KR20230044079A - Control method of node included in information centric network, and system - Google Patents

Abstract

Disclosed is a method for controlling a node, which comprises the steps of: receiving and storing one content; transmitting a virtual interest including information on the stored content to all surrounding nodes located within a certain number of hops from a node whenever the content is stored; receiving an interest requesting the content matching the virtual interest from at least one external node receiving the virtual interest; and transmitting the content matching the virtual interest to the external node. Therefore, the quality of a content service is improved.

Description

Control method of node included in information-centric network, and system { CONTROL METHOD OF NODE INCLUDED IN INFORMATION CENTRIC NETWORK, AND SYSTEM }

The present disclosure relates to a control method of a node included in an information-centric network or a method of providing content, and more particularly, an efficient access method for content stored in a cache of a node off a path to each node and system It is disclosed from the operating point of view.

TCP/IP technology, the basis of current Internet technology, was developed focusing on data transmission between two nodes based on location information such as IP addresses. However, considering the service patterns of current Internet users for large-capacity contents, TCP/IP technology causes communication inefficiency such as transmitting the same data repeatedly or generating large overhead to support the mobility of terminals. do.

To solve this problem, the concept of information-centric network (ICN) was introduced. In ICN, networking is performed based on content name information instead of content location information. In addition, copies of the content are stored in nodes on the network as the content is transmitted. Since the stored copy of the content is then serviced for requests for the same content, more efficient networking can be implemented. As a result, from the user's point of view, network delay is reduced by receiving content from a closer location, and as a result, a higher quality of service (QoS) can be experienced. In addition, network operators can efficiently operate networks by minimizing redundant data transmission. In addition, from a content provider's point of view, server infrastructure costs can be reduced by distributing the load on the server to the network cache.

Among various networking architectures that implement the ICN concept, NDN (Named Data Network) is a representative decentralized ICN architecture that has been studied mainly by academics in the US, Europe, and Asia, starting with UCLA. Compared to a centralized structure, NDN has the advantage of being easy to expand and maximizing the flexibility of routing, so research is being actively conducted.

Communication in NDN begins when a user sends a request message called an Interest to the network. Since the request message contains a unique content name, routing to the target content on the network can be performed based on the name information.

When a router (node) receives this request message, it searches a network cache called Content Store (CS) to see if the requested content exists. If a copy of the requested content exists in the CS, the router can deliver the requested content directly to the user in a data packet called a response without forwarding Interest to the next hop. If a copy of the requested content does not exist in the CS, the router can register the interest packet and the interface that received the interest packet in the PIT (Pending Interest Table). This is to ensure that the response packet that will arrive later can be sent to the interface where the interest packet came in (after all, the interest and response are delivered through the same routing path). Thereafter, the router may determine the next hop by performing longest prefix matching between entries registered in the Forwarding Information Base (FIB) based on the content name recorded on Interest. And, the router can forward the interest packet to the corresponding interface.

In the case of a response packet, routing may be performed through the following process. Specifically, when a router receives a response packet, it searches the PIT to find an entry with the same content name. In the matching PIT entry, it is also recorded through which interface the corresponding data request packet (Interest) was previously delivered. Therefore, the router forwards the response packet to the recorded interface so that it can be delivered to the user. At this time, the used PIT entry is deleted. If an entry with the same content name does not exist when searching for PIT, the delivered Response packet may be discarded. In addition, the router may separately store a copy of the content in the router's CS while the content is being delivered, so that when a request message for the same content arrives later, service can be provided through the CS.

On the other hand, in the above-described basic NDN method, interest can access only caches existing on a path toward a content server, which can often cause path inefficiency.

In relation to this, FIG. 1 is a diagram for explaining the disadvantages of a conventional Information Centric Network (ICN)-NDN.

Referring to FIG. 1 , assuming that user A receives a content service from a server, copies of the content may be stored in CSs of R2, R4, R5, R7, and R9, respectively. Then, when user B requests the same content, the interest message from user B passes through the routers of R6, R3, R1, and R2 constituting the shortest path, and user B receives the content service from R2's CS. .

What should be noted here is that although the content is stored in the CS of R7, as the interest inevitably moves along the routing path, user B receives a service from a more distant node R2.

To solve this problem, two types of methods, proactive and reactive, have been proposed in previous papers.

