EP3563558A1

EP3563558A1 - Computer network of nodes communicating with one another by peer-to-peer messages and associated method for interconnecting between nodes

Info

Publication number: EP3563558A1
Application number: EP17838185.1A
Authority: EP
Inventors: Adrien PLAGNOL; Chary TITOUAN
Original assignee: Worldline SA
Current assignee: Worldline SA
Priority date: 2016-12-27
Filing date: 2017-12-27
Publication date: 2019-11-06
Also published as: CN110431824A; WO2018122533A1; FR3061391A1; CN110431824B; FR3061391B1

Abstract

The invention concerns a computer network of nodes (1, 2) communicating with one another by peer-to-peer messages, message-producer nodes (1) sending their produced messages to message-consumer nodes (2), a consumer node (2) also being able to be a producer node (1) at another level of the network (6), characterised in that each message-producer node (1): chooses, among the nodes (1, 2) registered as active in the network, its consumer nodes (2) in order to interconnect with them; keeps up-to-date the list of its interconnected consumer nodes (2); manages the load distribution of its produced messages to its interconnected consumer nodes (2), according to the processing time, by its interconnected consumer nodes (2), of its produced messages, by sending its produced messages primarily to its interconnected consumer nodes (2) having low processing times; and, modifies its list of interconnected consumer nodes (2) according to the processing time, by its interconnected consumer nodes (2), of its produced messages, by deleting from its list, at least temporarily, those of its interconnected consumer nodes (2) which are too slow.

Description

COMPUTER NETWORK OF NODES COMMUNICATING BETWEEN THEM AND PAIR MESSAGES THEREFOR

INTERCONNECTION BETWEEN ASSOCIATED NODES

FIELD OF THE INVENTION

The invention relates to the field of node networks communicating with each other by peer-to-peer messages as well as the domain of interconnection processes between nodes in a network of nodes communicating with each other by peer-to-peer messages.

The network of nodes considered is for example an architecture of micro-services, respectively distributed on potentially several machines, having for example not the same addresses between them, not being for example not all on the same network, not presenting between them for example not the same computing capacity. In addition, each micro-service can have a potentially undefined life time. The heterogeneous nature of this micro-services architecture makes it more fragile compared to imponderable events, of the localized fault type, and therefore difficult to manage.

BACKGROUND OF THE INVENTION

The network of nodes, so that its nodes communicate with each other, goes initially in an initial phase to interconnect the nodes between them, then goes in a phase of update during the life of the network to maintain and to adapt the interconnections between the nodes of the network.

For example, the message distribution system will communicate micro services between them through a message passing system. For this, it can use an implementation centered on a broker of messages ("broker" in English language). According to a first prior art, there is known a network of nodes exchanging messages in peer-to-peer ("peer to peer") and an associated interconnection method, which are both managed by a central interconnection device . A disadvantage of this first prior art is that this central interconnection device is a weak link in the network, a single point of failure (SPF), including a failure, even temporary, will disrupt the operation of the network and significantly degrade its performance of transmission efficiency and message processing.

For example, the message broker-based system can be centralized (such as "RabbitMQ" or "ActiveMQ"), but it has a significant limitation because this centralized system will be a single point of failure that can disrupt the operation. network and degrade its capabilities.

According to a second prior art, it is known a network of nodes exchanging messages in peer-to-peer ("peer to peer" in English) and an associated interconnection method, which are both fully distributed. A disadvantage of this second prior art is that the robustness related to this fully distributed character, in particular in a network of nodes exchanging peer-to-peer messages, becomes a weakness when the size of the messages exchanged within the network becomes important. because a large volume of fully distributed data exchange, in a type of peer-to-peer communication, will tend to saturate the network and therefore also degrade its performance in transmission efficiency and message processing.

