acquires the current load of all the transmission nodes, respectively. All the transmitting nodes send the current load to each scheduling node separately, or all the transmitting nodes send the current load to a management node for maintaining the state of the transmitting node, and then the scheduling node periodically acquires all the transmitting nodes from the management node. Current load.
  • Step 201 is a preferred step compared to other embodiments.
  • the transmission node in the content distribution network may send the current load to a management node, and the management node sends the current load of the transmission node to a storage node, and the scheduling node periodically acquires the current load of each transmission node from the storage node.
  • Step 202 Construct a corresponding load interval for each transmission node according to load performance parameters of each transmission node.
  • the load performance parameter refers to a parameter describing the maximum load capacity of the transmission node, such as the maximum number of active connections, the maximum load, the maximum network bandwidth, etc., and the maximum load capacity of each transmission node is different, and the load interval is It means that a certain interval is allocated to each transmission node in the demarcated numerical range, so that the collection of the intervals of all the transmission nodes can occupy the entire numerical range. And each interval has no intersection and the boundary points are continuous, that is, the intervals allocated to any two transmission nodes have no common parts, and the two boundary points of one interval must be the boundary points of two adjacent intervals.
  • the scheduling node constructs a corresponding load interval for each transmission node, and is used for subsequently selecting a transmission node of the current stream data by calculating a load interval to which the current stream data belongs.
  • a preferred implementation manner may be to delimit a sum of values of load performance parameters of each transmission node as a maximum value, and a value range of zero to a minimum value, in which the load performance parameters of each transmission node are determined.
  • the value of the load is allocated to a certain load interval, so that the collection of the intervals of all the transmission nodes can occupy the entire range of values.
  • the specific allocation method of allocating a certain load interval according to the load performance parameters of each transmission node may not be limited, and may be set according to actual needs. Scale and then assign according to the set ratio.
  • the value of the load performance parameter of one transmission node to the value of the load performance parameter of the previous transmission node, and the interval between the values of the load performance parameter of the previous transmission node is not included as the load interval of the transmission node
  • any suitable hash function may be used to generate a hash value corresponding to the value of the load performance parameter of each transmission node, and then the hash value corresponding to the load performance parameter of each transmission node is used for each transmission.
  • the node constructs the corresponding load interval. Since the string corresponding to the value of the load performance parameter of the transit node may be non-standard, the hash function can be used to transform the load performance parameter of any length into a fixed-length hash value.
  • the width of the load interval of each transmission node is proportional to the value of the load performance parameter, and the implementation of the corresponding load interval for each transmission node according to the load performance parameter of each transmission node is implemented.
  • the method may be that a load interval is constructed for each transmission node by using the value of each load performance parameter as the width of the load interval. Therefore, the value of the load performance parameter of the transmission node is proportional to the load interval of the transmission node, so that the probability of selecting the transmission node is proportional to the maximum load capacity of the transmission node, optimizing the transmission node allocation efficiency, and making greater use of the difference.
  • the load of the transmission node with the maximum load capacity.
  • the width of the load interval of each transmission node is proportional to the value of the load performance parameter.
  • One implementation may be to delimit a sum of values of load performance parameters of each transmission node to a maximum value, and to minimize The value range of the value, when constructing the load interval of a transmission node, the end point of the above load interval may be used as the starting point, and the value of the load performance parameter of the transmission node is the interval width, and the end point of the current load interval is obtained, where the first The starting point of each load interval is zero, and the boundary point where two numerically adjacent load intervals intersect may belong to the previous load interval or may belong to the latter load interval. For example, there are three transfer nodes, the first one of which is the first transfer node.
  • the load performance parameter of the point is 10, the load performance parameter of the second transmission node is 30, and the load performance parameter of the third transmission node is 20, and a numerical range of 60 is the maximum value and 0 is the minimum value.
  • the load interval of the first transmission node is from 0 to 10
  • the load interval of the second transmission node is from 10 to 40
  • the load interval of the fourth transmission node is from 40 to 60.
  • Step 203 Search for a load interval to which the current stream data belongs according to the data identifier of the current stream data.
  • the data identifier of the current stream data is used to uniquely identify the stream data.
  • the data identifier of the current stream data is mapped to the value range in which the load interval of all the transit nodes is located, and the data identifier is mapped to the load interval, which is the load interval to which the current stream data belongs.
  • the specific calculation manner is not limited in the embodiment of the present invention. .
  • the method for searching for the load interval of the current stream data according to the data identifier of the current stream data may be: calculating an integer value corresponding to the data identifier; The sum of the integer value of the data identifier and the value of the load performance parameter of all the transit nodes is used as a remainder operation; the load interval corresponding to the result of the scavenging is searched as the load interval to which the current stream data belongs. Calculating the sum of the data attributes and the load performance parameters of all the transmitting nodes as a remainder operation; searching for the load interval to which the result of the redundancy is attributed.
  • the data identifier may be an arbitrary string, and the string is converted into an integer value by any feasible method.
  • the method of the present invention does not limit the specific method, for example, the CRC32 algorithm (full name cyclic redundancy check 32 algorithm) may be adopted.
  • the data identifier is converted to an integer value.
  • the integer value of the data identification can be divided by the sum of the values of the load performance parameters of all the transmission nodes, and the remainder obtained must be less than all
