KR20190120057A - Stochastic Routing Algorithm for Load-balancing Interconnection Network System - Google Patents

Abstract

The present invention proposes a probability-based adaptive routing method and system for load balancing. According to an embodiment of the present invention, the probability-based adaptive routing method for load balancing comprises the steps of: receiving a packet to identify a destination, calculating a minimum path and a minimum proximity path for a transmission path from a source to a destination; calculating queue length for each of the calculated paths, and calculating a selection probability according to a load for each path; and selecting a path according to the calculated selection probability and transmitting the packet.

Description

Stochastic Routing Algorithm for Load-balancing Interconnection Network System

The present invention relates to a probability based adaptive routing method and system for load balancing.

A technique for adaptive routing using aggregated communication congestion information (US 2018-0006950) can obtain congestion values based on local occupancy and remote occupancy values within a switch. As a technique, a packet is sent to an output port having the least congestion based on the congestion value obtained above.

Mechanism for controlling resource utilization by using adaptive routing (Korean Patent Publication No. 10-2016-0004348) is a technology using an adaptive routing algorithm based on CAAT (Content Aware Adaptive Throttle), which is a deterministic routing The congestion free path is selected between the determined path and the adaptive routing path.

The technique for adaptive routing using aggregated communication congestion information transmits all packets on one path with minimal congestion, resulting in bottleneck and load imbalance problems.

The mechanism for controlling resource utilization using adaptive routing uses deterministic routing instead of adaptive routing when the load of the adaptive routing path is above a certain level to alleviate the above problem. In the case of a mechanism for controlling resource utilization using adaptive routing, a packet is transmitted only through an adaptive routing path and a deterministic path, and thus there is a problem in that the devices existing in the network system cannot be fully utilized.

The technical problem to be achieved by the present invention is to provide a method and system that can efficiently distribute the load across the network through the probabilistic path selection based on the current load situation of each path in consideration of two or more paths between the source and the destination. It is.

In one aspect, the probability-based adaptive routing method for load balancing proposed by the present invention includes receiving a packet to identify a destination, calculating a minimum path and a minimum proximity path for a transmission path from a source to a destination, Computing the calculated queue length for each path, calculating a selection probability according to the load for each path, and selecting a path according to the calculated selection probability and transmitting a packet.

Receiving the packet to identify the destination, and calculating the minimum path and the minimum proximity path for the transmission path from the source to the destination, after determining the minimum and minimum set of paths from one node to another node, the minimum The load for all nodes on each path belonging to the set of paths and the set of least neighboring paths is updated at predetermined times, and the sum of all nodes in the paths for each path is defined as the load of the corresponding path.

Calculating the calculated queue length for each path, and calculating the selection probability according to the load for each path, it is necessary to reduce the load on each path when transmitting packets from one node to another in order to balance the load on each path. Calculate the selection probability with the higher probability.

In order to reduce the bottleneck, the step of transmitting a packet by selecting a path based on the calculated probability of selection uses a stochastic routing method in which the load is distributed by selecting a path with a higher probability as the load of each path is smaller.

In another aspect, the probabilistic adaptive routing system for load balancing proposed in the present invention receives a packet to identify a destination and calculates a minimum path and a minimum proximity path for a transmission path from a source to a destination. The path search unit may include a probability calculator configured to calculate the calculated queue length for each path, calculate a selection probability according to the load for each path, and transmit a packet by selecting a path according to the calculated selection probability.

The path search unit sets the minimum path set and the minimum proximity path set from one node to another node, and then updates the load for every node on each path belonging to the minimum path set and the minimum proximity path set at predetermined time intervals. For each path, the sum of all nodes in the path is defined as the load of the path.

In order to balance the load of each path, the probability calculation unit calculates a selection probability having a higher probability as the load of each path is smaller when transmitting a packet from one node to another node.

In order to reduce the bottleneck, the packet transmitter uses a probabilistic routing method in which the load is distributed by selecting a path with a higher probability as the load on each path is smaller.

According to embodiments of the present invention, in consideration of two or more paths between a source and a destination, load can be efficiently distributed throughout the network through probabilistic path selection based on current load conditions of each path. In addition, it is possible to efficiently use the equipment inside the network system, thereby improving the packet processing speed performance.