For example, the message broker-based system can be distributed (as for example "Kafka"), but it has an important limitation, because this distributed system will become operationally limited as soon as the size and the volume of the messages that he will have to pass will become too important. SUMMARY OF THE INVENTION

The object of the present invention is to provide a network of nodes and / or an interconnection method at least partially overcoming the aforementioned drawbacks.

More particularly, the object of the invention is to provide a network of nodes and / or an interconnection method, for which a good compromise between robustness of operation of the network on the one hand and efficiency of operation of this network on the other hand is realized. .

Indeed, on the one hand the robustness of operation of the network will allow the network of nodes to continue to operate by allowing the nodes of this network to continue to exchange messages with each other even if a given component, for example a central component, breaks down, at least in the case of a temporary breakdown, the most frequent case.

On the other hand, the operating efficiency of this network will allow the network node to continue to optimize the transmission and processing time of the messages exchanged, even if a given component, for example a central component, fails, at least for the case of a temporary breakdown, the most frequent case.

In order to achieve this compromise between robustness of operation and efficiency of network operation, the invention proposes a network in which the producing nodes, on the one hand each manage their list of consumer nodes, thus guaranteeing their autonomy with respect to a possible point of failure in the network, and on the other hand each optimize their list of consumer nodes in terms of processing time of their messages produced, by deleting, at least temporarily from their list, consumer nodes that are too slow to process the messages. received messages from the producer node corresponding to this list.

To this end, the present invention proposes a computer network of nodes communicating with each other by peer-to-peer messages, message producing nodes sending their product messages to message consumer nodes, a consumer node that can be also a producer node at another level of the network, characterized in that each message producing node: chooses, among the nodes registered as active in the network, its consumer nodes to interconnect with them, maintains the list of its nodes interconnected consumer nodes, manages the load balancing of its product messages to its interconnected consumer nodes, as a function of processing time, by its interconnected consumer nodes, of its product messages, by sending its product messages especially to its more interconnected consumer nodes. low processing time, modifies its list of interconnected consumer nodes according to the processing time, by its interconnected consumer nodes, its messages produced, removing from its list at least temporarily its interconnected consumer nodes too slow.

To this end, the present invention also proposes a method of interconnection and communication between nodes in a computer network according to the invention.

To this end, the present invention also proposes a method of interconnection between nodes in a computer network of nodes communicating with each other by peer-to-peer messages, message producing nodes sending their generated messages to message consumer nodes, a node a consumer which can also be a producer node at another level of the network, characterized in that each message producing node: chooses, from among the nodes registered as active in the network, its consumer nodes to interconnect with them, keeps up to date the list of its interconnected consumer nodes, modifies its list of interconnected consumer nodes according to the processing time, by its consumer nodes, of its generated messages, by deleting from its list at least temporarily its interconnected consumer nodes too slow.

According to preferred embodiments of the invention, the very large nature of the data exchanged and the critical real time character of the processing of these exchanged data make the use of node network according to the invention and the associated interconnection method particularly interesting.

According to preferred embodiments of the invention, the network will use a peer-to-peer message passing system that will be self-adaptive and highly resilient. This system will make it possible to establish a communication between micro-services through a communication channel functionally identical to a data bus, but whose implementation is based on direct communication between the microservices themselves. In this architecture, each micro-service is responsible for maintaining the list of its peers in the network, as well as load distribution when distributing messages to these peers. This system makes it possible to create data flows between micro-services that will be able to withstand even a large scale change, while maintaining a strong resilience in the event of the disappearance of nodes or in the event of communication links being cut off. the network.

According to preferred embodiments, the invention comprises one or more of the following features which can be used separately or in partial combination with one another or in total combination with one another, or with one or the other of the aforementioned objects of the invention. .

Preferably, the nodes register as active in the network with a centralized register in the network, which register diffuses to at least all the active producing nodes of the network, preferably to all the active nodes of the network, the list of at least all the active consumer nodes of the network, preferably the list of all the active nodes of the network. Broadcasting ("broadcast" in English) is a different type of communication from peer-to-peer messaging.

