WO2016098866A1 - 通信速度推定装置、通信速度推定方法及びプログラム - Google Patents
通信速度推定装置、通信速度推定方法及びプログラム Download PDFInfo
- Publication number
- WO2016098866A1 WO2016098866A1 PCT/JP2015/085405 JP2015085405W WO2016098866A1 WO 2016098866 A1 WO2016098866 A1 WO 2016098866A1 JP 2015085405 W JP2015085405 W JP 2015085405W WO 2016098866 A1 WO2016098866 A1 WO 2016098866A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- communication
- throughput
- information
- network
- server
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
- H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
- H04W72/044—Wireless resource allocation based on the type of the allocated resource
- H04W72/0446—Resources in time domain, e.g. slots or frames
Definitions
- the present invention relates to a communication speed estimation device, a communication speed estimation method, and a program for estimating a network communication speed when using a network.
- CDN Contents Delivery Network
- the present invention has been made paying attention to the above circumstances, and an object of the present invention is to provide a technique capable of estimating a usable communication speed in advance.
- a second means for causing When the user performs a new communication, in advance, based on at least one of a terminal, a server, and a network that the user intends to use, and a time zone that includes a time at which communication is to be performed And a third means for estimating a communication speed in the new communication based on the read throughput performance information.
- the block diagram which shows the structure of the communication-speed estimation apparatus which concerns on one Embodiment of this invention The block diagram which shows the structure of the communication log
- storage information of the calendar information database with which the communication-speed estimation apparatus shown in FIG. 1 is provided.
- the communication speed estimation device obtains information indicating a communication history, for example, information identifying a user terminal, a server, and a network used in the communication, and a communication time.
- a communication history information storage unit Each of the user terminal, the server, and the network is stored in a communication history information storage unit as communication log information, and information representing a time zone is included for each time zone of a predetermined granularity based on the stored communication log information.
- Throughput performance information is calculated and stored in the quality information storage unit.
- the communication speed estimation device obtains the throughput information information of the communication performed in the past same time zone using the user terminal, server and network used in the communication before the user starts a new communication. A search is made from the quality information storage unit, and the communication speed of the new communication is estimated based on the throughput record information.
- the communication speed estimation device calculates a plurality of pieces of communication log information in the communication history information storage unit when calculating the throughput performance information
- the same server the same server Communication log information relating to communication using the same network or the same user terminal is integrated, and based on the integrated communication log information, throughput performance information of each of the server, the network, and the user terminal is calculated for each time zone.
- the communication speed estimation device when calculating the throughput record information, the communication speed estimation device, for each of the plurality of integrated communication log information, for each of a server, a network, and a user terminal A throughput value is calculated for each of the calculated throughput values, an average of the throughput values included in the upper preset range among the calculated throughput values is calculated, and the calculated throughput average value is used as the throughput performance information.
- the communication speed estimation device when estimating the communication speed, corresponds to a time zone including a time at which communication is to be performed and tries to use a terminal used by a user.
- Throughput information corresponding to each of the server and the network is read from the quality information storage unit, and the read throughput performance information having the smallest value is selected from the read throughput performance information.
- the communication speed in the new communication is estimated.
- the communication speed estimation device calculates the throughput result information by calculating the upper and lower values of the throughput values calculated for each of the server, the network, and the user terminal for each time period. At least one of a ratio of values included in a preset range and a value representing a variance of the calculated throughput values is calculated. Then, when estimating the communication speed, the average value of the calculated throughput, the ratio of the calculated throughput value included in the preset upper and lower ranges, and the variance of the calculated throughput value The communication speed in the new communication is estimated based on at least one of the values representing the degree of.
- the communication speed is estimated by dividing by the time zone such as date and day of the week.
- TCP Transmission Control Protocol
- the present invention also uses this point for estimating the communication speed.
- the reason why the throughput performance information is used for estimating the communication speed is that the end-to-end throughput in communication has a network property that is limited by one bottleneck point. For example, when the user terminal has a processing capacity of 300 Mbps, the server has 100 Mbps, and the network has a processing capacity of 5 Mbps, the throughput of this route is 5 Mbps.
- throughput performance information is calculated for each of the server, the network, and the user terminal for each time period, and the communication speed is estimated based on the throughput performance information.
- the communication log information related to communication using the same server, network, or user terminal is integrated, Based on this integrated communication log information, it is preferable to calculate throughput performance information for each server, network, and user terminal for each time period. That is, for a plurality of communications performed in the past, those having the same server, network, and user terminal calculate and store common throughput record information.
- the number of pieces of throughput performance information to be stored can be significantly reduced compared to the case of calculating and storing the throughput performance information for each of the server, network, and user terminal individually for all communications. It is possible to reduce the capacity of the storage unit and reduce the search processing speed of the throughput record information.