  • the result of the redundancy can be distributed to each load interval, so that at least one transmission node selected for the current stream data has randomness, because the width of the load interval of the transmission node and the value of the load performance parameter of the transmission node become Proportional, so the probability that each transmission node is selected is proportional to the value of the load performance parameter of the transmission node, and the selection probability of higher load performance is greater, and load balancing of each transmission node is realized.
  • the scheduling node obtains a hash value corresponding to the data identifier of the current stream data by using any applicable hash function, and divides the hash value by the maximum value of the created numerical range, and takes the remainder, and obtains the remainder.
  • the interval in the range of values is the load interval to which the current stream data belongs.
  • any suitable hash function can be used to generate a hash value corresponding to the data identifier of the current stream data.
  • the MD5 digest function (MD5, full name Message Digest Algorithm MD5, Chinese name message digest algorithm fifth edition) can be used, and the first eight bytes of the MD5 digest function result can be used as a hash value.
  • the first four bytes of the MD5 digest function result can be taken as a hash value.
  • fewer, more, or different bytes may be used under conditions that only warrant sufficient fullness of the hash result.
  • the load interval to which the current stream data belongs is calculated according to the hash value corresponding to the data identifier of the current stream data.
  • the data identifier includes at least one of a client preset field, a client identifier, a timestamp information, a random number, and a stream data unique identifier.
  • the data identifier of the current stream data is used to uniquely identify the stream data, and may include, for example, at least one of a client preset field, a client identifier, a timestamp information, a random number, and a stream data unique identifier.
  • the client preset field may be a field specially set to distinguish the stream data according to a unified rule
  • the client identifier refers to the machine identifier or other network identifier of the client
  • the timestamp information refers to the string or the encoded information used to identify the stream.
  • the date and time of the data and the random number refer to the randomly generated number.
  • the unique identifier of the stream data refers to the identifier information that can uniquely identify the stream data. For example, MD5 value (MD5, full name Message Digest Algorithm MD5, Chinese name message digest algorithm fifth edition) .
  • Step 204 Select, from the current transmission node cluster, a transmission node corresponding to the belonging load interval.
  • the scheduling node selects, according to the load interval to which the current stream data belongs, the transport node corresponding to the load segment to which the load belongs.
  • a preferred implementation manner of estimating the actual load of the transit node to the cluster of the scheduling node may include steps 205 to
  • Step 205 Estimating the first load increment that is caused by the transmission node after the current stream data is allocated to the transit node.
  • the first load increment refers to the computing resource, the network bandwidth resource, and the like, and the current active connection, the current load, and the current load of the transit node, after the scheduling node selects the transport node.
  • the current network bandwidth usage and the like estimate the load that the current node data is allocated to the transit node and cause the transmission node to increase.
  • Step 206 Estimate, according to the first load increment, a second load increment that is caused by the scheduling node cluster scheduling the transmission node to increase.
  • scheduling a transport node by using a preset scheduling node cluster causes the The increased load increment of the transmission node, or the product of the first load increment increased by the transmission node and the set load multiplier after the current flow data is allocated to the transmission node, as the scheduling node cluster scheduling
  • the transmission node causes a second load increment that is increased by the transmission node.
  • an implementation of estimating, by the scheduling node cluster, the second load increment caused by the transmitting node after scheduling the transmitting node may be that the product of the first load increment and the set load multiplier is used as a second load increment that is caused by the transport node after the scheduling node clusters the scheduling of the transport node, and the set load is increased.
  • the multiple characterizes the number of times the scheduling node cluster has scheduled for the transmitting node.
  • the scheduling node in the cluster of the scheduling node selects at least one transmission node for the transmission of the stream data according to the stream data transmission request, and the scheduling nodes have no connection with each other, that is, one scheduling node selects one for the transmission of the stream data.
  • the transmission node, the other scheduling node cannot know the load increased by the transmission node due to the newly added transmission, and sets the load multiple to characterize the scheduling times of the scheduling node cluster to the transmission node.
  • After the current stream data is allocated to the transmission node causing the product of the first load increment increased by the transmission node and the set load multiple, as the scheduling node clusters the transmission node, causing the transmission The second load increment added by the node.
  • the load multiplier may be determined according to the number of scheduling nodes in the scheduling node cluster. The larger the number of scheduling nodes in the scheduling node cluster, the larger the set load multiplier, but the set load multiplier is smaller than the number of scheduling nodes in the scheduling node cluster. In practical applications, the load multiplier can be debugged according to actual usage. The present invention does not limit this. For example, if there are five scheduling nodes in the scheduling node cluster, the set load multiplier can be set to 3.5.
  • Step 207 The sum of the second load increment added by the transit node and the current load is caused by the cluster of the scheduling node, and the actual load of the cluster of the scheduling node is responded to by the transit node.
  • the cluster of the scheduling node causes the second load increment added by the transit node to be summed with the current load, and the actual load of the transport node to the cluster of the scheduling node is obtained, where the actual load is not necessarily the The actual load of the transmitting node at this time, but the load of the transmitting node estimated by the dispatching node.
  • Step 208 If the actual load exceeds the value of the load performance parameter of the transmission node, replace the transmission node with another transmission node whose value of the transmission node load performance exceeds the actual load.
  • the scheduling node selects the value of the load performance of the transmission node.
  • Another transmission node that exceeds the actual load specifically, one way may be to select the next transmission node in the load interval, and estimate the actual load of the transmission node, if the estimated actual load of the next transmission node exceeds
  • the load performance parameter of the transmission node is then selected from the next transmission node in the load interval until a transmission node whose actual load does not exceed the load performance of the transmission node is found.
  • Step 208 is a preferred step compared to other embodiments.
  • the current load of all the transport nodes in the current transport node cluster is periodically acquired, and the corresponding load interval is constructed for each transport node according to the load performance parameters of each transport node, according to the current flow data.