Preferably, said register does not establish for a given particular producer node the list of its interconnected consumer nodes, and said register does not send to any particular producer node the list of its interconnected consumer nodes. An advantage is the fact that, once constituted, the network of nodes becomes autonomous with respect to the registry service without which this network can then operate normally temporarily during a failure of limited duration of this register.

On the other hand, the good functioning of the register at least a part of the time remains useful, insofar as only the register can signal to the different producing nodes the changes occurring in the topology of the network, thus allowing these producing nodes to update each one. their list of consumer nodes, thereby allowing each producing node to operate with a list of consumer nodes that is optimized in terms of efficiency relative to the instant topology of the node network. Alternatively, the management of the evolution of the network topology in a distributed manner at each producing node remains possible, but this would reduce the operating efficiency of the network.

Thus, the distribution at the level of each producer node of the management of the list of its consumer nodes on the one hand associated with the centralized management of the only topology of the network with its permanent update on the other hand realizes an important additional improvement. compromise between robustness of operation and efficiency of network operation.

Preferably, the nodes register periodically as active in the network with the register, as long as they are able to process messages, the recording period preferably being less than or equal to 10 seconds, any active node that is missing at least two recording periods, preferably at least one recording period, are deleted from the list of active nodes by the register. Thus, even if this periodic recording is a little tedious for each consumer node, it allows on the other hand to further improve the compromise between robustness of operation and efficiency of network operation, insofar as it has the effect of temporarily removing any node a deficient consumer, and to suppress it only for approximately the duration of his disability, for the duration of a close registration period. Preferably, each message producing node can modify its list of interconnected consumer nodes by possibly adding a newly registered consumer node if it is able to consume messages produced by said producer node. Thus, each producing node can increase its list of consumer nodes in real time, and thus improve its load management in terms of product messages sent in real time as well.

Preferably, each consumer node sends, at least for some messages received from a producer node, its processing time of these messages or a representative weight of its processing time of these messages, representative of the difference between the arrival of these messages at said consumer node and their departure from said consumer node possibly after transformation, and preferably each producer node adds a processing time evaluation request in all or part of the messages that it sends to each of its interconnected consumer nodes . In this way, each consumer node can update the performance of all consumer nodes in its list and adapt the load management of its messages produced accordingly, just as often as it wants, it is sufficient for it to choose the proportion of his product messages to which he adds a request for evaluation of processing time.

Preferably, each message producing node manages the load balancing of its product messages to its interconnected consumer nodes, based on the processing times or weights sent by its interconnected consumer nodes, sending more product messages to the interconnected consumer nodes. whose treatment times are lower or the weights correspond to lower processing times. Thus each producing node improves the load management of its product messages, thus improving the overall efficiency of transmission of the messages produced and their processing in the network. Preferably, each producer node calculates a sliding average weight over time for each of its interconnected consumer nodes, which it uses to manage the load balancing of its product messages to its interconnected consumer nodes. In this way, the calculated parameter is on the one hand well representative of the processing speed of the messages received by a consumer node, while smoothing point variations not representative of this parameter, reducing the risks of instability and divergence of enslavement according to the variations of this parameter.

Preferably, each producer node loads its interconnected consumer nodes in a weighted random manner, so that the probability for its interconnected consumer nodes to receive a message produced by said producer node depends on a random draw which is weighted by the weights of said nodes. interconnected consumers. This weighted random load management achieves a good compromise between the fair and distributed use of all available consumer nodes, a factor that improves the robustness of network operation by avoiding overloading high-performance consumer nodes to avoid their collapse factor of instability in the network, and secondly the operating efficiency of the network by using priority and more often the fastest consumer nodes.

Preferably, each node is a micro-service that exchanges messages with at least one other node, each micro-service being either a processor that processes data received from one node of the network to then send that processed data to another node of the network, preferably said processing being reduced to a single processing task performed on these data, either a source that produces data to be injected into the network to at least one processor, or a sink that exports the processed data by at least one processor to a third-party system on the network. This type of network is particularly adapted to the use of the invention, in the as its nodes perform a lot of relatively simple operations on a lot of data.

