DISTRIBUTED COMPUTING SYSTEM FOR DETECTING ERRORS BASED ON STREAM FILTRATION AND MONITORING

Abstract

A distributed computing system comprises one or more computing devices and one or more communication links that communicate electronically with each other. Each communication link contains one or more nodes and a connection switch. The one or more nodes communicate electronically with each other via a communication link, and the communication link is connected to the distributed computing system via the connection switch. The connection switch executes instructions to identify a unique online stream profile for a specific stream and to determine an expected deviation of the unique stream profile within a defined time interval.The connection switch determines a difference between a unique offline stream profile and the unique online stream profile, compares an absolute difference value with the expected deviation of the unique stream profile, and, in response to finding that the difference is greater than the expected deviation of the unique stream profile, determines that an error has occurred.

Description

INTRODUCTION

[0001] The present disclosure relates to a distributed computing system for detecting errors based on stream filtering and stream monitoring.

[0002] Distributed computing systems comprise multiple computing devices, also called nodes, that transmit and receive information over a communication link. The communication link can be implemented in hardware, such as a bus-based communication network, or wirelessly. For example, a distributed computing system may use the 10Base-T network communication protocol, with the nodes communicating electronically with each other over a twisted-pair cable.

[0003] A specific example of a distributed computing system using the 10Base-T protocol includes one or more multidrop buses that communicate with one or more computing devices. In this example, the nodes may be smart electrical devices, such as smart sensors or smart actuators, that have limited processing capabilities. The computing devices include greater processing power than the nodes, and therefore, functional and logical computing can be performed on the computing devices. Each bus includes a port switch that connects the nodes to the remaining buses that are part of the distributed computing system and the computing devices. Sometimes, one or more nodes, or one or more of the buses that are part of the distributed computing network, experience a failure that causes one or more nodes to go offline.A node may go offline because the connection between the relevant bus and the relevant node is interrupted, or when the relevant node becomes inoperable. Computing devices that are part of the distributed computing network are required to receive notification when one or more nodes or the bus goes offline.

[0004] Therefore, although distributed computing systems serve their intended purpose, there is a need in the art for a method to determine when one or more nodes or buses are no longer connected to the distributed computing system. SUMMARY

[0005] According to several aspects, a distributed computing system comprises one or more computing devices and one or more communication links in electronic communication with the one or more computing devices, each communication link including one or more nodes and a port switch, and the one or more nodes in electronic communication with each other through a communication link, the communication link being connected to the distributed computing system through the port switch. The port switch executes instructions to determine that the distributed computing system is online.In response to determining that the distributed computing system is online, the port switch identifies a unique online stream profile for a specific stream, where the specific stream corresponds to an application executing through one of the nodes in electronic communication with the port switch during a specified time interval. The port switch determines an expected deviation of the unique stream profile during a specified time interval. At the end of the specified time interval, the port switch determines a difference between a unique offline stream profile and the unique online stream profile, where the unique offline stream profile represents the behavior of the specific stream when the distributed computing system is offline. The port switch compares an absolute difference value to the expected deviation of the unique stream profile.In response to determining that the difference is greater than the expected deviation of the unique stream profile, the port switch determines that an error has occurred within the communication link.

[0006] In another aspect, in response to determining that the fault occurred within the communication link, the connection switch transmits a notification to each of the computing devices via a distributed computing system.

[0007] According to yet another aspect, the fault affects one of the nodes and / or the communication link.

[0008] In one aspect, the specific stream generated by the application during the specified time interval includes a single traffic parameter.

[0009] According to another aspect, the port switch determines that the error has occurred by subtracting a value of the individual traffic parameter of the specific stream represented by the unique offline stream profile from a value of the individual traffic parameter of the specific stream represented by the unique online stream profile to determine a difference of the individual traffic parameter, comparing the difference of the individual traffic parameter to the expected deviation of the unique stream profile, and in response to determining that the difference of the individual traffic parameter is greater than the expected deviation of the unique stream profile, determining that the error has occurred.

[0010] According to another aspect, the specific stream generated by the application during the specified time interval comprises more than one traffic parameter.

[0011] In one aspect, the connection switch determines the error by equating the expected deviation of the unique stream profile with a correlation factor of a first vector representing the unique offline stream profile and a second vector representing the unique online stream profile.

[0012] According to another aspect, the expected deviation of the unique stream profile represents a predetermined deviation value that one or more traffic parameters of the specific stream allow without detecting the error within the communication link.

[0013] According to yet another aspect, the traffic parameters include the packet transmission frequency and / or the packet length and / or the burst size and / or the packet priority.