Proactive methods (Y. Wang et al, J. M. Wang et al) enable a router to utilize a cache other than a routing path by periodically exchanging content information stored in its cache with neighboring nodes. This method has been reported to have problems that information in the cache may be outdated due to 1) overhead occurring in the process of exchanging the entire information of the cache and 2) volatility of the cache.

Reactive methods (R. Chiocchetti et al, E. J. Rosensweig et al) allow routers to find content on neighboring nodes by copying the interest and sending it to additional interfaces. In the case of this method, as request messages are duplicated, redundant transmission of data may be unnecessary, and as a result, network bandwidth may be wasted. In order to solve this problem, a method of introducing a 'probe' packet to check the existence of content and transmitting interest in a direction with a high possibility of content existence has been proposed (I.V. Bastos et al), but data redundant transmission is inevitable in this method, and probe There is a disadvantage that unnecessary data transmission and calculation occur depending on the accuracy of the packet result.

In order to solve the above-mentioned problems of the prior art, the present disclosure provides a novel proactive-based method capable of effectively accessing content stored on an off-path.

Specifically, the present disclosure provides a control method of a node that notifies neighboring nodes individually whenever content is stored in a cache (called heartbeating), unlike existing proactive-based methods that periodically exchange all cache information.

The objects of the present disclosure are not limited to the above-mentioned objects, and other objects and advantages of the present disclosure not mentioned above can be understood by the following description and will be more clearly understood by the embodiments of the present disclosure. Further, it will be readily apparent that the objects and advantages of the present disclosure may be realized by means of the instrumentalities and combinations indicated in the claims.

A control method of a first node according to an embodiment of the present disclosure includes the steps of receiving and storing at least one piece of content, each time the content is stored, a neighboring node located within a certain number of hops from the first node Transmitting a virtual interest including information about the stored content to everyone, receiving an interest requesting content matching the virtual interest from a second node corresponding to a neighboring node that has received the virtual interest, and transmitting content matching the virtual interest to the second node.

The virtual interest includes information about a packet type indicating the virtual interest, and information about a period (ε) during which the information about the virtual interest is to be temporarily stored on a Forwarding Information Base (FIB) of the second node. can do.

In this case, in the control method of the first node, when new second content is received while the first content matching the virtual interest is stored in the first node, the first content is stored in the first node and selecting content to be stored in the first node from among the first content and the second content by comparing the existing time (A) with the period (ε).

Also, the control method of the first node may include setting the period ε according to an estimated value R for a time at which content is stored in the first node.

In this case, the control method of the first node may further include periodically updating the estimated value R according to the following equation. R = αR + (1-α)R _t. Here, R _t is the minimum residence time of contents stored in the first node during the latest period, and α is a constant in the range of (0, 1).

In addition, the control method of the first node may include, when new second content is received while the first content matching the virtual interest is stored in the first node, the first content is stored in the first node It may include comparing the time (A) and the estimated value (R) and updating the estimated value (R).

Meanwhile, the step of transmitting the virtual interest is a value obtained by dividing the number of interests received from neighboring nodes by the number of virtual interests transmitted by the first node in order to request content matching the virtual interest transmitted by the first node. (B) may be monitored, a transmission probability may be set according to the monitored value (B), and the virtual interest may be transmitted to neighboring nodes at a frequency matching the set transmission probability.

In addition, the transmitting of the virtual interest may include monitoring a time interval between interests received from neighboring nodes in order to request content matching the virtual interest transmitted by the first node, and a period corresponding to the monitored time interval. Each time, contents received and stored by the first node may be identified, and one virtual interest including the identified contents may be transmitted to neighboring nodes.

In addition, the transmitting of the virtual interest may include the virtual interest based on the number of interests pending on average in the PIT (Pending Interest Table) of the first node and the number of contents serviced in the cache of the first node. A transmission frequency may be set, and the virtual interest may be transmitted to neighboring nodes according to the set transmission frequency.

According to an embodiment of the present disclosure, a method for controlling a system including a first node and a second node includes receiving and storing, by the first node, at least one piece of content; the first node, storing the content; whenever it is possible, transmitting a virtual interest including information about the stored content to all neighboring nodes located within a certain number of hops from the first node, and receiving the virtual interest by the second node. , the second node temporarily storing information on content matching the virtual interest on a Forwarding Information Base (FIB) of the second node, the interest requesting at least one content is transmitted to the second node If received, comparing, by the second node, the information on the content stored on the FIB with information on the content requested according to the interest, the information on the content stored on the FIB for the requested content information, the second node transmitting the interest to the first node, the first node transmitting content matching the interest received from the second node to the second node Include steps.