- certain communication log information using a server is information indicating that the throughput is obtained when the server is used, and therefore may be limited by other elements such as a network and a user terminal. . For this reason, even if there is a variation in throughput in multiple communications performed under the same conditions, the influence of the variation is affected by using the higher quantile value that excludes such a possible lower value. This makes it possible to estimate speed with higher reliability.
- the throughput information calculated for each of the user terminal, the server, and the network is selected with the smallest value to set the communication speed. It is good to estimate. In this way, it is possible to accurately estimate the end-to-end communication speed according to the nature of the network that the end-to-end throughput in communication is limited by one bottleneck point.
- the communication speed in the new communication may be estimated by referring to these variance values and upper or lower percentage values.
- FIG. 1 is a block diagram showing a configuration of a communication speed estimation apparatus according to an embodiment of the present invention.
- the communication speed estimation apparatus 1 includes a processing unit including a CPU (Central Processing Unit) and a storage unit using a nonvolatile memory that can be written and read at any time.
- the storage unit is provided with a communication history database 40, a network database 50, a calendar information database 60, and a quality database 70 as storage units according to this embodiment.
- the communication history database 40 is used to store communication log information generated for each communication by the communication history analysis unit 10 described later.
- the calendar information database 60 a holiday weekday flag for identifying whether the date is a weekday or a holiday is associated with the date.
- FIG. 5 shows an example of the storage result.
- the network database 50 stores information for identifying the network in association with the network ID (NW_ID). Specifically, for a fixed line, an IP address defined for each Internet service provider (Internet Service Provider: IPS) is used, and for a mobile phone line (Public Land Mobile Mobile Network: PLMN), a base station ID is a mobile phone number ( (PLMN number) and information indicating the network type.
- FIG. 6 is a diagram showing an example.
- the quality database 70 includes a network quality database 71, a server quality database 72, and a terminal quality database 73.
- the network quality database 71 includes network identification information (NWID) used in communication performed in the time zone of a predetermined granularity, for example, every hour, and a communication speed calculation unit 20 described later.
- NWID network identification information
- FIG. 7A shows an example of the storage result.
- the server quality database 72 includes server identification information (server ID) used in communication performed in the time zone every hour, and a communication speed calculation unit 20 described later.
- the average throughput value and the throughput deviation value calculated by are stored.
- FIG. 7B shows an example of the storage result.
- the identification information (user ID) of the user terminal used in the communication performed in the time zone, and the throughput average value calculated by the communication speed calculation unit 20 described later are stored in the terminal quality database 73. And the throughput deviation value is stored.
- the terminal quality database 73 also stores data in the same format as the database format shown in FIGS. 7A and 7B.
- the processing unit includes a communication history analysis unit 10, a communication rate calculation unit 20, and a communication rate prediction unit 30 as control function units according to this embodiment.
- Each of these control function units is realized by causing the CPU to execute a program stored in a program memory (not shown).
- each of the communication history analysis unit 10, the communication speed calculation unit 20, and the communication speed prediction unit 30 may be a hardware circuit having a corresponding processing logic.
- the communication history analysis unit 10 generates communication log information based on the communication history every time communication is performed and accumulates it in the communication history database 40. As shown in FIG. A network ID conversion unit 12, a calendar information conversion unit 13, and a log input unit 14 are provided.
- the syntax analysis unit 11 receives communication history information of the communication from the communication management device every time communication is performed.
- the communication history information includes communication date / time, server ID, user ID, network information, and throughput information.
- the syntax analysis unit 11 extracts necessary parameters from the communication history information according to the format of the communication log information.
- the network ID conversion unit 12 refers to the network database 50 to convert the network information included in the communication history information into a network ID according to the granularity of the network used for estimating the communication speed.
- the calendar information conversion unit 13 refers to the calendar information database 60 to determine whether the communication date / time included in the communication history information is a holiday or a weekday, and also according to the granularity of time used for estimating the communication speed. Convert communication date and time to time zone.
- the log input unit 14 obtains the parameters extracted by the syntax analysis unit 11, that is, the server ID, user ID, and throughput information, the network ID converted by the network ID conversion unit 12, and the calendar information conversion unit 13.
- the obtained calendar information is stored in the communication history database 40 in association with each other.
- the communication speed calculation unit 20 periodically calculates the throughput for each of the server, the network, and the user terminal for each time zone based on the communication log information stored in the communication history database 40.
- FIG. As shown, an update determination unit 21 and a speed update unit 22 are provided.
- the update determination unit 21 searches the network database 50, calendar information database 60, and communication history database 40 each time a startup request is periodically input by a timer (not shown), so that new communication log information is added. It is determined whether or not. Then, when new communication log information is added, the communication log information is supplied to the speed update unit 22.
- the speed update unit 22 calculates a throughput average value and a throughput deviation value for each of the server, the network, and the user terminal for each time zone based on the communication log information supplied from the update determination unit 21. Then, the average throughput value and the throughput deviation value calculated for each of the server, the network, and the user terminal are stored in the network quality database 71, the server quality database 72, and the terminal quality database 73 of the quality database 70, respectively.