  • the data identifier is used to search for the load interval to which the current stream data belongs, and the transport node corresponding to the load interval that belongs to the current transport node cluster is selected, and the transport node is selected by the dispatch node cluster for the transmission of the stream data, and the solution is solved.
  • a single scheduling node selects the processing capability bottleneck problem of the transmission node for the transmission of the stream data, and improves the efficiency of selecting the transmission node.
  • the result of the redundancy can be distributed to each load interval, so that at least one transmission node selected for the current stream data has randomness, due to the width of the load interval of the transmission node and the value of the load performance parameter of the transmission node. It is proportional, so the probability that each transmission node is selected is proportional to the value of the load performance parameter of the transmission node, and the selection probability of higher load performance is greater, and load balancing of each transmission node is realized.
  • the transmission node is replaced by another transmission node whose value of the load performance of the transmission node exceeds the actual load, because the current load cannot reflect the real-time load condition of the transmission node, and there are multiple scheduling nodes that can simultaneously allocate transmission for the transmission of the stream data.
  • each scheduling node can estimate the actual load of the transmission node, and correct the transmission node with excessive actual load to a transmission node with no excessive load, thereby avoiding that the scheduling node in the scheduling node cluster is simultaneously stream data.
  • the transmission selects the problem that the transmission node is overloaded by a single transmission node.
  • FIG. 3 it is a schematic flowchart of the steps of the method for selecting a stream data transmission node according to the third embodiment of the present invention, which may specifically include the following steps:
  • Step 301 Receive a scheduling request of a mobile client for a transmission resource of current stream data.
  • the transmission resource refers to at least one transmission node that transmits current stream data, and specifically may be a transmission node or a transmission node cluster, where the transmission node cluster refers to a transmission node cluster composed of at least one transmission node, and the scheduling node may At least one transport node in the transport node cluster is selected as the transport node of the current stream data.
  • the transmission of stream data is mostly stored.
  • the content distribution network the content distribution network is set up to deploy transmission resources in different regions. Because a transmission node cannot withstand high concurrency, there are usually multiple transmission nodes in the content distribution network, and the transmission nodes that distribute stream data are transmitted by the scheduling node. .
  • Step 301 is a preferred step compared to other embodiments.
  • the mobile client sends a scheduling request to the scheduling node, requests transmission resources of the current stream data, and the scheduling node receives the scheduling request.
  • Step 302 Obtain information about a location where the mobile client is located, and select a transport node cluster of a corresponding area of the location where the mobile client is located as a current transport node cluster.
  • a transport node cluster refers to a transport node cluster composed of at least one transport node, and the scheduling node may select at least one transport node in the transport node cluster as a transport node of current stream data.
  • the location may be a geographical area, or may be a geographical positioning point, and may also be other precision or representation information.
  • the location information may be the identifier of the location, specific
  • the content may also be information of other dimensions of the location. Taking the location as the geographic area as an example, the location information may be the actual content of the area, the area identifier of the area, the crowd information in the area, and the like.
  • the scheduling node can obtain information about the location of the mobile client in various manners, and the present invention does not limit this, for example, extracting information about the location of the mobile client from the received scheduling request, or after receiving the scheduling request. Sending a location acquisition request to the mobile client, and the mobile client feeds back the information of the location to the scheduling node.
  • the scheduling node After obtaining the information about the location of the mobile client, the scheduling node searches for the area to which the location indicated by the information belongs according to the obtained location information, and selects at least one transmission resource in the area, for example, the location information of the mobile client is Beijing.
  • Zhongguancun, Haidian District belongs to Haidian District of Beijing, and selects the transmission resources in Haidian District of Beijing.
  • the content distribution network is set up to deploy transmission resources in different regions. For example, a transmission node or a transmission node cluster will be deployed in Haidian District, Beijing.
  • the scheduling node may select multiple transmission resources according to the information of the location of the obtained mobile client, or select multiple transmission resources according to the obtained multiple information of the location of the mobile client, where each Information selects one or more transmission resources.
  • the obtaining, by the location information of the mobile client, the location information of the mobile client is determined according to the at least one location indication information of the mobile client. Information.
  • the location indication information is at least one type, and may include a network address, location information, and the like. After obtaining the at least one location indication information of the mobile client, the location indication may be determined according to each location indication information that is obtained. The location corresponding to the information can also be combined with the acquired A plurality of position indication information collectively determine a position corresponding to the plurality of position indication information.
  • the determining, according to the location indication information of the mobile client, the information about the location of the mobile client is: extracting from the scheduling request The network address information of the mobile client; searching for information about the geographical area where the mobile client is indicated corresponding to the network address information.
  • the mobile client sends a scheduling request to the scheduling node, where the scheduling request carries the network address information of the mobile client, and the scheduling node extracts the network address from the scheduling request; the scheduling node can call the network address database, the slave network address
  • the geographical area information corresponding to the extracted network address is queried in the database, that is, the geographical area information of the mobile client is obtained, for example, Haidian District of Beijing.
  • another implementation manner of determining information about a location where the mobile client is located according to the at least one location indication information of the mobile client is: receiving the mobile client call
  • the location information obtained by the location locator is obtained by accessing the geographic area information of the operating system to obtain information about the geographical area of the mobile client indicated by the location information obtained by the interface.