The node/system control method according to the present disclosure improves the quality of content service by quickly sharing content stored in a short distance on an off-path.

The control method of a node/system according to the present disclosure performs heartbeating that transmits virtual interest including content information whenever content is stored, so that information on stored content is efficiently delivered through minimal traffic. .

1 is a diagram for explaining the disadvantages of a conventional Information Centric Network (ICN);
2 is a diagram for explaining a node that communicates with a plurality of external nodes according to an embodiment of the present disclosure;
3 is a flowchart for explaining the operation of a node according to an embodiment of the present disclosure;
4 is a diagram for explaining a configuration of virtual interest generated by a node according to an embodiment of the present disclosure;
5A is a diagram for explaining an operation in which a node updates a transmission probability of virtual interest according to the number of transmissions of virtual interest and the number of receptions of interest according to an embodiment of the present disclosure;
FIG. 5B is a diagram for explaining an operation in which a node bundles and transmits interests in a plurality of contents based on a reception interval of interest matching a temporarily generated Forwarding Information Base (FIB) entry according to an embodiment of the present disclosure. ,
6 is a sequence diagram for explaining the operation of a plurality of nodes according to an embodiment of the present disclosure;
7 is graphs for comparing cache hit ratios and average number of hops between a system including a node according to an embodiment of the present disclosure and a system according to the prior art;
8 are graphs for comparing network overhead of a system including a node according to an embodiment of the present disclosure and systems according to the prior art;
9 shows an example of a Rocketfuel topology used for performance evaluation of a large-scale network;
10 is graphs for comparing cache hit rates and average hop counts of a system including a node according to an embodiment of the present disclosure and a system according to the prior art, respectively, in a Rocketfuel topology; and
11 is graphs for comparing network overhead of a system including a node according to an embodiment of the present disclosure and systems according to the prior art in a Rocketfuel topology.

Prior to a detailed description of the present disclosure, the method of describing the present specification and drawings will be described.

First, terms used in the present specification and claims are general terms in consideration of functions in various embodiments of the present disclosure. However, these terms may vary depending on the intention of a technician working in the art, legal or technical interpretation, and the emergence of new technologies. In addition, some terms are arbitrarily selected by the applicant. These terms may be interpreted as the meanings defined in this specification, and if there is no specific term definition, they may be interpreted based on the overall content of this specification and common technical knowledge in the art.

In addition, the same reference numerals or numerals in each drawing attached to this specification indicate parts or components that perform substantially the same function. For convenience of description and understanding, the same reference numerals or symbols are used in different embodiments. That is, even if all components having the same reference numerals are shown in a plurality of drawings, the plurality of drawings do not mean one embodiment.

Also, in the present specification and claims, terms including ordinal numbers such as “first” and “second” may be used to distinguish between elements. These ordinal numbers are used to distinguish the same or similar components from each other, and the meaning of the term should not be construed as being limited due to the use of these ordinal numbers. For example, the order of use or arrangement of elements associated with such ordinal numbers should not be limited by the number. If necessary, each ordinal number may be used interchangeably.

In this specification, singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "comprise" or "consist of" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other It should be understood that the presence or addition of features, numbers, steps, operations, components, parts, or combinations thereof is not precluded.

In the embodiments of the present disclosure, terms such as “module,” “unit,” and “part” are terms used to refer to components that perform at least one function or operation, and these components are hardware or software. It may be implemented or implemented as a combination of hardware and software. In addition, a plurality of "modules", "units", "parts", etc. are integrated into at least one module or chip, except for cases where each of them needs to be implemented with separate specific hardware, so that at least one processor can be implemented as

In addition, in an embodiment of the present disclosure, when a part is said to be connected to another part, this includes not only a direct connection but also an indirect connection through another medium. In addition, the meaning that a certain part includes a certain component means that it may further include other components without excluding other components unless otherwise stated.

2 is a diagram for explaining a node that communicates with a plurality of external nodes according to an embodiment of the present disclosure.

Referring to FIG. 2 , a node 100 according to an embodiment of the present disclosure performs direct/indirect communication with one or more external nodes 200-1, 2, 3, 4, 5, … within the NDN system. can do.

In addition, the node 100 may communicate with various user terminals and one or more content servers providing content. The node 100 may perform direct communication with a user terminal, content server, etc., or may perform communication with a user terminal, content server, etc. via one or more external nodes.