Preferably, at least one of the processors, preferably the majority of the processors, still more preferably all the processors, filters the received data and then routes them to different types of nodes in the network. This is a particularly representative example of simple operations achievements in large numbers on a large volume of data.

Preferably, a type of producer node is a filter which separates the received messages according to a criterion to be filled, to direct them towards three types of consumer nodes respectively according to whether said criterion is fulfilled, or not filled, or perhaps filled. Here again is a simple operation, but to achieve very quickly and a lot of times.

In a first preferred embodiment, the messages are representative of banking transactions, the criterion is the fraudulent nature or not of a banking transaction. The very large character of the exchanged data and the critical real time character of the processing of these exchanged data make the use of the node network according to the invention as well as the associated interconnection method particularly interesting.

In a second preferred embodiment, the messages are representative of images filmed by CCTV cameras, the criterion is the character of presence or absence of delictual risk related to the content of a filmed image or a sequence of images. images filmed by these CCTV cameras. The very large character of the exchanged data and the critical real time character of the processing of these exchanged data make the use of the node network according to the invention as well as the associated interconnection method particularly interesting.

Other features and advantages of the invention will appear on reading the following description of a preferred embodiment of the invention, given by way of example and with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 schematically shows an example of a network part according to one embodiment of the invention.

FIG. 2 diagrammatically represents an example of a network part according to one embodiment of the invention.

Figure 3 schematically shows an example of a network part according to one embodiment of the invention.

FIG. 4 schematically represents an example of a scale of the cumulative weights allowing management of the peas of the consumer nodes in a network according to one embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION FIG. 1 schematically represents an example of a network portion according to one embodiment of the invention.

The network 6 comprises producing nodes 1 and consuming nodes 2. In the network 6 are represented the nodes A, B, C and D. Here, the node A is a producer node 1. The node B is a consumer node 2 at the level of the link between the nodes A and B in the network 6, while the same node B is a producer node 1 at the link between the nodes B and C in the network 6 as well as at the link between the nodes. nodes B and D in the network 6. Node B is a producer node 1 or a consumer node 2, depending on the level of the network 6 considered. On the other hand here, node A is only a producer node

1, while the nodes C and D are only consumer nodes 2.

A register 3 broadcasts to all the nodes A, B, C and D of the network 6, initially the original topology of the network 6, then each update of this topology during the life of the network 6. At the level of the link between the nodes A and B in the network 6, the messages produced by the producer node A go to the consumer node B and only circulate in this direction, to be consumed by the consumer node B. At the level of the link between the nodes B and C in the network 6, the messages produced by the producer node B go to the consumer node C and only circulate in this direction, to be consumed by the consumer node C. At the link between the nodes B and D in the network 6, the messages produced by the producer node B go to the consumer node D and only circulate in this direction, to be consumed by the consumer node D.

Each producer node 1 maintains a list of consumer nodes 2 to which it may have to transmit messages as part of the flow to which it belongs. In the flow of FIG. 1, the node A knows the node B to which it transmits the messages that it produces. Node B knows the nodes C and D to which it transmits the messages it produces, each message being sent to either node C or node D.

This list of consumer nodes 2 is initialized by a request to a register 3 to which each producer node 1 or consumer 2 belongs at its instantiation. This register 3 is only useful when a change occurs in the topology of the nodes. Thus, if the register 3 disappears, for example if it is damaged or if its link with the network 6 becomes inoperative, the operation of the flow represented in FIG. 1 is not affected.

Each producer node 1 or consumer 2 register, at its instantiation, in the register 3. It will send thereafter a new registration request at regular intervals, for example every 5 or 10 seconds, with the register 3. From this way, if the register 3 becomes faulty, this node 1 or 2 will quickly receive the complete topology of the network 6 of nodes during the re-instantiation of this register 3.