  • the positioning information refers to the latitude and longitude information
  • the scheduling node sends a request for acquiring the geographic area to the mobile client.
  • the mobile client After receiving the request, the mobile client starts the location locating program to obtain the current latitude and longitude information, for example, acquires the GPS (full name: Global Positioning System, Chinese name: Global Positioning System) The latitude and longitude information fed back by the chip, accessing the geographic area information acquisition interface of the operating system, and obtaining the information of the geographical area where the mobile client is indicated by the positioning information.
  • the geographic area information obtaining interface of the operating system may obtain the corresponding geographical area according to the latitude and longitude information. Since the positioning information obtained by the location locating program is not affected by the domain name hijacking or the network address error, the accuracy of obtaining the location information of the mobile client is improved.
  • Step 303 Select at least one transport node for the current stream data from the current transport node cluster.
  • Step 304 According to the current load of the selected transmission node acquired in advance, the actual load of the transmission node to the scheduling node cluster is estimated.
  • Step 305 Correct the selected transmission node according to the actual load.
  • the transmission node cluster of the corresponding area of the information selects at least one transmission node for the current stream data from the current transmission node cluster, realizes determining the transmission node cluster according to the location, and selects the transmission node for the transmission of the stream data, and solves the problem.
  • a single scheduling node selects the processing capability bottleneck of the transmission node for the transmission of stream data, and improves the efficiency of selecting the transmission node.
  • each scheduling node can estimate the actual load of the transmission node, and correct the transmission node with excessive actual load to the transmission node with no excessive load, and avoid the scheduling node in the cluster of the scheduling node to simultaneously select the transmission of the stream data.
  • the method may include the following steps:
  • Step 401 Select at least one transport node for the current stream data from the current transport node cluster.
  • Step 402 Estimate, according to the current load of the selected transmission node that is obtained in advance, the actual load of the transmission node to the cluster of the scheduling node.
  • Step 403 Correct the selected transmission node according to the actual load.
  • Step 404 respectively detecting the transmission speeds of the selected plurality of transmission nodes.
  • the transmission speed refers to the speed of data transmission between the mobile client and the transmission node, and the data packet can be sent to each other through the mobile client and the transmission node, and the time taken to send the data packet is obtained by the data packet.
  • the transmission speed is obtained when the size and transmission time are used.
  • the selected plurality of transmission nodes can be used as candidate transmission nodes, so the plurality of candidate transmission nodes are measured at a speed to select a transmission node with a faster transmission speed, and the plurality of transmission nodes are the selected transmission nodes.
  • the scheduling node may acquire information about multiple locations of the mobile terminal according to different manners. The location of the information of different locations may be the same or different depending on the detection precision of different manners. Therefore, different multiple candidate transmission nodes are further selected according to different locations.
  • the scheduling node may determine whether to detect the uplink transmission speed or the downlink transmission speed of the mobile client to the transmission node according to whether the current stream data transmission is uplink transmission or downlink transmission, or may simultaneously detect the mobile client to transmit according to actual needs.
  • the uplink transmission speed and downlink transmission speed of the node may be determined whether to detect the uplink transmission speed or the downlink transmission speed of the mobile client to the transmission node according to whether the current stream data transmission is uplink transmission or downlink transmission, or may simultaneously detect the mobile client to transmit according to actual needs. The uplink transmission speed and downlink transmission speed of the node.
  • the implementation manner of separately detecting the transmission speeds of the selected multiple transmission nodes may be: determining, according to the scheduling request of the mobile client for the transmission resource of the current stream data, Corresponding type of speed measurement; transmitting a speed measurement strategy set for the speed type to the mobile client, and receiving a feedback transmission speed.
  • the scheduling node determines the currently required speed measurement type according to the scheduling request of the mobile client for the transmission resource of the current stream data, for example, the uplink speed measurement and the downlink speed measurement, and the speed measurement type is sent.
  • the mobile client sends the speed measurement according to the speed measurement type, and feeds the speed measurement result to the scheduling node, thereby determining which measurement speed to perform according to the current streaming data transmission requirement.
  • the method for determining the corresponding speed measurement type according to the scheduling request of the mobile client for the transmission resource of the current stream data may be: if the scheduling request corresponds to the request uplink When the resource is transmitted, the speed measurement type is an uplink speed measurement.
  • the speed measurement policy includes: sending a speed measurement request carrying the first data packet to each of the transmission nodes, and performing speed measurement according to the second data packet fed back by the transmission node for the first data packet, where the first data packet is greater than the first Set the amount of data.
  • the first set data amount refers to a minimum amount of data required for detecting a transmission speed
  • the mobile client separately sends a speed measurement request to each transmission node, where the speed measurement request carries the first data packet, and the transmission node receives the data packet.
  • the speed measurement request carries the first data packet
  • the transmission node receives the data packet.
  • the first data packet is received.
  • the transmitting node sends the second data packet to the mobile client, and the mobile client starts to send the first data packet and start receiving the second data packet.
  • the transmission speed of the first data packet is transmitted, that is, the uplink speed from the mobile client to the transmission node is obtained.
  • the second data packet when detecting the uplink speed, is smaller than the data amount of the first data packet, which reduces unnecessary network transmission, and further the second data packet is preferably smaller than the second set data amount, wherein the second setting data Amount is the minimum amount of data used to detect network connectivity.
  • another implementation manner of determining the corresponding speed measurement type may be: if the scheduling request corresponds to the request For downlink transmission resources, the speed measurement type is downlink speed measurement.