The node 100 may store content received from the outside on a cache of the node 100 . The node 100 may provide stored content to an external node as a result of receiving an interest in content from at least one external node.

3 is a flowchart illustrating an operation of a node according to an embodiment of the present disclosure. 3, it is assumed that the node 100 is a cache node for storing content.

Referring to FIG. 3 , the node 100 may receive and store at least one piece of content (S310).

At this time, the content may be content received by the node 100 through one or more nodes by at least one interest. Also, the node 100 may store the received content on a cache of the node 100 .

As such, whenever content is stored, the node 100 may transmit virtual interest including information about the stored content to all neighboring nodes located within a certain number of hops from the node 100 (S320). That is, the node 100 may inform neighboring nodes that the content is stored on the cache through virtual interest, and in this case, the virtual interest may be transmitted only once when the content is stored.

The above-described constant number of hops corresponds to a criterion for a distance between nodes for transmitting virtual interest, and may be 1 hop or 2 hops, but is not limited thereto.

Neighboring nodes receiving the virtual interest may create a temporary entry related to the virtual interest and content in a Forwarding Information Base (FIB).

Thereafter, interest in the corresponding content can be generated by at least one user (user terminal), and when the generated interest arrives at at least one of the neighboring nodes (receiving the virtual interest), interest is generated by this temporary FIB entry. can be routed toward the node 100 located on the off-path.

As a result, the node 100 may receive interest in the corresponding content from an external node corresponding to a neighboring node that previously received virtual interest (S330).

Then, the node 100 may transmit the corresponding content to an external node (S340).

Since this method does not generate a copy of interest, it has an effect of preventing duplicate content transmission in the existing reactive technique. In addition, in the existing proactive techniques, access information for each content is transmitted several times as the entire cache information is periodically exchanged, whereas in the node control method according to the present disclosure, access information for each content is stored in the cache once. Since they are exchanged, overhead due to redundant transmission of access information can be avoided. Nonetheless, problems (request retransmission and service delay increase) due to outdated access information may be prevented by individually adjusting the lifetime of the temporary FIB.

4 is a diagram for explaining a format of virtual interest generated by a node according to an embodiment of the present disclosure.

Referring to FIG. 4, the format of the virtual interest packet is basically the same as that of the existing interest packet, but the PacketType (0x02) representing virtual interest is separately defined, and the interestLifeTime field contains the lifetime of the FIB entry that is created temporarily. The difference is in how they are transmitted.

For example, when content k is stored in the cache of node i, a virtual interest in content k may be created.

Here, the lifetime (ε _k ) of the temporary FIB entry may be set to “R _i , which is an estimated value of the time the content stays in the cache of node i”. The value can be calculated while being periodically updated according to the equation below.

Here, R _i,t denotes the minimum residence time of contents stored in the cache of node i during the latest period, and α denotes a constant in the range of (0, 1).

The lifetime (ε _k ) of a temporary FIB entry must be less than or equal to the time the content stays in the actual cache. Otherwise, when interest is routed by the temporary FIB entry and visits node i, the corresponding content may not exist in node i's cache, and retransmission may occur as a result.

To avoid this problem, in the control method according to the present disclosure, before the node puts the newly arrived content into the cache (that is, before removing the existing content from the cache), the ε of the content (eg, content k) to be kicked out of the cache We can compare _k and the measure of how long content k stays in the actual cache (before new content arrives) (A _k ).

If A _k is less than ε _k , the node may keep the existing content k in the cache (ie, not delete the existing content from the cache) without putting the newly arrived content into the cache. As such, even if newly arrived content is temporarily not put into the cache, there may not be a big difference in caching performance. This is because it is not known which will be more popular, the newly arrived content or the content that will be evicted from the cache.

In addition, the aforementioned A _k is also compared with the current R _i , and if A _k is smaller than R _i , the node immediately updates R _i to A _k to reflect the latest time the content stays in the cache to R _i immediately. can

Algorithm 1 below is intended to explain the operation of a node (caching node) that generates virtual interest and transfers it to neighboring nodes.

Algorithm 2 below is for explaining the operation of neighboring nodes receiving virtual interest.

Referring to Algorithm 2, a node (a neighboring node) receiving a virtual interest may create a temporary entry on the FIB according to the virtual interest. Then, when an interest in at least one piece of content is received, it may be identified whether the corresponding content is stored in a cache in the node.