FIG. 2 diagrammatically represents an example of a network part according to one embodiment of the invention. The part of the network 6 is first shown at a time T0 and then at a time T1.

At a time T0, the node A is interconnected with the node B and the node B is interconnected with the node C. The node A sends messages towards the node B which sends messages towards the node C.

Then, register 3 broadcasts the information that node D has registered with it. Neither node A nor node C interconnect with node D. On the other hand, the node B is interconnected with the node D. At time T1, the node A is interconnected with the node B and the node B is interconnected with the nodes C and D. The node A sends messages to node B that sends messages to either node C or node D.

The register 3 has the role of transmitting all the events it processes broadcast to all the nodes 1 and 2 of the network 6 that it knows. Thus, when a new node 1 or 2 is declared, all the other nodes 1 or 2 are informed and the producer nodes 1 choose to connect to it depending on whether this new node is part of their stream or not.

In FIG. 2, for example at the time T1, the node D is declared, the register 3 broadcasts the information. Node A does not connect to Node D, Node B connects to Node D because Node D is part of the type of service included in its flow.

Each consumer node 2 maintains the median response time on a sliding window of N message sends, which is called Tmed. When sending a message to a consumer node 2, if this target consumer node 2 has not responded after 2 * Tmed, the request is aborted. This request will be immediately reissued under the same conditions. After a certain number of such re-transmissions, for example after 2 or 3 retransmissions, the sending of the message is considered as failed and the target consumer node 2 is removed from the list of target consumer nodes 2 of the producer node 1 unsuccessfully trying to send him messages. This procedure makes it possible to quickly eliminate the consumer nodes 2 whose response time becomes abnormally high. These deleted consumer nodes 2 will then be able to re-register with the register 3, when they will be operational again and they will again be able to guarantee a reasonable time of processing of the messages received from the producer nodes 1. The register 3 will broadcast this information which will in turn be intercepted by the producer node 1 which is transmitting messages. This mechanism makes it possible to temporarily exclude the nodes consumers 2 failing or suddenly too slow, before reintegrating them into the flow, once they become operational again.

In a micro-service architecture network 6 organized in flow and communicating with each other so that each message producing micro-service manages its own list of micro-services consumers of messages, the loss, at least temporary, of the register 3 has no impact on the operation of the network 6. The register 3 is used to distribute the knowledge of the topology of micro-services, but once it acquired, the network 6 becomes autonomous for its operation, and n ' again needs register 3 only when changing network topology 6.

The network 6 comprises producer nodes 1 and consumer nodes 2. In the network 6 are represented the nodes PI, P2, C1, C2 and C3. Here, the nodes PI and P2 are producing nodes 1, at the level of links considered. Here, the nodes C1, C2 and C3 are consumer nodes 2, at the level of links considered.

A register 3 diffuses to all the nodes PI, P2, C1, C2 and C3 of the network 6, initially the original topology of the network 6, then each update of this topology during the life of the network 6.

The messages produced by the producer node PI go to the consumer nodes C1 and C2, and only circulate in this direction, to be consumed by the consumer nodes C1 and C2. The messages produced by the producer node P2 go to the consumer nodes C1, C2 and C3, and only circulate in this direction, to be consumed by the consumer nodes C1, C2 and C3.

On such an architecture having N producing nodes 1, that is to say sending messages, as well as M consumer nodes 2, that is to say receiving these messages, it is useful to intelligently distribute the workload between the different nodes. Load balancing is a technique used by large networks to balance the solicitation of the servers. The network 6 according to the invention has set up a dynamic, intelligent and adaptive load distribution.

Each producer node 1, for example a producer micro-service 1, has a load distribution module which is grafted at its output interface and which communicates with the input interface of each of the consumer nodes 2 of its list. , for example consumer micro-services 2. Each of the N producer nodes 1 knows its list of consumer nodes 2 among the M consumer nodes 2, all M consumer nodes 2 having at initialization an identical weight, for example a value of 1 .