[0014] In one aspect, the port switch performs per-stream filtering and monitoring based on the Institute of Electrical and Electronics Engineers (IEEE) 802.1Qci standard.

[0015] In another aspect, the system further comprises a switched network connecting the communication links to the one or more computing devices.

[0016] In one aspect, the communication link is a bus implemented based on the 10Base-T1S network communication protocol.

[0017] According to another aspect, a method for detecting a fault by a distributed computing system is disclosed. The method includes determining that the distributed computing system is online by a connection switch, wherein the distributed computing system comprises one or more communication links, each of the communication links including one or more nodes and a connection switch, wherein the one or more nodes are in electronic communication with each other through the communication link, and the communication link is connected to the distributed computing system through the connection switch.The method comprises, in response to the connection switch determining that the distributed computing system is online, identifying a unique online stream profile for a specific stream, wherein the specific stream corresponds to an application executed by one of the nodes in electronic communication with the connection switch during a specified time interval. The method comprises, by the connection switch, determining an expected deviation of the unique stream profile during a specified time interval. The method further comprises, at the end of the specified time interval, determining, by the connection switch, a difference between a unique offline stream profile and the unique online stream profile, wherein the unique offline stream profile represents the behavior of the specific stream when the distributed computing system is offline.The method includes comparing an absolute difference value with the expected deviation of the unique stream profile by the connection switch. The method includes, in response to determining that the difference is greater than the expected deviation of the unique stream profile, determining that a fault has occurred within the communication link.

[0018] In another aspect, the method includes, in response to determining that the error occurred within the communication link, transmitting a notification by the port switch via the distributed computing system to one or more computing devices.

[0019] According to yet another aspect, a distributed computing system comprises one or more computing devices and one or more communication links in electronic communication with the one or more computing devices, each communication link including one or more nodes and a port switch, and the one or more nodes in electronic communication with each other through a communication link, the communication link being connected to the distributed computing system through the port switch, and the port switch performing per-stream filtering and monitoring based on IEEE standard 802.1Qci. The port switch executes instructions to determine that the distributed computing system is online.In response to determining that the distributed computing system is online, the port switch identifies a unique online stream profile for a specific stream, where the specific stream corresponds to an application executing through one of the nodes in electronic communication with the port switch during a specified time interval. The port switch determines an expected deviation of the unique stream profile during a specified time interval. At the end of the specified time interval, the port switch determines a difference between a unique offline stream profile and the unique online stream profile, where the unique offline stream profile represents the behavior of the specific stream when the distributed computing system is offline. The port switch compares an absolute difference value to the expected deviation of the unique stream profile.In response to determining that the difference is greater than the expected deviation of the unique stream profile, the port switch determines that a fault has occurred within the communication link. In response to determining that the fault has occurred within the communication link, the port switch transmits a notification to each of the computing devices via a distributed computing system.

[0020] According to another aspect, the error affects one of the nodes and / or the communication link.

[0021] According to yet another aspect, the specific stream generated by the application during the specified time interval contains a single traffic parameter.

[0022] In one aspect, the port switch determines that the error has occurred by subtracting a value of the individual traffic parameter of the specific stream represented by the unique offline stream profile from a value of the individual traffic parameter of the specific stream represented by the unique online stream profile to determine a difference of the individual traffic parameter, comparing the difference of the individual traffic parameter to the expected deviation of the unique stream profile, and in response to determining that the difference of the individual traffic parameter is greater than the expected deviation of the unique stream profile, determining that the error has occurred.

[0023] According to another aspect, the specific stream generated by the application during the specified time interval comprises more than one traffic parameter.

[0024] In yet another aspect, the connection switch determines the error by equating the expected deviation of the unique stream profile with a correlation factor of a first vector representing the unique offline stream profile and a second vector representing the unique online stream profile.

[0025] Further areas of applicability will become apparent from the description provided herein. It should be understood that the description and specific examples are for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present disclosure in any way; they show: Fig. 1 is a schematic representation of a vehicle including the disclosed distributed computing system including one or more computing devices and one or more communication links including one or more nodes, according to an exemplary embodiment; Fig. 2A is a software representation of one of the nodes that, according to an exemplary embodiment, are part of the distributed computing system that is Fig. 1 is shown; Fig. 2B is a software representation of an Ethernet switch that is part of the distributed computing system described in Fig. 1 is shown; and Fig. 3 is a process flow diagram illustrating a method for detecting a fault within one of the communication links that are part of the distributed computing system, according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or uses.