  • the speed measurement policy includes: sending a speed measurement request carrying the first data packet to each of the transmission nodes, and performing speed measurement according to the second data packet fed back by the transmission node for the first data packet, where the second data packet is greater than the first Set the amount of data.
  • the first set data amount refers to a minimum amount of data required for detecting a transmission speed
  • the mobile client separately sends a speed measurement request to each transmission node, where the speed measurement request carries the first data packet, and the transmission node receives the data packet.
  • the speed measurement request carries the first data packet
  • the transmission node receives the data packet.
  • the first data packet is received.
  • the transmitting node sends the second data packet to the mobile client, and the mobile client starts to send the first data packet and complete the second data packet by calculation.
  • the transmission speed of the transmission of the second data packet is obtained, that is, the downlink speed from the mobile client to the transmission node is obtained.
  • the first data packet when detecting the downlink speed, is smaller than the data volume of the second data packet, thereby reducing unnecessary network transmission, and further, the first data packet is preferably smaller than the second set data amount, wherein the second setting data is Quantity means The minimum amount of data used to detect network connectivity.
  • another implementation manner of determining the corresponding speed measurement type may be: if the scheduling request corresponds to the request For the uplink and downlink transmission resources, the speed measurement type is an uplink speed measurement and a downlink speed measurement.
  • the speed measurement policy includes: transmitting, to each of the transmission nodes, a speed measurement request carrying the first data packet, and performing speed measurement according to the second data packet fed back by the transmission node for the first data packet, the first data packet and the second data packet The data packets are all larger than the first set data amount.
  • the first set data amount refers to a minimum amount of data required for detecting a transmission speed
  • the mobile client separately sends a speed measurement request to each transmission node, where the speed measurement request carries the first data packet, and the transmission node receives the data packet. After the speed measurement request, the first data packet is received. After the first data packet is received, the transmitting node sends the second data packet to the mobile client, and the mobile client starts to send the first data packet and the transmitting node completes receiving the first data.
  • the time between the data packets is obtained, and the transmission speed of the first data packet is transmitted, that is, the uplink speed from the mobile client to the transmission node is obtained, and then the second data packet is started to be transmitted through calculation and the second data packet is received by the mobile client. Between the time, the transmission speed of the transmission of the second data packet is obtained, that is, the downlink speed from the mobile client to the transmission node is obtained.
  • another implementation manner of determining the corresponding speed measurement type may be: if the scheduling request corresponds to the request For data communication, the speed measurement type is network connectivity detection.
  • the speed measurement policy includes: transmitting, to each of the transmission nodes, a speed measurement request carrying the first data packet, and performing speed measurement according to the second data packet fed back by the transmission node for the first data packet, the first data packet and the second data packet The data packets are all smaller than the second set data amount.
  • the second set data amount refers to a minimum amount of data used for detecting network connectivity
  • the mobile client separately sends a speed measurement request to each transmission node, where the speed measurement request carries the first data packet, and the transmission node receives the data packet.
  • the first data packet is started to be received.
  • the transmission node sends the second data packet to the mobile client, thereby completing the detection of the network connectivity. Since the first data packet and the second data packet are both smaller than the second set data amount, unnecessary network transmission is reduced, and the network connectivity is detected to be more timely.
  • Step 405 Select a transmission node whose transmission speed meets the setting requirement as an actual transmission node.
  • the transmission speed meets the setting requirement, which is a transmission speed requirement that can meet the transmission requirement, and the specific setting requirement can be set according to actual needs, which is not limited by the embodiment of the present invention, and the scheduling node detects the selection. After the transmission speed of multiple transmission nodes, select one of them A transfer node that meets the set requirements.
  • the transmission node by selecting at least one transmission node for the current stream data from the current transmission node cluster, the transmission node is selected by the scheduling node cluster for the transmission of the stream data, and the single scheduling node is solved as the stream.
  • the transmission of data selects the processing capability bottleneck of the transmission node, which improves the efficiency of selecting the transmission node.
  • each scheduling node can estimate the actual load of the transmission node, and correct the transmission node with excessive actual load to be not excessively loaded.
  • the transmission node avoids the problem that the scheduling node in the scheduling node cluster simultaneously selects the transmission node for the transmission of the stream data and the load of the single transmission node is too large.
  • the transmission speeds of the selected plurality of transmission nodes are respectively detected, and the transmission node whose transmission speed meets the set requirement is selected, and the transmission nodes whose transmission speeds meet the requirements are selected by performing speed measurement on the plurality of selected transmission nodes, thereby realizing A transit node that is faster than uploading or downloading is preferred.
  • FIG. 5 it is a structural block diagram of a device for selecting a stream data transmission node according to Embodiment 5 of the present invention, which may specifically include the following modules:
  • the transmission node selection module 501 is configured to select at least one transmission node for the current stream data from the current transmission node cluster;
  • the load estimation module 502 is configured to estimate, according to the pre-acquired current load of the selected transmission node, the actual load of the transmission node to the cluster of the scheduling node;
  • the transmission node modification module 503 is configured to modify the selected transmission node according to the actual load.
  • the device further comprises:
  • the load interval construction module is configured to construct a corresponding load interval for each transmission node according to load performance parameters of each transmission node before selecting at least one transmission node for the current stream data from the current transmission node cluster, and each load interval has no intersection and boundary The point is continuous.
  • the width of the load interval of each transmission node is proportional to the value of the load performance parameter