If the content is not stored in the cache, the node may compare information about the content stored on the FIB with information about content matching the interest. If the information about the content stored on the FIB matches the information about the content that matches the interest, the node (: neighboring node) may forward the interest to the node (: cache node) that previously transmitted the virtual interest. .

Meanwhile, as described above, an operation in which the node (cache node 100) transmits virtual interest to neighboring nodes once whenever content is stored may be defined as heartbeating.

In general, since it is most effective in terms of cost-benefit to notify the nodes of the two-hop range of content information stored in the cache of the node 100, the node 100 according to an embodiment of the present disclosure is a virtual interest can be delivered up to the two-hop range. To this end, the value of the “Scope” field of FIG. 4 may be set to 2, and the corresponding value may be decreased by 1 each time forwarding is performed to the next hop. When the value becomes 0, forwarding can be stopped.

The frequency of heartbeating described above may be set according to various embodiments.

In the basic heartbeating technique, whenever content is stored in the cache, virtual interest is generated and transmitted to neighboring nodes.

However, at least one of the two embodiments (stochastic heartbeating and delayed heartbeating) may be utilized in order to minimize the cost incurred while generating virtual interest each time.

As a first embodiment related to the frequency of heartbeating, a stochastic heartbeating technique may be used.

Because more popular content is usually stored in caches near network edges, heartbeating gains may differ between network edges and cores. In other words, considering the fact that the gain of heartbeating varies depending on the location of the cache, the heartbeating frequency (a kind of cost) is intended to be varied.

In this regard, referring to FIG. 5A , the node 100 may update the transmission probability of virtual interest according to the number of transmissions of virtual interest and the number of receptions of interest.

In this case, the number of receptions of interest may mean the number of times interest based on an FIB entry generated according to the virtual interest transmitted by the node 100 is received by the node 100.

And, the virtual interest transmission probability means the probability that the node 100 actually transmits the virtual interest to neighboring nodes as content is stored.

Specifically, referring to Equation 2, as a result of monitoring the number of virtual interests transmitted for each face and the number of interests serviced by the FIB entry generated by the virtual interest thus transmitted, the B value can be periodically calculated. there is.

In addition, the transmission probability (p) for transmitting virtual interest may be set differently according to the change in B value. For example, the B value for the face f may be referred to as B _f and the transmission probability may be referred to as P _f .

In this case, whenever m number of virtual interests are transmitted, B _f may be updated as shown in Equation 3 below. Here, B _f,i corresponds to a value B calculated as a value observed in the i-th period (number of interest served/number of virtual interest transmitted). And κ is a coefficient of a moving average and may be a constant in the range of (0,1).

one example. The initial transmission probability value may be set to 1. That is, P _f = 1, and the basic probability change amount can be set to reduce by g (0 < g < 0.1).

If the value of B _f,i+1 is greater than or equal to B _f,i (if the cost benefit is increasing), the node applies the existing probability change value (ΔP _f,i ) as it is (ΔP _{f ,i+1} = ΔP _f,i ) P _f is updated, otherwise - ΔP _f,i can be applied to (ΔP _f,i+1 = -ΔP _f,i ) P _f can be updated. After all, this method has the effect of maximizing the cost-benefit by sending virtual interest less frequently to nodes that do not sufficiently use the temporary FIB entries created as a result of heartbeating.

As a second embodiment related to the frequency of heartbeating, a delayed heartbeating technique may be used.

In relation to this, FIG. 5B shows the time at which virtual interest is generated/transmitted on a node (cache node) in the heartbeating process (lower side) and the time at which interests arrive from neighboring nodes that receive virtual interest to the node (cache node) (upper side). ).

For example, in order to reduce the number of generated/transmitted virtual interests while delivering information about stored contents in a timely manner, a technique of aggregating heartbeating to occur between two interests that arrive consecutively may be used (FIG. 5B). safe aggregation).

Specifically, safe aggregation is a technique in which a period is set according to an interval at which interest requesting content (including interest according to virtual interest) is received, and virtual interests are bundled and transmitted at once within the set period.

However, there is a disadvantage of safe aggregation in that it is not easy to know the safe aggregation interval in the node transmitting the virtual interest, and the gain due to aggregation may be insignificant because the safe aggregation interval is not large enough.

Therefore, the node/system control method according to an embodiment of the present disclosure may utilize a relaxed aggregation technique instead of safe aggregation.