- the communication speed prediction unit 30 has a function of responding to communication speed information in the new communication when an inquiry request for communication speed information is input prior to the new communication. As shown in FIG. Unit 31, network ID conversion unit 32, calendar information conversion unit 33, and throughput prediction unit 34.
- the syntax analysis unit 31 extracts the server ID and the user ID from the communication speed inquiry request in the same manner as the syntax analysis unit 11 of the communication history analysis unit 10 described above. Similar to the network ID conversion unit 12 of the communication history analysis unit 10, the network ID conversion unit 32 refers to the network database 50 so that the network information included in the communication speed information inquiry request is estimated as a communication speed. It is converted into a network ID according to the granularity of the network used for.
- the calendar information conversion unit 33 refers to the calendar information database 60 to determine whether the communication date and time included in the communication speed information inquiry request is a holiday or a weekday. In addition to the determination, the communication date and time is converted into a time zone according to the granularity of time used for estimating the communication speed.
- the throughput prediction unit 34 is based on the server ID and user ID extracted by the syntax analysis unit 31, the network ID converted by the network ID conversion unit 32, and the calendar information obtained by the calendar information conversion unit 33.
- the throughput record information of the server, the network, and the user terminal in the same time zone that is, the throughput average value and the throughput deviation value are read from the quality database 70.
- the communication speed of a new communication is predicted based on the read throughput information, and the predicted communication speed is returned to the request source.
- the request source is, for example, a user terminal, a server that provides a content distribution service, or the like.
- the communication speed estimation device 1 may be realized by causing a computer including a CPU and a memory to execute a program, or by using a hardware circuit having logic corresponding to each functional unit. You can also
- FIG. 8 shows an example of a hardware configuration when the communication speed estimation apparatus 1 is realized by a computer.
- the communication speed estimation apparatus 1 includes a CPU 101, a memory 102, an HDD (hard disk drive) 103, and an input / output unit 104.
- Programs corresponding to the processes of the communication history analysis unit 10, the communication speed calculation unit 20, and the communication speed prediction unit 30 are stored in the memory 102 or the HDD 103.
- the CPU 101 reads out and executes the program, thereby performing processing of the communication history analysis unit 10, the communication speed calculation unit 20, and the communication speed prediction unit 30.
- Each database is stored in the memory 102 or the HDD 103. According to the program, the CPU 101 accesses the database and reads / writes data.
- the input / output unit 104 performs data input / output with the outside.
- syntax analysis processing of the communication history information is performed by the syntax analysis unit 11 of the communication history analysis unit 10.
- server ID, user ID, and throughput information which are parameters necessary for creating communication log information, are extracted in accordance with the format of communication log information.
- the throughput information in the communication history information input from the communication management device is, for example, the throughput measured at the user terminal when the user terminal downloads content from the server.
- the user ID and server ID in the communication history information correspond to the user terminal and the server.
- the throughput is a data transfer amount per unit time and is an example of network quality. As the network quality, another index may be used instead of the throughput.
- the network ID conversion unit 12 performs a process of converting the network information included in the communication history information into a network ID according to the granularity of the network used for estimating the communication speed by referring to the network database 50. Is called.
- the fixed line has a granularity in units of ISP and is converted into an IP address issued for each ISP.
- the cellular phone line has a base station granularity, and the base station ID included in the network information is output as it is without being converted.
- the calendar information conversion unit 13 refers to the calendar information database 60 to determine whether the communication date / time included in the communication history information is a holiday or a weekday. At the same time, the communication date and time is converted into a time zone in units of one hour, for example, according to a granularity of time set in advance.
- the server ID, user ID, and throughput information extracted by the syntax analysis unit 11, the network ID converted by the network ID conversion unit 12, and the calendar information output from the calendar information conversion unit 13 are logged. They are associated with each other by the input unit 14 and stored in the communication history database 40 as communication log information.
- the server is classified by time zone based on the communication log information stored in the communication history database 40.
- Throughput is calculated for each of the network and the user terminal. That is, first, the update determination unit 21 searches the network database 50, calendar information database 60, and communication history database 40, respectively, and determines whether or not new communication log information has been added. When new communication log information is added, the communication log information is supplied to the speed update unit 22 as an update target list.
- the speed update unit 22 based on the update target list including the communication log information supplied from the update determination unit 21, the average throughput value and the throughput deviation value for each of the server, the network, and the user terminal for each time zone. Are calculated respectively.
- the throughput TPS j (t) of the server j at the date and time t is obtained by extracting only the data of the user who used the server j in the time zone including the date and time t from the target communication log information, It is obtained by taking the average of x% values.
- the “throughput average value” here is, for example, when the user A uses the server j 10 times in the time period (when there is 10 throughput data), and adds the throughput data. It is the value divided by 10. Such an average throughput value is obtained for each user using the server j.