  • the load interval construction module is specifically configured to construct a load interval for each transmission node with the value of each load performance parameter as the width of the load interval.
  • the transmission node selection module includes:
  • a load interval finding submodule configured to search, according to the data identifier of the current stream data, a load interval to which the current stream data belongs;
  • the transmission node selects a sub-module, and is configured to select, from the current transmission node cluster, a transmission node corresponding to the belonging load interval.
  • the load interval finding submodule comprises:
  • the integer value is calculated from the unit, and is used to calculate an integer value corresponding to the data identifier;
  • a remainder operation subunit configured to perform a remainder operation by using an sum of an integer value of the data identifier and a value of a load performance parameter of all the transmission nodes
  • the load interval search sub-unit is configured to search for a load interval corresponding to the spare result as a load interval to which the current flow data belongs.
  • the data identifier includes at least one of a client preset field, a client identifier, a timestamp information, a random number, and a stream data unique identifier.
  • the device further comprises:
  • the current load periodic acquisition module is configured to periodically acquire the current load of all the transit nodes in the current transport node cluster before selecting at least one transport node for the current stream data from the current transport node cluster.
  • the load estimation module includes:
  • a first load increment estimation submodule configured to estimate a first load increment that is caused by the transmission node after the current stream data is allocated to the transit node;
  • a second load increment estimation sub-module configured to estimate, according to the first load increment, a second load increment that is caused by the scheduling node to cluster the transport node and cause the transport node to increase;
  • An actual load submodule configured to cause, by the cluster of the scheduling node, a sum of a second load increment that is added by the transit node and the current load, as an actual load of the transit node to the cluster of the scheduling node.
  • the second load increment estimation sub-module is specifically configured to use the product of the first load increment and a set load multiplier to schedule the transmission node as the scheduling node cluster to cause the transmission.
  • the second load increment added by the node, the set load multiplier characterizing the number of times the scheduling node cluster schedules the transport node.
  • the transmission node modification module is specifically configured to: if the actual load exceeds a value of a load performance parameter of the transmission node, replace the transmission node with a value of a load performance of the transmission node that exceeds an actual load. Another transit node.
  • the device further comprises:
  • a scheduling request receiving module configured to receive a scheduling request of the mobile client for the transmission resource of the current stream data before selecting the transmission node for the current stream data from the current transmission node cluster;
  • a transmission node cluster selection module configured to acquire information about a location of the mobile client, and select a transmission node cluster of a corresponding area of the location where the mobile client is located as a current transmission node cluster.
  • the selected transmission node comprises a plurality
  • the device further comprises:
  • a transmission speed detecting module configured to separately detect transmission speeds of the selected plurality of transmission nodes
  • the selection module is used to select a transmission node whose transmission speed meets the setting requirement as an actual transmission node.
  • the transmission node selection module is configured to select at least one transmission node for the current stream data from the current transmission node cluster according to a history record; wherein the history record indicates a transmission used by the corresponding stream data node.
  • the history record may be saved in the selection device of the stream data transmission node, or stored in a certain module of the device, or may be obtained by requesting from a third party, and the present invention does not Make a limit.
  • the transmission node by selecting at least one transmission node for the current stream data from the current transmission node cluster, the transmission node is selected by the scheduling node cluster for the transmission of the stream data, and the single scheduling node is solved as the stream.
  • the transmission of data selects the processing capability bottleneck of the transmission node, which improves the efficiency of selecting the transmission node.
  • each scheduling node can estimate the actual load of the transmission node, and correct the transmission node with excessive actual load to the load. Too large a transmission node avoids the problem that the scheduling node in the cluster of scheduling nodes simultaneously selects the transmission node for the transmission of the stream data and the load of the single transmission node is too large.
  • modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment.
  • the modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components.
  • any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods so disclosed, or All processes or units of the device are combined.
  • Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.
  • the various component embodiments of the present invention may be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof.
  • a microprocessor or digital signal processor may be used in practice to implement some or all of the components of the method and apparatus for selecting a streaming data transmission node in accordance with an embodiment of the present invention.
  • the invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing some or all of the methods described herein.
  • Such a program implementing the present invention may be stored on a computer readable medium or may have There are one or more signal forms. Such signals may be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
  • FIG. 6 shows an electronic device in which a method of selecting a streaming data transmission node according to the present invention can be implemented.
  • the electronic device conventionally includes a processor 610 and a computer program product or computer readable medium in the form of a memory 620.
  • the memory 620 may be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), an EPROM, a hard disk, or a ROM.
  • Memory 620 has a storage space 630 that stores program code 631 for performing any of the method steps described above.
  • storage space 630 storing program code may store various program codes 631 for implementing various steps in the above methods, respectively.
  • the program code can be read from or written to one or more computer program products.
  • Such computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks.
  • Such a computer program product is typically a portable or fixed storage unit such as that shown in FIG.
  • the storage unit may have a storage section, a storage space, and the like arranged similarly to the storage 620 in the electronic device of FIG.
  • the program code can be compressed, for example, in an appropriate form.
  • the storage unit stores computer readable program code 631' for performing the method steps of the present invention, i.e., code readable by a processor, such as 610, which, when executed by the electronic device, causes the electronic
  • the apparatus performs the various steps in the methods described above.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