1 is a diagram illustrating a probability-based adaptive routing algorithm for load balancing according to an embodiment of the present invention.
2 is a flowchart illustrating a probability-based adaptive routing method for load balancing according to an embodiment of the present invention.
3 is a diagram illustrating a configuration of a probability-based adaptive routing system for load balancing according to an embodiment of the present invention.

The present invention proposes a high performance computing system (HPC) and a method for probabilistically determining a plurality of paths as candidates for routing paths from one node to another node in a data center. A data center is a central processing unit that manages large servers and manages databases by connecting hundreds to thousands of computers through switches.

In the present invention, the load of the node is defined as follows. The dictionary meaning is 'the number of packets to be processed that have not yet been processed by the node', and may be defined as 'the estimated waiting time until the node exits when a packet arrives at the node.' After determining the minimum set of paths from one node to another and the set of paths that are closest to the minimum, the load on all nodes on each path in each of the two sets is updated at regular intervals, and for each path, The sum of all nodes is defined as the load of the path. Now, in transmitting a packet from one node to another node, the smaller the load value of each path is, the higher the probability is to select the path and set the routing path. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a probability-based adaptive routing algorithm for load balancing according to an embodiment of the present invention.

Adaptive routing algorithms in the prior art select one minimum path when the best path is congested. At this time, a problem of load imbalance may occur.

를 선택할 확률

는 각 경로의 로드를 고려하여 결정한다. 각 경로의 로드가 균형을 이루도록 하는

설정을 통한 확률적 라우팅으로, 종래기술의 적응형 라우팅 알고리즘들의 로드 불균형 문제를 해결할 수 있다. According to an embodiment of the invention, the k th path of k paths for any source-destination pair

Probability of choosing

Is determined by considering the load of each path. To balance the load on each path

Probabilistic routing through configuration can solve the load imbalance problem of the adaptive routing algorithms of the prior art.

2 is a flowchart illustrating a probability-based adaptive routing method for load balancing according to an embodiment of the present invention.

Probability-based adaptive routing for the proposed load balancing method receives a packet to identify the destination, calculates the minimum path and the minimum proximity path for the transmission path from the source to the destination (210), the queue by the calculated path Calculating the length, calculating a selection probability according to the load for each path (220) and selecting a path according to the calculated selection probability and transmitting a packet (230).

In step 210, a packet is received 211 to identify a destination 212, and a minimum path and a minimum proximity path for a transmission path from a source to a destination are calculated 213. After determining the minimum and minimum set of paths from one node to another, update the loads for all the nodes on each path in the set of minimum and minimum paths at predetermined times, and For a path, the sum of all nodes in the path is defined as the load of that path.

를 계산한다(222). 각 경로의 로드가 균형을 이루도록 하기 위해 하나의 노드에서 다른 노드로 패킷을 전송할 때, 각 경로의 로드가 적을 수록 높은 확률을 갖는 선택 확률을 계산한다. In step 220, the calculated queue length for each path is calculated (221), and the selection probability according to the load for each path.

Calculate (222). In order to balance the load of each path, when a packet is transmitted from one node to another node, the smaller the load of each path, the higher the probability of selection is calculated.

설정 과정에 대하여 설명한다. For load balancing according to an embodiment of the present invention

The setting process will be described.

The total k paths between source and destination can be expressed as follows when the minimum path is N and the minimum path is M.

N minimum paths

M minimum proximity paths

를 선택할 확률

는 다음과 같이 나타낼 수 있다: Where kth path

Probability of choosing

Can be written as:

는 경로

의 큐 길이이고, T는 큐 길이 업데이트의 주기이다. here,

The path

Is the queue length, and T is the period of queue length update.

는 다음과 같이 나타낼 수 있다: Load of kth path at this time

Can be written as:

에 따른 경로를 선택(231)하여 패킷을 전송한다(232). 병목현상을 감소시키기 위해 각 경로의 로드가 적을 수록 높은 확률로 경로를 선택하여 로드를 분산시키는 확률론적 라우팅 방법을 이용한다. In step 230, the calculated selection probabilities

In step 232, the path is selected and the packet is transmitted. In order to reduce bottlenecks, we use stochastic routing method that selects paths with high probability as load of each path is smaller.