[0028] In Fig. 1 is a schematic diagram illustrating an exemplary distributed computing system 10 that is part of a vehicle 12. According to the non-limiting embodiment shown in Fig. 1, the distributed computing system 10 is implemented as part of a vehicle, such as, but not limited to, a sedan, a truck, an SUV, a van, or a camper. It should be recognized, however, that Fig. 1 is merely exemplary in nature, and the disclosed distributed computing system 10 is not limited to a vehicle. Indeed, the distributed computing system 10 may be implemented in a variety of other applications, such as industrial automation control systems and the Internet of Things (IoT).

[0029] The distributed computing system 10 includes one or more communication links 22 that are in electronic communication with each other through a switched network 24. The switched network 24 includes a plurality of interconnected switches or a single switch. The one or more communication links 22 are also in electronic communication with one or more computing devices 32. Each communication link 22 is connected to the distributed computing system 10 through a port of a switch, each port being referred to as a port switch 34, which is Fig. 1. In the case where a wireless protocol is used, the connection switch 34 is a wireless access point.

[0030] According to the embodiment shown in Fig. 1, the distributed computing system 10 includes three communication links 22 and three computing devices 32; however, it should be appreciated that Fig. 1 is merely exemplary in nature. The communication links 22 are implemented either in hardware, such as a bus communication system, or wirelessly. According to the non-limiting embodiment shown in Fig. As shown in Figure 1, the communication link 22 is a bus implemented based on the 10Base-T1S network communication protocol. According to another embodiment, the network communication protocol is based on switched Ethernet.

[0031] The communication links 22 each contain nodes 30 up to a number i, where i is any integer greater than zero. In other words, the communication links 22 each contain one or more nodes 30. According to the embodiment shown in Fig. 1, each communication link 22 may include up to eight nodes 30 (or i = 8), where the communication links 22 are implemented based on the 10Base-T1S network communication protocol, however, it should be appreciated that a maximum number of nodes 30 supported by each communication link 22 varies based on the network communication protocol. According to one non-limiting embodiment, the nodes 20 may represent devices, such as smart sensors and smart actuators, that have limited computing power. It should be appreciated that the one or more computing devices 32 in embodiments may have higher computing power compared to the nodes 30, and therefore, the functional and logical computing of the distributed computing system 10 may be performed on the computing devices 32.

[0032] Nodes 30 are designated as N0m,N2m,…Nim where m denotes a total number of communication links 22 included as part of the distributed computing system 10 and m > 0. The communication links 22 are respectively designated as B1, B2,...B m the connection switches 32 are each designated as S1,S2,....S l and the one or more computing devices 32 are designated as C1, C2,··· C k where I denotes a total number of connection switches 34 and k denotes a total number of computing devices 32 included as part of the distributed computing system 10.

[0033] Fig. Figure 2A shows a software representation of one of the nodes 30 of the distributed computing system 10, wherein the node 30 is Nim Each node 30 executes applications Ai,jm up to a number j, where the application Ai,jm has the form of a service and j is any integer greater than zero (or j > 0). Each node 30 contains a physical address Eim, which is a unique value for the distributed computing system 10. According to the embodiment shown in the figures, in which the communication links 22 are implemented on the basis of the 10Base-T1S network communication protocol, the physical address Eim a media access control address (MAC address). Each application Ai,jm of a specific node 30 contains a connection value Pi,jm, which for every application Ai,jm is unique for a specific node 30. Qi,jm refers to a service identifier of the application Ai,jm and is for a specific service that the application Ai,jm executes, clearly. Ri,jm denotes a service instance identifier of the application Ai,jm and is for a specific service instance of the service identifier Qi,jm clearly. Si,jm refers to a stream that is sent by the application Ai,jm is generated.

[0034] Fig. Figure 2B shows a software representation of an exemplary Ethernet switch illustrating one of the port switches 32 designated as S lIt should be appreciated that each port switch 34 that is part of the distributed computing system 10 performs per-stream filtering and monitoring based on the IEEE (Institute of Electrical and Electronics Engineers) 802.1Qci standard. It should also be appreciated that each port switch 34 that is part of the distributed computing system 10 is configured to generate and forward notifications to the computing devices 32 based on a unicast and / or multicast approach. The port switches 34 each include one or more processors for performing the per-stream filtering and monitoring. For example, according to one embodiment, the port switches 34 include one or more central processing units (CPUs) based on a reduced instruction set computer (RISC) architecture.