- TPS j (t) ave (rep x (L m
- SI m j)) (1)
- the function ave is a function for obtaining an average value
- the function rep x is a function for extracting only a value (percentage value) corresponding to a ratio included in the upper predetermined range x among given numerical values.
- the value obtained by the above equation corresponds to the “throughput average value” in FIG. 7B.
- the “throughput deviation value” is calculated by taking the square root of the distribution of log information using the server j as a value indicating how much the throughput when the server j is used varies.
- Throughputs TPN k (t) and TPU i (t) are calculated for the network and user terminals in the same manner.
- the throughputs TPS j (t), TPN k (t), and TPU i (t) calculated for each of the server, network, and user terminal as described above are the server quality database 72 and network quality of the quality database 70, respectively. It is stored in the database 71 and the terminal quality database 73.
- the communication speed prediction unit 30 uses the new throughput information stored in the quality database 70 to A process for estimating the communication speed of a simple communication is performed as follows.
- the syntax analysis unit 31 extracts the server ID and the user ID from the communication speed inquiry information request.
- the network ID conversion unit 32 converts the network information included in the communication speed information inquiry request into a network ID according to the granularity of the network used for estimating the communication speed by referring to the network database 50.
- the calendar information conversion unit 33 refers to the calendar information database 60 to determine whether the communication date / time included in the communication speed information inquiry request is a holiday or a weekday, and for estimating the communication speed.
- the communication date and time is converted into a time zone according to the granularity of the time used. Note that these conversion processes are the same as the processes performed by the communication history analysis unit 10 described above.
- the server ID and user ID extracted by the syntax analysis unit 31 the value and ID converted by the network ID conversion unit 32, and the calendar obtained by the calendar information conversion unit 33.
- the throughputs TPS j (t), TPN k (t), and TPU i (t) of the server, the network, and the user terminal in the communication performed in the past same time zone are read from the quality database 70. .
- a new communication speed is predicted based on the read throughput information.
- L m is a value indicating TI m indicating the day, day of the week and time of communication
- UI m indicating a value uniquely indicating a user
- SI m indicating a value uniquely indicating a server
- TP m which is a value indicating throughput.
- a value uniquely indicating a network an IP address group collected in units of ISP as described above, or a base station ID of a mobile phone line is used.
- TP (i, t) min (TPS j (t), TPN k (t), TPU i (t)) (2) Is required.
- j and k are the server and network identifiers that the user i intends to use.
- TP (i, t) may be expressed as TP (i, j, k, t).
- the one with the smallest value is selected from the throughputs TPS j (t), TPN k (t), and TPU i (t) of the server, the network, and the user terminal. If any of the values of TPS j (t), TPN k (t), and TPU i (t) is not obtained, the value is excluded and only the throughput of the remaining elements is used. Calculated.
- TPSj (t), TPNk (t), and TPUi (t) may be set to TP (i, t), and the smaller one of the two values is set to TP (i, t). It is good.
- the communication speed calculation unit 20 calculates the throughput
- the ratio of the values included in the preset ranges at the upper and lower levels of the throughput values calculated for each of the server, the network, and the user terminal for each time zone At least one of (percentage value) and the dispersion value of each calculated throughput value is calculated in advance. Then, when newly estimating the communication speed expected for communication, the average value of the calculated throughput, the calculated upper and lower percentage values, and the calculated throughput value variance A value may be selectively used to estimate the communication speed in the new communication.
- the communication speed calculation unit 20 gives a total of 100 throughput values (corresponding to the average throughput values shown in FIGS. 7A and 7B) for a plurality of servers, a plurality of networks, and a plurality of user terminals.
- the throughput value A is defined as the upper value
- the throughput value B is defined as the lower value.
- the ratio of the number of values greater than or equal to the throughput value A among the 100 values and the ratio of the number of values equal to or less than the throughput value B of the 100 values are calculated.
- a variance value of 100 values is calculated in advance.
- the ratio of the number of values greater than or equal to the throughput value A, the ratio of the number of values less than or equal to the throughput value B, and the dispersion value of 100 values may be calculated.
- the throughput TP (i, t) expected for the user i in the time period is obtained by the equation (2).
- the ratio of the number of values equal to or greater than the throughput value A is greater than a predetermined value, a large throughput can be expected in the time period, and therefore, adjustment is performed to increase TP (i, t).
- the ratio of the number of values equal to or less than the throughput value B is larger than a predetermined value, there is a high possibility that the throughput will be small in the time period, so adjustment is performed to reduce TP (i, t).
- the variance value is large, it can be predicted that the variation in network quality or the like is large in the time zone. Therefore, adjustment for decreasing TP (i, t) is performed in consideration of safety.
- the throughput TP (i, j, k, t) calculated as described above is returned to the request source as estimated communication speed information for the new communication.