In other words, the present invention is a probability-based routing method that can significantly improve the performance of the deterministic routing method.

In addition, while many existing routing methods can be used limited to some specific topologies, the present invention is a routing technique applicable to all topologies.

Therefore, the routing scheme according to the embodiment of the present invention can greatly reduce the bottleneck that may occur when data is transferred between nodes in the supercomputer / data center, and thus may have a fault-tolerant feature. have.

3 is a diagram illustrating a configuration of a probability-based adaptive routing system for load balancing according to an embodiment of the present invention.

The probability-based adaptive routing system 300 for load balancing according to the present embodiment may include a processor 310, a bus 320, a network interface 330, a memory 340, and a database 350. Memory 340 may include an operating system 341 and a probabilistic based adaptive routing routine 342 for load balancing. The processor 310 may include a path searcher 311, a probability calculator 312, and a packet transmitter 313. In other embodiments the probability based adaptive routing system 300 for load balancing may include more components than the components of FIG. 3. However, it is not necessary to clearly show most of the prior art components. For example, the probability based adaptive routing system 300 for load balancing may include other components, such as a display or a transceiver.

The memory 340 is a computer-readable recording medium, and may include a permanent mass storage device such as random access memory (RAM), read only memory (ROM), and a disk drive. In addition, the memory 340 may store program code for the operating system 341 and the probability-based adaptive routing routine 342 for load balancing. These software components may be loaded from a computer readable recording medium separate from the memory 340 using a drive mechanism (not shown). Such a separate computer-readable recording medium may include a computer-readable recording medium (not shown) such as a floppy drive, a disk, a tape, a DVD / CD-ROM drive, a memory card, and the like. In other embodiments, the software components may be loaded into the memory 340 via the network interface 330 rather than the computer readable recording medium.

The bus 320 may enable communication and data transfer between components of the probability-based adaptive routing system 300 for load balancing. Bus 320 may be configured using a high-speed serial bus, a parallel bus, a storage area network (SAN) and / or other suitable communication technology.

The network interface 330 may be a computer hardware component for connecting the probability-based adaptive routing system 300 for load balancing to a computer network. The network interface 330 may connect the probability based adaptive routing system 300 for load balancing to a computer network via a wireless or wired connection.

The database 350 may serve to store and maintain all information necessary for probability-based adaptive routing for load balancing. In FIG. 3, a database 350 is built and included in the probability-based adaptive routing system 300 for load balancing. However, the present invention is not limited thereto and may be omitted depending on a system implementation method or environment. It is also possible for all or part of the database to exist as an external database built on a separate, different system.

The processor 310 may be configured to process instructions of a computer program by performing input and output operations of the probabilistic adaptive routing system 300 for basic arithmetic, logic, and load balancing. The instructions may be provided to the processor 310 by the memory 340 or the network interface 330 and via the bus 320. The processor 310 may be configured to execute program codes for the path searcher 311, the probability calculator 312, and the packet transmitter 313. Such program code may be stored in a recording device such as memory 340.

The path search unit 311, the probability calculator 312, and the packet transmitter 313 may be configured to perform the steps 210 to 230 of FIG. 2.

The probability-based adaptive routing system 300 for load balancing may include a path searcher 311, a probability calculator 312, and a packet transmitter 313.

The path search unit 311 receives the packet to identify the destination, and calculates the minimum path and the minimum proximity path for the transmission path from the source to the destination. After determining the minimum and minimum set of paths from one node to another, update the loads for all the nodes on each path in the set of minimum and minimum paths at predetermined times, and For a path, the sum of all nodes in the path is defined as the load of that path.

를 계산한다. 각 경로의 로드가 균형을 이루도록 하기 위해 하나의 노드에서 다른 노드로 패킷을 전송할 때, 각 경로의 로드가 적을 수록 높은 확률을 갖는 선택 확률을 계산한다. The probability calculator 312 calculates the calculated queue length for each path and selects a probability according to the load for each path.

Calculate In order to balance the load of each path, when a packet is transmitted from one node to another node, the smaller the load of each path, the higher the probability of selection is calculated.