Relaxed aggregation is a technique for aggregating heartbeating that occurs between two consecutive interests in which content access is actually made by an FIB entry generated by a virtual interest among interests arriving at neighboring nodes.

The relaxed aggregation interval has an advantage in that the node (cache node) transmitting the virtual interest can grasp through received interests without exchanging a separate message. In addition, since heartbeating can be performed in longer time units in the relaxed aggregation interval compared to the safe aggregation interval, benefits due to aggregation may be greater.

When the relaxed aggregation interval is utilized, the node (cache node) may update the interval by monitoring in real time the interval between interests received by the FIB entry.

However, when the corresponding interval is initially set, the initial interval may be set according to user input.

Alternatively, the initial interval may be set according to an average value of intervals of interest (interest received according to an FIB entry) received during the warmup time. During the warmup time, aggregation is not performed and virtual interest may be output individually whenever each content is stored. Alternatively, during the warm-up time, heartbeating may be performed in a manner of outputting virtual interest by bundling information on stored contents at predetermined time intervals set arbitrarily.

The two techniques described above with reference to FIGS. 5A and 5B are not incompatible with each other and can be independently adopted.

Meanwhile, in relation to heartbeating, adaptive heartbeating in various other ways can be utilized.

For example, in terms of load balancing, if the number of packets processed by a router (node) (ex. the average number of pending interest on the PIT + the number of contents served from the cache) exceeds a certain threshold, the node and system reduce the number of heartbeats. It can reduce the amount of interest delivered from off-path nodes.

In addition, the node and the system are used for content with high content importance (ex. content where a content provider pays additionally for a special purpose and wants a differentiated caching service, or where the distance to the content server is relatively far or greater than a critical distance). A differentiated caching service can be provided by selectively performing heartbeating only for contents).

Meanwhile, FIG. 6 is a sequence diagram illustrating operations of a plurality of nodes constituting an NDN system according to an embodiment of the present disclosure.

FIG. 6 assumes a situation in which node 100 corresponds to a cache node and node 200 is a neighboring node receiving virtual interest from node 100 . However, it goes without saying that the node 200 may also serve as a cache node, and conversely, the node 100 may also serve as a neighboring node based on the node 200.

Referring to FIG. 6 , the node 100 may receive and store at least one piece of content in a cache (S610 and S620). Then, whenever the content is stored, the node 100 may transmit virtual interest including information about the stored content to all neighboring nodes located within a certain number of hops from the first node (S630).

In this case, the node 100 may transmit one virtual interest including information on received contents for each of the above-described relaxed aggregation intervals, but is not limited thereto.

When the virtual interest transmitted by the node 100 is received, the node 200 may create a temporary entry related to the virtual interest on the FIB (S640). Specifically, the node 200 may temporarily store information on content matching virtual interest on the FIB.

Thereafter, when interest requesting at least one content is received from an external node (S650), the node 200 may compare information about content stored on the FIB with information about content requested according to interest.

Specifically, the node 200 may first compare information about content stored in the cache of the node 200 with information about content requested according to interest. And, if the content requested according to the interest is not stored in the cache of the node 200, the node 200 may perform a search for the FIB.

Here, when the information on the content (temporary entry) stored on the FIB of the node 200 matches the information on the requested content (S660), the node 200 may forward the interest to the node 100 (S670).

Then, the node 100 may transmit content (stored in the cache) matching the interest to the node 200 (S680). As a result, content requested according to interest may be delivered to the user terminal via the node 200 or the like.

Meanwhile, performance comparison of the node/system control method (HEARTBEAT) according to the present disclosure with Inform (R. Chiocchetti et al) and D-FIB (O. Ascigil et al), which are representative conventional off-path content access techniques. evaluation was performed. In this performance comparison evaluation, both stochastic heartbeating and delayed heartbeating were used.

First, an experimental environment targeting a small network (eg, FIG. 1) is as follows. One server provides the contents requested by users, and the total number of contents is 1,000,000. 3 consumer applications (a total of 9) are requesting 20 contents per second for each of the user terminals, user A, B, and C, and the popularity of the requested contents follows the zipf-Mandelbrot distribution. Each router has an LRU (Least Recently Used) cache installed, and performance evaluation was conducted while changing the cache size from 0.1% to 0.5% of the total content. The bandwidth of all links is 100Mbps and the propagation delay is set to 5ms. In the node/system control method according to the present disclosure, the heartbeating range was set to 2 hops, and the probability change g value was set to 0.01. α and κ for the moving average were applied as 0.5. The B value is set to be updated whenever the number of virtual interests corresponding to (cache size) × 0.1 is transmitted. For performance comparison evaluation, the total simulation progress time is 3000 seconds, and the results for 2000 seconds after the first 1000 seconds are provided as a graph.