- the communication speed estimation device communicates information indicating a communication history, for example, information identifying a user terminal, a server, and a network used in the communication, and communication each time communication is performed.
- Information representing a time zone including the time is stored in the communication history database 40 as communication log information, and the throughput for each of the user terminal, server, and network for each time zone of a predetermined granularity based on the communication log information.
- Result information is calculated and stored in the quality database 70.
- the communication speed estimation device uses the terminal, server, and network to be used for the communication before the user starts a new communication, and displays the throughput performance information of the communication performed in the past same time zone.
- a search is made from the quality database 70, and the communication speed of the new communication is estimated based on this throughput record information.
- the communication speed is estimated. Therefore, the end-to-end communication speed can be estimated prior to communication.
- an average value of the throughput values included in the upper preset range among a plurality of throughput values respectively calculated for the same server, network, and user terminal in the same time zone is calculated, and this throughput average
- the value is stored as throughput performance information. That is, a technique is adopted in which the upper quantile value is taken for the throughput of communication log information of communication using the server, the network, and the user terminal. With this method, it is possible to omit lower values that are likely to be rate-determining due to factors other than the factor of interest, and even if there are variations in throughput among multiple communications performed under the same conditions, the effects of these variations are not affected. This makes it possible to estimate the communication speed with higher reliability.
- the communication speed is estimated by selecting the throughput information calculated for each of the user terminal, the server, and the network with the smallest value. For this reason, it is possible to accurately estimate the end-to-end communication speed in consideration of the nature of the network that the end-to-end throughput in communication is limited by one bottleneck point.
- second means for calculating throughput performance information for each of the server, network, and user terminal used for the communication, and storing the calculated throughput performance information in a quality information storage unit For each band, second means for calculating throughput performance information for each of the server, network, and user terminal used for the communication, and storing the calculated throughput performance information in a quality information storage unit; and
- the corresponding throughput record information is preliminarily determined based on at least one of a terminal, a server, and a network that the user intends to use and a time zone that includes a time at which communication is to be performed.
- Third communication speed estimation and means are provided that.
- the communication history analysis unit 10 of the present embodiment is an example of the first means.
- the communication speed calculation unit 20 is an example of the second means.
- the communication speed prediction unit 30 is an example of the third means.
- the throughput record information is provided for each of the server, network, and user terminal used for the communication for each time period. Calculated and stored.
- a time zone including a time at which communication with at least one of a terminal used by the user, a server to be used, and a network is included is included in advance.