에 따른 경로를 선택하여 패킷을 전송한다. 병목현상을 감소시키기 위해 각 경로의 로드가 적을 수록 높은 확률로 경로를 선택하여 로드를 분산시키는 확률론적 라우팅 방법을 이용한다. The packet transmitter 313 calculates the selected probability

Select the path according to the packet transmission. In order to reduce bottlenecks, we use stochastic routing method that selects paths with high probability as load of each path is smaller.

The routing scheme according to an embodiment of the present invention can greatly reduce a bottleneck that may occur when data is transmitted between nodes within a supercomputer / data center, and thus may have a fault-tolerant feature.

The present invention can be adopted and used as a routing technique for all types of topology supercomputers / data centers that have been implemented in the past, and has a high versatility over several existing routing patents. In addition, if the existing routing scheme is replaced with the routing proposed in the present invention, the bottleneck can be greatly reduced in the case of a packet congestion in a specific node. Reduced bottlenecks will significantly improve overall system performance because the more nodes that are bottlenecked, the less traffic will arrive on time, the more packets will be lost and retransmitted, which will adversely affect overall system performance. Can be. Therefore, when adopting the technology proposed by the present invention in many companies that implement the existing supercomputer / data center, it is expected that it can represent a significant performance improvement.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable arrays (FPAs), It may be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of explanation, one processing device may be described as being used, but one of ordinary skill in the art will appreciate that the processing device includes a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as parallel processors.

The software may include a computer program, code, instructions, or a combination of one or more of the above, and configure the processing device to operate as desired, or process it independently or collectively. You can command the device. Software and / or data may be any type of machine, component, physical device, virtual equipment, computer storage medium or device in order to be interpreted by or to provide instructions or data to the processing device. It can be embodied in. The software may be distributed over networked computer systems so that they may be stored or executed in a distributed manner. Software and data may be stored on one or more computer readable recording media.

The method according to the embodiment may be embodied in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

Although the embodiments have been described by the limited embodiments and the drawings as described above, various modifications and variations are possible to those skilled in the art from the above description. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different manner than the described method, or other components. Or even if replaced or substituted by equivalents, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are within the scope of the claims that follow.

Claims (8)

Receiving a packet to identify a destination, and calculating a minimum path and a minimum proximity path for the transmission path from the source to the destination;
Calculating a calculated queue length for each path, and calculating a selection probability according to the load for each path; And
Selecting a path according to the calculated selection probability and transmitting a packet
Adaptive routing method comprising a. The method of claim 1,
Receiving the packet to identify the destination, and calculating the minimum path and the minimum proximity path for the transmission path from the source to the destination,
After deciding the minimum set of paths and the closest set of paths from one node to another, update the loads for all nodes on each path in the set of minimum and set of closest paths at predetermined times, For a path, the sum of all nodes in the path is defined as the load of that path.
Adaptive Routing Method. The method of claim 1,
Calculating the calculated queue length for each path, and calculating the selection probability according to the load for each path,
In order to balance the load on each path, when a packet is sent from one node to another node, the smaller the load on each path, the higher the probability of selection is calculated.
Adaptive Routing Method. The method of claim 1,
Selecting a path according to the calculated selection probability and transmitting the packet,
To reduce bottlenecks, we use stochastic routing method that distributes loads by selecting paths with higher probability as the load on each path is less.
Adaptive Routing Method. A path search unit for receiving a packet to identify a destination and calculating a minimum path and a minimum proximity path for a transmission path from a source to a destination;
A probability calculator which calculates the calculated queue length for each path and calculates a selection probability according to the load for each path; And
Packet transmission unit for transmitting a packet by selecting a path based on the calculated selection probability
Adaptive routing system comprising a. The method of claim 5,
The route search unit,
After deciding the minimum set of paths and the closest set of paths from one node to another, update the loads for all nodes on each path in the set of minimum and set of closest paths at predetermined times, For a path, the sum of all nodes in the path is defined as the load of that path.
Adaptive Routing System. The method of claim 5,
Probability calculation unit,
In order to balance the load on each path, when a packet is sent from one node to another node, the smaller the load on each path, the higher the probability of selection is calculated.
Adaptive Routing System. The method of claim 5,
The packet transmission unit,
To reduce bottlenecks, we use stochastic routing method that distributes loads by selecting paths with higher probability as the load on each path is less.
Adaptive Routing System.

Images