Every q seconds, for example every 5 seconds, each producing node 1 will place, within some of its messages produced and sent, a header specifying a request for evaluation of the processing time for the message considered. Thus consumer nodes 2 will have weights that are dynamic because they are recalculated regularly.

Each consumer node 2 has a queue of messages to be processed, as well as a processing function that consumes these messages one after the other.

As soon as a message incorporating a processing time request is inserted into the queue, it is labeled its system arrival time, and as soon as it leaves the processing function, it will be labeled its system time of initially, the difference between this system start time and this arrival system time corresponding to the message processing time by this consumer node 2.

As an acknowledgment message, the consumer node 2 responds to the producer node 1 having sent the processed message, preferably returning to it a parameter representative of the difference between its start time and its arrival time, for example a weight, alternatively, the treatment time itself.

Here, from this difference, a weight is generated proportional to the execution speed of the consumer node 2. Let the weight of a consumer node i:

which corresponds to formula (1).

This weight is updated in the consumer node table of the producing node concerned. Each producer node independently harvests the load information of each consumer node. Each producer node therefore maintains its own weight table of consumer nodes, which can therefore be different from that of the other producing nodes.

In order to obtain a precise weight, the invention preferentially uses the notion of sliding average weight, defined by the average of the weights over a time window δ, defining a number of iterations making it possible to calculate a sliding average, given for a node consumer in particular:

If '^ W ^ W ^ Wi ^ Win}

which corresponds to formula (2).

Let the average sliding weight of the consumer i, noted Wigliss, for a set of measurements of size n: and Wigliss = 1 for n = 0

which corresponds to formula (3).

This sliding weight is used to weight the random draw used to select the destination consumer node of the next message.

The weighting is done by comparing the value drawn with a scale on which the consumer nodes are positioned according to their weight.

This scale goes from 0 to the value of the sum of Wigliss. The value of each terminal of the scale is calculated according to the accumulated values of the previous terminals.

With N the number of consumer nodes:

For i from 0 to N-1: Borne ^ = Borne ^ _! + Wigliss and Bound ₀ = 0 For example, if we take the network part of Figure 3, we obtain the following calculation.

After initialization of the network, the producer nodes PI and P2 both have the same weight table:

TABLE 1

During a first iteration, the weights are cumulated on a scale, and we obtain the table 2.

TABLE 2

To send their next message, the producer nodes PI and P2 derive a value between 0 and 3. If this value is between 0 and 1, the next message will be sent to the consumer node C1, if the value is between 1 and 2. , the message will be sent to the consumer node C2, if the value is between 2 and 3, the message will be sent to the consumer node C3. By convention, if the value is exactly 1, it will be sent to the consumer node C1, and if the value is exactly 2, it will be sent to the consumer node C2.

For example, consider the draw value 1.5 for the producer node PI and 2.2 for the producer node P2. The producer node PI sends its message to the consumer node C2, while the producer node P2 sends its message to the consumer node C3. The consumer node C2 responds in 5 seconds to the producer node PI and the consumer node C3 responds in 2 seconds to the producer node P2.

Applying the formulas (1), (2) and (3) presented above, for the producer node PI, we obtain:

1 1

W _c2 o =; - = - = 0.2

n.arrivee n.depart ^

ô _c2 = {0.2}

1

W _c2 gliss = - * W _{c2 0} = 0.2

and for the producer node P2:

1 1

W _c3 o =; - = j = 0.5

n.arrivee ^ -i.depart ^

O _C3 : {0.5}

W _c3 gliss = - * W _{c3 0} = 0.5

The producer nodes PI and P2 now have different weights for each of the consumer nodes C1, C2 and C3, which gives the table 3.

TABLE 3

In a second iteration, weights are accumulated on a scale that is updated, and we get table 4.