- Corresponding throughput record information is read, and the communication speed of the new communication is estimated.
- the communication speed is estimated by referring to the throughput performance information for each time zone by the server, network, and terminal used in the communication performed under the same previous conditions. Is done. Therefore, the end-to-end communication speed can be estimated more accurately prior to communication.
- the installation location of the communication speed estimation apparatus according to the present invention may be any of a server on the service provider side, a user terminal, a server operated by a provider, and a server operated by a third party.
- each configuration of a communication speed estimation device, a communication history analysis unit, a communication speed calculation unit, and a communication speed prediction unit, a communication log information format and throughput calculation processing procedure and its processing content, a processing procedure used for communication speed estimation, and The processing contents can be variously modified and implemented without departing from the gist of the present invention.
- the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage.
- various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
前記記憶された通信ログ情報に基づいて、時間帯別に、前記通信に使用したサーバ、ネットワーク及びユーザ端末の各々についてスループット実績情報を算出し、当該算出されたスループット実績情報を品質情報記憶部に記憶させる第2の手段と、
前記ユーザが新たな通信を行う場合に、事前に、当該ユーザが利用しようとする端末、サーバ及びネットワークのうちの少なくとも1つと、通信を行おうとする時刻が含まれる時間帯とに基づいて、対応するスループット実績情報を前記品質情報記憶部から読み出し、当該読み出されたスループット実績情報をもとに前記新たな通信における通信速度を推定する第3の手段と
を具備する通信速度推定装置が提供される。
先ずこの発明の実施形態の概要について説明する。
(構成)
図1は、この発明の一実施形態に係る通信速度推定装置の構成を示すブロック図である。
次に、以上のように構成された装置の動作を説明する。
通信が行われ、通信管理装置から当該通信の通信履歴情報が入力されると、先ず通信履歴解析部10の構文解析部11により上記通信履歴情報の構文解析処理が行われ、これにより通信ログ情報のフォーマットに合わせて、通信ログ情報の作成に必要なパラメータであるサーバID、ユーザID及びスループット情報が抽出される。なお、ここで通信管理装置から入力される通信履歴情報におけるスループット情報は、例えば、ユーザ端末がサーバからコンテンツをダウンロードした際に、当該ユーザ端末で測定されたスループットである。また、通信履歴情報におけるユーザID、及びサーバIDは、当該ユーザ端末と当該サーバに対応する。なお、スループットは単位時間あたりのデータ転送量であり、ネットワーク品質の一例である。ネットワーク品質として、スループットに代えて、他の指標を用いても良い。
通信速度計算部20では、タイマにより定期的に起動要求が入力されるごとに、上記通信履歴データベース40に記憶された通信ログ情報に基づいて、時間帯別に、サーバ、ネットワーク及びユーザ端末の各々についてスループットが算出される。すなわち、先ず更新判定部21において、ネットワークデータベース50、カレンダ情報データベース60及び通信履歴データベース40がそれぞれ検索され、新たな通信ログ情報が追加されたか否かが判定される。そして、新たな通信ログ情報が追加されていると、当該通信ログ情報が更新対象リストとして速度更新部22に供給される。
ここで、関数ave は平均値を求める関数、関数repxは与えられた数値のうち、上位の所定範囲xに含まれる割合に該当する値(パーセント値)のみを取り出す関数である。また、「Lm|SIm=j」は、通信ログ情報の中で、サーバjを利用したログ情報を示す。上記の式で得られた値は、図7Bにおける「スループット平均値」に相当する。「スループット偏差値」については、サーバjを利用した時のスループットがどの程度ばらついているかを示す値として、サーバjを利用したログ情報の分散の平方根をとることで算出される。ネットワーク、ユーザ端末についても同様の方針でスループットTPNk(t)、TPUi(t)が計算される。
新たな通信の開始に先立ち、通信速度情報問い合わせ要求が入力されると、通信速度予測部30では上記品質データベース70に記憶されたスループット実績情報をもとに、上記新たな通信の通信速度を推定するための処理が以下のように行われる。
TP(i,t)=min(TPSj(t),TPNk(t),TPUi(t)) …(2)
により求められる。ただし、j,kはそれぞれ、ユーザiが利用しようとするサーバ、及び、ネットワークの識別子である。なお、上記のTP(i,t)をTP(i,j,k,t)と表記してもよい。
以上詳述したように一実施形態では、通信速度推定装置は、通信が行われるごとに、通信履歴を表す情報、例えば当該通信で使用されたユーザ端末、サーバ及びネットワークを識別する情報と、通信時刻を含む時間帯を表す情報を通信ログ情報として通信履歴データベース40に記憶すると共に、当該通信ログ情報をもとに、所定の粒度の時間帯別に、上記ユーザ端末、サーバ及びネットワークの各々についてスループット実績情報を算出して品質データベース70に記憶する。そして、通信速度推定装置は、ユーザが新たな通信を開始する前に、当該通信に利用しようとする端末、サーバ及びネットワークを使用して過去の同一時間帯に行われた通信のスループット実績情報を上記品質データベース70から検索し、このスループット実績情報をもとに上記新たな通信の通信速度を推定するようにしている。
本発明の実施の形態により、ユーザが実行した通信の履歴を表す通信ログ情報を生成して通信履歴情報記憶部に記憶させる第1の手段と、前記記憶された通信ログ情報に基づいて、時間帯別に、前記通信に使用したサーバ、ネットワーク及びユーザ端末の各々についてスループット実績情報を算出し、当該算出されたスループット実績情報を品質情報記憶部に記憶させる第2の手段と、前記ユーザが新たな通信を行う場合に、事前に、当該ユーザが利用しようとする端末、サーバ及びネットワークのうちの少なくとも1つと、通信を行おうとする時刻が含まれる時間帯とに基づいて、対応するスループット実績情報を前記品質情報記憶部から読み出し、当該読み出されたスループット実績情報をもとに前記新たな通信における通信速度を推定する第3の手段とを具備する通信速度推定装置が提供される。
なお、この発明は上記実施形態に限定されるものではない。例えば、この発明に係る通信速度推定装置の設置場所は、サービス事業者側のサーバ、ユーザ端末、プロバイダが運用するサーバ、第三者が運用するサーバのいずれであってもよい。
10…通信履歴解析部
11、31…構文解析部
12、32…ネットワークID変換部
13、33…カレンダ情報変換部
14…ログ投入部
20…通信速度計算部
21…更新判定部
22…速度更新部
30…通信速度予測部
34…スループット予測部
40…通信履歴データベース