In relation to this, FIG. 7 is graphs for comparing cache hit rates and average hop counts of a system including a node (HEARTBEAT) according to an embodiment of the present disclosure and systems (DEFAULT, INFORM, DFIB) according to the prior art, respectively. . The average number of hops represents a distance to a point where content is serviced.

As shown in the left graph of FIG. 7 , the highest cache hit rate is shown in the node/system control method according to the present disclosure regardless of the size of the cache. In addition, when the average number of hops to accessed content is based on the performance improvement of Inform, it can be confirmed that the control method according to the present disclosure shows a performance improvement of about 3 times. That is, it is confirmed that more content can be serviced closer through the node/system control method according to the present disclosure.

8 are graphs for comparing network overhead between a system including a node according to an embodiment of the present disclosure and systems according to the prior art. Specifically, FIG. 8 shows graphs indicating how many hops each of interest and response messages are delivered to provide service.

Referring to FIG. 8 , despite the performance improvement as shown in FIG. 7 , it is confirmed that the number of forwarding of interest and response in the network is the smallest in the control method according to the present disclosure. As a result, the node/system control method according to the present disclosure enables the most effective use of the cache while generating the minimum overhead, thereby minimizing the delay time required for service provision and enabling efficient use of network bandwidth at the same time It is confirmed that

Meanwhile, performance comparison evaluation targeting large-scale networks was also performed.

9 is a diagram showing an example of a Rocketfuel topology used for performance evaluation of a large-scale network. A total of 4 servers, one for each producer-side edge router, provide unique content, and each consumer-side edge router is configured to request content from 5 consumer applications (total of 65 consumers). Other simulation parameters, including the total number of contents and the popularity distribution of contents, are the same as the performance evaluation for the small network described above. The bandwidth of all links is 100 Mbps, and the propagation delay is set to 10 ms for the network core and 5 ms for the network edge.

10 is a comparison of the cache hit rate and average number of hops of a system (HEARTBEAT) including a node according to an embodiment of the present disclosure and systems (DEFAULT, INFORM, DFIB) according to the prior art, respectively, in a Rocketfuel topology. graphs for

Referring to FIG. 10 , similar to the performance comparison evaluation of the small network described above, the control method (HEARTBEAT) according to the present disclosure produces the largest cache hit rate and provides content service at the nearest location.

As the size of the topology increases, the number of caches increases, and the control method according to the present disclosure can more effectively use the increased caches, and as a result, it is confirmed that the performance difference with the comparison group is larger.

11 is graphs for comparing network overhead of a system including a node according to an embodiment of the present disclosure and systems according to the prior art in a Rocketfuel topology. Specifically, FIG. 11 is graphs showing how many hops interest and response are delivered. Similar to the above-described simulation results for the small network, it is confirmed that the control method (HEARTBEAT) according to the present disclosure generates a small amount of overhead compared to comparable techniques.

However, in the control method (HEARTBEAT) according to the present disclosure, it can be seen that the amount of interest transmission increases as the cache size increases, which means that as the cache size increases, the number of interests that can be served in the cache increases, and as a result This phenomenon occurs when the relaxed aggregation interval becomes smaller.

In the case of a 0.5% cache, the interest transmission amount of the control method (HEARTBEAT) according to the present disclosure is larger than that of the DFIB technique, but the response transmission amount is still the smallest in the control method (HEARTBEAT) according to the present disclosure. It is confirmed.

In general, considering that the size of the response packet is 30 to 100 times larger than the size of the interest packet, it was confirmed that the control method (HEARTBEAT) according to the present disclosure shows the best performance while generating the least amount of traffic. .

Meanwhile, various embodiments described above may be implemented in a recording medium readable by a computer or a similar device using software, hardware, or a combination thereof.

According to the hardware implementation, the embodiments described in this disclosure are application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), and field programmable gate arrays (FPGAs). ), processors, controllers, micro-controllers, microprocessors, and electrical units for performing other functions.

In some cases, the embodiments described herein may be implemented by a processor itself. According to software implementation, embodiments such as procedures and functions described in this specification may be implemented as separate software modules. Each of the software modules described above may perform one or more functions and operations described herein.