La présente invention concerne un procédé et un appareil destinés à sélectionner une ressource de transmission pour des données de diffusion en continu. Le procédé comporte les étapes consistant à: sélectionner au moins un nœud de transmission pour des données actuelles de diffusion en continu à partir d'un groupe actuel de nœuds de transmission; en fonction d'une charge actuelle du nœud de transmission sélectionné obtenu préalablement, pré-estimer une charge réelle du nœud de transmission pour s'adapter à un groupe de nœuds de programmation; et corriger le nœud de transmission sélectionné en fonction de la charge réelle. Selon la solution technique de la présente invention, un nœud de transmission est sélectionné pour la transmission de données de diffusion en continu au moyen d'un groupe de nœuds de programmation; le problème du goulot d'étranglement des performances de traitement dans la sélection du nœud de transmission par un nœud de programmation unique pour la transmission de données de diffusion en continu est résolu; le rendement de sélection du nœud de transmission est amélioré; et le problème d'une charge excessivement élevée d'un nœud de transmission unique du fait de la sélection simultanée de nœuds de transmission par des nœuds de programmation dans le groupe de nœuds de programmation pour la transmission des données de diffusion en continu est résolu.
PCT/CN2017/103067 2016-09-23 2017-09-25 Procédé et appareil de sélection d'un nœud de transmission de données de diffusion en continu WO2018054369A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610849757.7 2016-09-23
CN201610849757.7A CN106453122B (zh) 2016-09-23 2016-09-23 一种流数据传输节点的选取方法和装置