50…ネットワークデータベース
60…カレンダ情報データベース
70…品質データベース
71…ネットワーク品質データベース
72…サーバ品質データベース
73…端末品質データベース
101…CPU
102…メモリ
103…HDD
104…入出力部
Claims (7)
- ユーザが実行した通信の履歴を表す通信ログ情報を生成して通信履歴情報記憶部に記憶させる第1の手段と、
前記記憶された通信ログ情報に基づいて、時間帯別に、前記通信に使用したサーバ、ネットワーク及びユーザ端末の各々についてスループット実績情報を算出し、当該算出されたスループット実績情報を品質情報記憶部に記憶させる第2の手段と、
前記ユーザが新たな通信を行う場合に、事前に、当該ユーザが利用しようとする端末、サーバ及びネットワークのうちの少なくとも1つと、通信を行おうとする時刻が含まれる時間帯とに基づいて、対応するスループット実績情報を前記品質情報記憶部から読み出し、当該読み出されたスループット実績情報をもとに前記新たな通信における通信速度を推定する第3の手段と
を具備することを特徴とする通信速度推定装置。 - 前記第2の手段は、前記通信履歴情報記憶部に複数の通信ログ情報が記憶されている場合に、同一のサーバ、同一のネットワークまたは同一のユーザ端末を使用した通信に係る通信ログ情報を統合し、この統合された通信ログ情報に基づいて、時間帯別に、サーバ、ネットワーク及びユーザ端末各々のスループット実績情報を算出することを特徴とする請求項1記載の通信速度推定装置。
- 前記第2の手段は、前記統合された複数の通信ログ情報の各々について、時間帯別に、サーバ、ネットワーク及びユーザ端末各々のスループット値を算出し、この算出された各スループット値のうち上位の予め設定した範囲内に含まれるスループット値の平均を算出して、当該算出されたスループット平均値を前記スループット実績情報とすることを特徴とする請求項2記載の通信速度推定装置。
- 前記第3の手段は、通信を行おうとする時刻が含まれる時間帯に対応し、かつユーザが利用する端末、サーバ及びネットワークの各々に対応するスループット実績情報を前記品質情報記憶部から読み出し、当該読み出された各スループット実績情報のうちスループット値が最小のものを選択し、この選択されたスループット実績情報をもとに前記新たな通信における通信速度を推定することを特徴とする請求項1記載の通信速度推定装置。
- 前記第2の手段は、前記時間帯別に、サーバ、ネットワーク及びユーザ端末の各々について算出された各スループット値の上位及び下位の予め設定した範囲に含まれる値の割合と、前記算出された各スループット値の分散の度合いを表す値との少なくとも一方を算出する手段を、さらに備え、
前記第3の手段は、前記算出されたスループット平均値と、前記算出された上位及び下位の予め設定した範囲に含まれる値の割合と、前記算出されたスループット値の分散の度合いを表す値の少なくとも1つに基づいて、前記新たな通信における通信速度を推定することを特徴とする請求項4記載の通信速度推定装置。 - コンピュータと記憶部を備えた通信速度推定装置が実行する通信速度推定方法であって、
ユーザが実行した通信の履歴を表す通信ログ情報を生成して通信履歴情報記憶部に記憶させる第1の過程と、
前記記憶された通信ログ情報に基づいて、時間帯別に、前記通信に使用したサーバ、ネットワーク及びユーザ端末の各々についてスループット実績情報を算出し、当該算出されたスループット実績情報を品質情報記憶部に記憶させる第2の過程と、
前記ユーザが新たな通信を行う場合に、事前に、当該ユーザが利用しようとする端末、サーバ及びネットワークのうちの少なくとも1つと、通信を行おうとする時刻が含まれる時間帯とに基づいて、対応するスループット実績情報を前記品質情報記憶部から読み出し、当該読み出されたスループット実績情報をもとに前記新たな通信における通信速度を推定する第3の過程と
を具備することを特徴とする通信速度推定方法。 - 請求項1乃至5のいずれかに記載の通信速度推定装置が具備する各手段が行う処理を、前記通信速度推定装置が備えるコンピュータに実行させるプログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016564908A JP6383009B2 (ja) | 2014-12-18 | 2015-12-17 | 通信速度推定装置、通信速度推定方法及びプログラム |
US15/536,361 US10412182B2 (en) | 2014-12-18 | 2015-12-17 | Communication speed estimation apparatus, communication speed estimation method, and program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014-256412 | 2014-12-18 | ||
JP2014256412 | 2014-12-18 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016098866A1 true WO2016098866A1 (ja) | 2016-06-23 |
Family
ID=56126749
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/085405 WO2016098866A1 (ja) | 2014-12-18 | 2015-12-17 | 通信速度推定装置、通信速度推定方法及びプログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US10412182B2 (ja) |
JP (1) | JP6383009B2 (ja) |
WO (1) | WO2016098866A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019024154A (ja) * | 2017-07-21 | 2019-02-14 | 日本電信電話株式会社 | 品質制御装置、通信品質予測方法、及びプログラム |
CN114585013A (zh) * | 2022-03-01 | 2022-06-03 | 北京中网华通设计咨询有限公司 | 一种网络质量评估方法及*** |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10681416B2 (en) * | 2015-02-04 | 2020-06-09 | Nippon Telegraph And Telephone Corporation | Quality-of-experience optimization system, quality-of-experience optimization apparatus, recommend request apparatus, quality-of-experience optimization method, recommend request method, and program |
US10498615B2 (en) * | 2017-07-10 | 2019-12-03 | Google Llc | Method, apparatus and system for providing wireless network speed information |
JP7003705B2 (ja) * | 2018-02-06 | 2022-01-21 | 日本電信電話株式会社 | サーバ選択装置、サーバ選択方法及びプログラム |
JP7131394B2 (ja) * | 2019-01-04 | 2022-09-06 | 日本電信電話株式会社 | 品質推定システム、品質推定方法、及びプログラム |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5913041A (en) * | 1996-12-09 | 1999-06-15 | Hewlett-Packard Company | System for determining data transfer rates in accordance with log information relates to history of data transfer activities that independently stored in content servers |
JP2010166137A (ja) * | 2009-01-13 | 2010-07-29 | Hitachi Ins Software Ltd | マルチパス通信端末、その制御方法及びシステム |