TABLE 4

C1 C2 C3

PI 0 1 1.2 2.2

P2 0 1 2 2.5 The new random draw gives a value of 1.1 for the producer node PI and a value 0.3 for the producer node P2.

The producer node PI sends its message to the consumer node C2 and the producer node P2 sends its message to the consumer node C1.

The consumer node C2 responds to the producer node PI in 2 seconds and the consumer node C1 responds to the producer node P2 in 10 seconds.

Applying the formulas (1), (2) and (3) again for the producer node PI, we obtain:

1 1

W _c2! = - = - = 0.5

ti.arrivee ^ -i.depart ^

o _c2 = {0.2, 0.5}

W _c2 gliss = - * (W _{c2 0} + W _{c2 1} ) = 0.35

and for the producer node P2, we obtain:

1 1

W _CL o = = - = 0.1

.arrivee .depart - ^LU

6 _C1 : {0.1}

W _cl gliss = * W _{c1 0} = 0.1

The PI and P2 producer nodes now have weights again modified for the consumer nodes C1, C2 and C3, which gives the table

TABLE 5

PI P2

Cl 1 0.1

C2 0.35 1

C3 1 0.5 In a third iteration, weights are accumulated on a scale that is updated, and we get table 6.

TABLE 6

Thus, the final weights for consumer nodes C1 to C3 are dynamic sliding average weights.

The cumulative weight scale represents the three consumer nodes C1, C2 and C3 at the time t0 in a tree, each consumer node being represented by a sheet 4 comprising the name of the node and the value of its weight.

The cumulative weight scale represents the three consumer nodes C1, C2 and C3 at time t1 in the same tree, each consumer node being represented by a sheet 5 comprising the name of the node and the value of its weight, which may have been modified.

The cumulative weight scale is stored as a cumulative weight table via an interval tree ("range tree" in English), which brings average complexity with this technique in 0 (log (n)).

The leaves of the tree correspond to the boundaries of the intervals. The tree structure provides access to values in a given interval quickly and recalculating weights involves known algorithms.

We see, for the nodes C1, C2, C3, the limits of the intervals which are respectively 1/2/3, at time t0, and 1 / 1,2 / 2,2 at time tl.

This tree data structure storing this accumulated weight scale not only satisfies the storage problem of a scale of values, but also presents a useful additional advantage which is to have very good access and modification performance to the values stored therein.

Of course, the present invention is not limited to the examples and to the embodiment described and shown, but it is capable of numerous variants accessible to those skilled in the art.

Claims

A computer network of nodes (1, 2) communicating with each other by peer-to-peer messages, message producing nodes (1) sending their generated messages to message consumer nodes (2), a consumer node (2) capable of also being a producer node (1) at another level of the network (6), characterized in that each producer node (1) of message:

chooses, among the nodes (1, 2) registered as active in the network, its consumer nodes (2) to interconnect with them,

maintains a list of its interconnected consumer nodes (2),

manages the load balancing of its product messages to its interconnected consumer nodes (2), as a function of the processing time, by its interconnected consumer nodes (2), of its product messages, by sending its messages mainly to its consumer nodes ( 2) interconnected with lower processing time,

modifies its list of interconnected consumer nodes (2) according to the processing time, by its interconnected consumer nodes (2), of its generated messages, by deleting from its list at least temporarily its interconnected consumer nodes (2) too slow.

2. Computer network according to claim 1, characterized in that the nodes (1, 2) register as active in the network near a centralized register (3) in the network (6), which register diffuses to at least all the producing nodes (1) of the network (6), preferably at all the active nodes (1, 2) of the network (6), the list of at least all the active consumer nodes (2) of the network (6) ), preferably the list of all active nodes (1, 2) of the network (6).

3. Computer network according to claim 2, characterized in that said register (3) establishes for no particular producer node (1) given the list of its interconnected consumer nodes (2), does not send to any producing node (1 ) particular given the list of its interconnected consumer nodes (2).