Publications (1)

Publication Number Publication Date
WO2018054369A1 true WO2018054369A1 (fr) 2018-03-29

Family

ID=58169566

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/103067 WO2018054369A1 (fr) 2016-09-23 2017-09-25 Procédé et appareil de sélection d'un nœud de transmission de données de diffusion en continu

Country Status (2)

Country Link
CN (1) CN106453122B (fr)
WO (1) WO2018054369A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111934947A (zh) * 2020-07-17 2020-11-13 中国联合网络通信集团有限公司 测速方法、测速调度服务器、终端设备及可读存储介质
CN112700292A (zh) * 2020-12-24 2021-04-23 航天信息股份有限公司 一种通过自动调度税控设备进行电子发票开具的方法及系统
CN113949741A (zh) * 2021-10-14 2022-01-18 北京奇艺世纪科技有限公司 一种调度方法、调度装置、电子设备及存储介质

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106453122B (zh) * 2016-09-23 2019-06-04 北京奇虎科技有限公司 一种流数据传输节点的选取方法和装置
CN109347869B (zh) * 2018-11-28 2021-07-06 京东数字科技控股有限公司 集群间通信安全因子的生成方法、装置、介质及电子设备
CN111683132B (zh) * 2020-06-04 2022-07-19 重庆金窝窝网络科技有限公司 一种基于微服务架构的业务分发方法及相关装置
CN113452767B (zh) * 2021-06-23 2022-11-25 新华三大数据技术有限公司 一种应用于服务集群内的负载均衡方法及装置
CN115242780A (zh) * 2022-07-26 2022-10-25 北京知道创宇信息技术股份有限公司 文件下载方法、装置、下载管理器及可读存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120036215A1 (en) * 2010-08-04 2012-02-09 Laura Perryman Optimized Data Stream Upload
CN102394931A (zh) * 2011-11-04 2012-03-28 北京邮电大学 一种基于云的用户访问请求调度方法
US20140215075A1 (en) * 2013-01-30 2014-07-31 Electronics And Telecommunications Research Institute Load balancing apparatus and method based on estimation of resource usage
CN104834722A (zh) * 2015-05-12 2015-08-12 网宿科技股份有限公司 基于cdn的内容管理系统
CN105007312A (zh) * 2015-07-03 2015-10-28 叶秀兰 一种云计算服务器自适应负载均衡控制方法及控制系统
CN105959405A (zh) * 2016-06-24 2016-09-21 北京兰云科技有限公司 Cdn视频调度系统、方法以及cdn调度服务器和客户端
CN106453122A (zh) * 2016-09-23 2017-02-22 北京奇虎科技有限公司 一种流数据传输节点的选取方法和装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101626399B (zh) * 2009-08-11 2012-03-28 华中科技大学 一种音乐在线播放的调度及控制方法
CN101695050A (zh) * 2009-10-19 2010-04-14 浪潮电子信息产业股份有限公司 一种基于网络流量自适应预测的动态负载均衡方法
CN104852857B (zh) * 2014-02-14 2018-07-31 航天信息股份有限公司 基于负载均衡的分布式数据传输方法和系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120036215A1 (en) * 2010-08-04 2012-02-09 Laura Perryman Optimized Data Stream Upload
CN102394931A (zh) * 2011-11-04 2012-03-28 北京邮电大学 一种基于云的用户访问请求调度方法
US20140215075A1 (en) * 2013-01-30 2014-07-31 Electronics And Telecommunications Research Institute Load balancing apparatus and method based on estimation of resource usage
CN104834722A (zh) * 2015-05-12 2015-08-12 网宿科技股份有限公司 基于cdn的内容管理系统
CN105007312A (zh) * 2015-07-03 2015-10-28 叶秀兰 一种云计算服务器自适应负载均衡控制方法及控制系统
CN105959405A (zh) * 2016-06-24 2016-09-21 北京兰云科技有限公司 Cdn视频调度系统、方法以及cdn调度服务器和客户端
CN106453122A (zh) * 2016-09-23 2017-02-22 北京奇虎科技有限公司 一种流数据传输节点的选取方法和装置

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111934947A (zh) * 2020-07-17 2020-11-13 中国联合网络通信集团有限公司 测速方法、测速调度服务器、终端设备及可读存储介质
CN112700292A (zh) * 2020-12-24 2021-04-23 航天信息股份有限公司 一种通过自动调度税控设备进行电子发票开具的方法及系统
CN112700292B (zh) * 2020-12-24 2024-03-12 航天信息股份有限公司 一种通过自动调度税控设备进行电子发票开具的方法及系统
CN113949741A (zh) * 2021-10-14 2022-01-18 北京奇艺世纪科技有限公司 一种调度方法、调度装置、电子设备及存储介质
CN113949741B (zh) * 2021-10-14 2023-07-21 北京奇艺世纪科技有限公司 一种调度方法、调度装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN106453122A (zh) 2017-02-22
CN106453122B (zh) 2019-06-04

Similar Documents

Publication Publication Date Title
WO2018054369A1 (fr) Procédé et appareil de sélection d'un nœud de transmission de données de diffusion en continu
KR102305064B1 (ko) 비디오 라이브 방송 방법 및 장치
EP2721504B1 (fr) Procédé de traitement de fichier, système et système à grappes de serveurs pour stockage en nuage
CN113497817B (zh) 流量调度方法、装置、cdn网络的边缘节点及服务器
US9867011B2 (en) Identifying proximity history of computer devices
CN106357776B (zh) 一种流数据的传输资源的选取方法和装置
US20140025723A1 (en) Cloud storage system and data storage and sharing method based on the system
CN109558065B (zh) 数据删除方法及分布式存储系统
CN110402567B (zh) 信息为中心的网络中基于中心性的缓存
CN106657371B (zh) 一种传输节点的调度方法和装置
CN106453460B (zh) 一种文件分发方法、装置和系统
JP6904169B2 (ja) タスク配備プログラム、タスク配備方法、およびタスク配備装置
CN106686101B (zh) 一种流数据的传输集群的调度方法和装置
CN102333130A (zh) 一种访问缓存服务器的方法、系统及缓存智能调度器
US20210392200A1 (en) Systems and methods for remote network topology discovery
WO2019080232A1 (fr) Procédé et appareil de transmission d'informations de tâches dans un système de flux de travaux, et dispositif informatique
CN105656964B (zh) 数据推送的实现方法及装置
CN101150593A (zh) 一种上传数据的方法及系统
US20150006571A1 (en) Method And Apparatus For Enabling Queries In An Information-Centric Network
CN117278628B (zh) 数据传输方法、装置、系统、计算机设备和存储介质
CN106790354B (zh) 一种防数据拥堵的通信方法及其装置
CN113094350A (zh) 一种基于区块链的分布式文件存储方法及装置
CN112019604A (zh) 边缘数据传输方法和系统
CN103944972A (zh) 一种上传、下载数据的方法及网关
CN114466031B (zh) 一种cdn系统节点配置方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17852436

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17852436

Country of ref document: EP

Kind code of ref document: A1