On the other hand, computer instructions or computer programs for performing processing operations in the electronic device 100 according to various embodiments of the present disclosure described above are provided on a non-transitory computer-readable medium can be stored in Computer instructions or computer programs stored in such a non-transitory computer readable medium, when executed by a processor of a specific device, cause the above-described specific device to perform processing operations in the node/electronic device according to various embodiments described above.

A non-transitory computer readable medium is a medium that stores data semi-permanently and is readable by a device, not a medium that stores data for a short moment, such as a register, cache, or memory. Specific examples of the non-transitory computer readable media may include CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM, and the like.

Although the preferred embodiments of the present disclosure have been shown and described above, the present disclosure is not limited to the specific embodiments described above, and is common in the technical field belonging to the present disclosure without departing from the gist of the present disclosure claimed in the claims. Of course, various modifications and implementations are possible by those with knowledge of, and these modifications should not be individually understood from the technical spirit or perspective of the present disclosure.

100: node (cache node) 200: node (neighboring node)

Claims (11)

In the control method of the first node,
receiving and storing at least one piece of content;
whenever content is stored, transmitting a virtual interest including information about the stored content to all neighboring nodes located within a certain number of hops from the first node;
receiving an interest requesting content matching the virtual interest from a second node corresponding to a neighboring node that has received the virtual interest; and
Transmitting the content matching the virtual interest to the second node; including, the control method of the first node. According to claim 1,
The virtual interest is,
A first node, including information on a packet type indicating the virtual interest, and information on a period (ε) during which the information on the virtual interest will be temporarily stored on a Forwarding Information Base (FIB) of the second node. control method. According to claim 2,
The control method of the first node,
When new second content is received in a state where the first content matching the virtual interest is stored in the first node, the time (A) that the first content was stored in the first node and the period (ε) Comparing, selecting the content to be stored in the first node from among the first content and the second content; including, the control method of the first node. According to claim 2,
The control method of the first node,
and setting the period (ε) according to an estimated value (R) for a time during which content is stored in the first node. According to claim 4,
The control method of the first node,
Further comprising: periodically updating the estimated value (R) according to the following equation,
R = αR + (1-α)R _t
R _t is the minimum residence time of the contents stored in the first node during the latest period,
Wherein α is a constant in the range of (0, 1), the control method of the first node. According to claim 4,
The control method of the first node,
When new second content is received while the first content matching the virtual interest is stored in the first node, the time (A) when the first content was stored in the first node and the estimated value (R) ), and updating the estimated value (R); including, the control method of the first node. According to claim 1,
The step of transmitting the virtual interest,
Monitoring a value (B) obtained by dividing the number of interest received from neighboring nodes by the number of virtual interests transmitted by the first node in order to request content matching the virtual interest transmitted by the first node;
Set the transmission probability according to the monitored value (B),
A control method of a first node for transmitting the virtual interest to neighboring nodes at a frequency matching the set transmission probability. According to claim 1,
The step of transmitting the virtual interest,
Monitoring a time interval between interests received from neighboring nodes in order to request content matching the virtual interest transmitted by the first node;
For each period corresponding to the monitored time interval, identifying contents received and stored by the first node and transmitting one virtual interest including the identified contents to neighboring nodes. According to claim 1,
The step of transmitting the virtual interest,
Setting a transmission frequency of virtual interest based on the number of interests pending on average in the PIT (Pending Interest Table) of the first node and the number of contents serviced in the cache of the first node;
A control method of a first node for transmitting the virtual interest to neighboring nodes according to the set transmission frequency. In the control method of a system including a first node and a second node,
receiving and storing, by the first node, at least one piece of content;
transmitting, by the first node, virtual interest including information about the stored content to all neighboring nodes located within a certain number of hops from the first node, whenever content is stored;
receiving, by the second node, the virtual interest;
temporarily storing, by the second node, information on content matching the virtual interest on a Forwarding Information Base (FIB) of the second node;
When an interest requesting at least one piece of content is received by the second node, comparing, by the second node, information about the content stored in the FIB with information about the content requested according to the interest;
transmitting, by the second node, the interest to the first node when the information on the content stored on the FIB matches the information on the requested content; and
Transmitting, by the first node, content matching the interest received from the second node to the second node; including, the system control method. In the computer readable recording medium,
A computer-readable recording medium having stored thereon; at least one instruction that is executed by a processor of an electronic device and causes the electronic device to perform the control method of claim 1.

Images