4. Computer network according to any one of claims 2 to 3, characterized in that the nodes (1, 2) periodically register as active in the network (6) with the register (3), as soon as they are able to process messages, the recording period preferably being less than or equal to 10 seconds, any active node (1, 2) which lacks at least two recording periods, preferably at least one recording period, is removed from the list of active nodes (1, 2) by the register (3).

A computer network according to any one of the preceding claims, characterized in that each message producer node (1) can modify its interconnected consumer node list (2) by optionally adding a newly registered consumer node (2) if the one it is able to consume the messages produced by said producer node (1).

6. Computer network according to any one of the preceding claims, characterized in that each consumer node (2) sends, at least for certain messages received from a producer node (1), its processing time of these messages or a weight representative of its processing time of these messages, representative of the difference between the arrival of these messages at said consumer node (2) and their departure from said consumer node (2) possibly after transformation, preferably each producer node (1 ) adds a processing time evaluation request in all or some of the messages that it sends to each of its interconnected consumer nodes (2).

7. The computer network as claimed in claim 6, characterized in that each message producing node (1) manages the load distribution of its product messages to its interconnected consumer nodes (2), as a function of the processing times or the weights sent. through its interconnected consumer nodes (2), sending more product messages to the interconnected consumer nodes (2) whose processing times are lower or the weights corresponding to lower processing times.

Computer network according to claim 7, characterized in that each producer node (1) calculates a sliding average weight over time for each of its interconnected consumer nodes (2), which it uses to manage the load distribution. from its product messages to its interconnected consumer nodes (2).

A computer network according to claim 7 or 8, characterized in that each producer node (1) loads its interconnected consumer nodes (2) in a weighted random manner, so that the probability for its interconnected consumer nodes (2) to receive a message produced by said producer node (1), depends on a random draw which is weighted by the weights of said interconnected consumer nodes (2).

Computer network according to one of the preceding claims, characterized in that:

each node (1, 2) is a micro-service that exchanges messages with at least one other node (1, 2),

each micro-service being:

o is a processor that processes data received from a node (1, 2) of the network (6) to then send the processed data to another node (1, 2) of the network (6), preferably said processing being reduced to a single processing task performed on these data, o is a source that produces data to be injected into the network (6) to at least one processor,

or a sink that exports the data processed by at least one processor to a third party system (6).

11. The computer network as claimed in claim 10, wherein at least one of the processors, preferably the majority of the processors, and even more preferably all the processors, filters the received data and then routes them to different types of nodes. (1, 2) in the network (6).

12. Computer network according to any one of the preceding claims, characterized in that a producer node type (1) is a filter that separates the received messages according to a criterion to be filled, to direct them to three types of nodes consumers (2) respectively according to whether said criterion is: completed, or not filled, or perhaps filled.

13. Computer network according to claim 12, characterized in that the messages are representative of banking transactions, the criterion is the fraudulent nature or not of a bank transaction.

14. Computer network according to claim 12, characterized in that the messages are representative of images filmed by CCTV cameras, the criterion is the character of presence or absence of delictual risk related to the content of a filmed image or a sequence of images filmed by these CCTV cameras.

15. A method of interconnection and communication between nodes (1, 2) in a computer network (6) according to any one of the preceding claims.

16. A method of interconnection between nodes (1, 2) in a computer network (6) of nodes (1, 2) communicating with each other by means of messages peer-to-peer message producing nodes (1) sending their generated messages to message consumer nodes (2), a consumer node (2) also being a producer node (1) at another network level (6) characterized in that each producer node (1) of message:

chooses, among the nodes (1, 2) registered as active in the network (6), its consumer nodes (2) to interconnect with them, maintains the list of its interconnected consumer nodes (2),

modifies its list of interconnected consumer nodes (2) according to the processing time, by its consumer nodes (2), of its generated messages, by deleting from its list at least temporarily its interconnected consumer nodes (2) too slow.