JP2013200632A (ja) * | 2012-03-23 | 2013-10-03 | Fuji Xerox Co Ltd | サーバ、端末、プログラム |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8417811B1 (en) * | 2009-12-17 | 2013-04-09 | Amazon Technologies, Inc. | Predicting hardware usage in a computing system |
US9450839B2 (en) * | 2012-11-09 | 2016-09-20 | Lenovo Enterprise Solutions (Singapore) Pte. Ltd. | Efficient network bandwidth utilization in a distributed processing system |
US10491754B2 (en) * | 2016-07-22 | 2019-11-26 | Level 3 Communications, Llc | Visualizing predicted customer bandwidth utilization based on utilization history |
-
2015
- 2015-12-17 US US15/536,361 patent/US10412182B2/en active Active
- 2015-12-17 WO PCT/JP2015/085405 patent/WO2016098866A1/ja active Application Filing
- 2015-12-17 JP JP2016564908A patent/JP6383009B2/ja active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5913041A (en) * | 1996-12-09 | 1999-06-15 | Hewlett-Packard Company | System for determining data transfer rates in accordance with log information relates to history of data transfer activities that independently stored in content servers |
JP2010166137A (ja) * | 2009-01-13 | 2010-07-29 | Hitachi Ins Software Ltd | マルチパス通信端末、その制御方法及びシステム |
JP2013200632A (ja) * | 2012-03-23 | 2013-10-03 | Fuji Xerox Co Ltd | サーバ、端末、プログラム |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019024154A (ja) * | 2017-07-21 | 2019-02-14 | 日本電信電話株式会社 | 品質制御装置、通信品質予測方法、及びプログラム |
CN114585013A (zh) * | 2022-03-01 | 2022-06-03 | 北京中网华通设计咨询有限公司 | 一种网络质量评估方法及*** |
Also Published As
Publication number | Publication date |
---|---|
JPWO2016098866A1 (ja) | 2017-09-21 |
US20170374167A1 (en) | 2017-12-28 |
JP6383009B2 (ja) | 2018-08-29 |
US10412182B2 (en) | 2019-09-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6383009B2 (ja) | 通信速度推定装置、通信速度推定方法及びプログラム | |
US10091348B1 (en) | Predictive model for voice/video over IP calls | |
US10601698B2 (en) | Techniques for managing telemetry data for content delivery and/or data transfer networks | |
US9609052B2 (en) | Distributing application traffic to servers based on dynamic service response time | |
US10715638B2 (en) | Method and system for server assignment using predicted network metrics | |
US8271656B2 (en) | Decreasing latency in anonymity networks | |
US8849990B2 (en) | Optimized video streaming to client devices | |
US10771355B2 (en) | Method for measuring end-to-end internet application performance | |
US20160330500A1 (en) | Method for obtaining network information by a client terminal configured for receiving a multimedia content divided into segments | |
US20180077217A1 (en) | Method and apparatus for controlling multi-connection for data transmission rate improvement | |
CN110771122A (zh) | 使内容传送网络能够处理非预期流量激增的方法和网络节点 | |
Medagliani et al. | Overlay routing for fast video transfers in CDN | |
Viola et al. | Predictive CDN selection for video delivery based on LSTM network performance forecasts and cost-effective trade-offs | |
CN108632680B (zh) | 一种直播内容的调度方法、调度服务器及终端 | |
CN107113332A (zh) | 媒体客户端稳定状态的服务器侧预测 | |
US9537707B2 (en) | System for self organization of an internet protocol television grid | |
Timmerer et al. | Quality of experience of commercially deployed adaptive media players | |
US11470145B2 (en) | Server selection apparatus, server selection method and program | |
US9307000B2 (en) | Content distribution system, content distribution device, content distribution method and program | |
US9710861B2 (en) | Optimizing social information signaling | |
US20240163225A1 (en) | Providing estimated bandwidth ranges to network-attached entities | |
Suzuki et al. | Overlay Optimization for Cost Efficient P2P Streaming Service |
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: 15870068 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016564908 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15536361 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15870068 Country of ref document: EP Kind code of ref document: A1 |