[0035] In the Fig. 1, Fig. 2A and Fig. 2B shows a unique offline stream profile Πi,jm the behavior of a specific stream Si,jm, which, when the distributed computing system 10 is offline, is used by an application Ai,jm, which during a specified time interval ΔTi,jm executed by a specific node 30. It should be recognized that the unique offline stream profile Πi,jm can be updated when the distributed computing system 10 is online through a software update of the connection switches 34. During the design time of the distributed computing system 10, the unique offline stream profile Πi,jm defined and each connection switch 34 has knowledge of the unique offline stream profile Πi,jm. The unique offline stream profile ∏i,jm is a function of a corresponding physical address Eim, a corresponding connection value Pi,jm, a corresponding service identifier Qi,jm and a corresponding service instance identifier Ri,jm of the specific node 30 (or Πi,jm=f(Eim,Pi,jm,Qi,jm,Si,jm) )). The unique offline stream profile Πi,jm indicates one or more traffic parameters for the specific stream Si,jm. Some examples of traffic parameters include, but are not limited to, packet transmission frequency, packet length, burst size, and packet priority. It should be recognized that the unique offline stream profile Πi,jm can be a scalar value or a multidimensional vector depending on the number of traffic parameters contained.

[0036] It should be appreciated that the communication link 22 comprises a multidrop topology. Therefore, each stream generated by a node 30 that is part of a specific communication link 22 passes through the corresponding connection switch 34. Once the distributed computing system 10 goes online, a unique online stream profile πi,jm for the specific stream Si,jm detected and identified by the connection switch 34. Specifically, each stream Si,jm, generated by one of the nodes 30 of the distributed computing system 10, by a corresponding port switch 34 and is detected and identified by the corresponding port switch 34 based on the IEEE 802.1Qci standard. The corresponding port switch 34 detects the unique online stream profile. πi,jm for a corresponding, unique stream Si,jm, where the unique online stream profile πi,jm the same dimensions as the unique offline stream profile Πi,jm It should be appreciated that during design time and until the distributed computing system 10 goes online (ie, switches to transmission), each port switch 34 of the distributed computing system 10 is configured with a unique stream and a unique stream profile, where the stream refers to a data flow and the stream profile refers to a specific traffic behavior followed by a unique stream. The unique offline stream profile Πi,jm for the specific stream Si,jm is in the corresponding connection switch S l , which is connected to the communication link B m for all applications Ai,jm, from each of the nodes 30 that are part of the communication link B m are running, connected, configured.

[0037] A connection switch 34 corresponding to the relevant communication link 33 determines during the specified time interval ΔTi,jm an expected deviation of the unique stream profile ΔΠi,jm, where the expected deviation of the unique stream profile ΔΠi,jm a predetermined deviation value that the one or more traffic parameters of the specific stream Si,jm allow without an error Fi,jm within the corresponding communication link 22. The predetermined deviation value is set when the distributed computing system 10 is offline. It should be appreciated that the predetermined deviation value varies due to the accuracy required by a specific application on which the distributed computing system 10 is implemented. It should be appreciated that in embodiments, one or more nodes 30 may not be able to generate a regular stream profile. In this scenario, a less resource-intensive application Ai,zm be used or an artificial stream profile can be set instead of the regular stream profile.

[0038] At the end of the specified time interval ΔTi,jm the corresponding connection switch 34 determines a difference between the unique offline stream profile Πi,jm and the unique online stream profile πi,jm and compares an absolute difference value with the expected deviation of the unique stream profile Δ∏i,jm. In response to determining that the difference between the unique offline stream profile Πi,jm and the unique online stream profile πi,jm greater than the expected deviation of the unique stream profile ΔΠi,jm the corresponding connection switch 34 determines that an error Fi,jm, which is expressed as |Π - π| > ΔΠ, has occurred within the communication link 22. Specifically, the error Fi,jm to a specific application Ai,jm, which is executed by one of the nodes 30 that are part of the communication link 22. In response to determining the error Fi,jm the corresponding connection switch 34 transmits a notification via the distributed computing system 10 to each of the computing devices 32 that are part of the distributed computing system 10. The notification indicates that the error Fi,jm occurred.

[0039] It must be acknowledged that the error Fi,jm one of the nodes 30 that are part of a specific communication link 22, or the specific communication link 22 itself. If the error Fi,jm in all applications executed by a specific node 30 of the specific communication link 22, then the error concerns Fi,jm the specific node 30. If the error Fi,jm However, if the error is present in all applications executed by all nodes 30 that are part of the specific communication link 22, then the error affects Fi,jm the specific communication link 22.

[0040] Determining the error Fi,jm through the specific communication link 22 will now be described. It should be appreciated that the connection switch 34, when the specific stream Si,jm, of the application Ai,jm during a specified time interval ΔTi,jm generated contains a single traffic parameter, by subtracting a value of the single traffic parameter of the specific stream determined by the unique offline stream profile Πi,jm from a value of the individual traffic parameter of the specific stream, which is determined by the unique online stream profile πi,jm to determine a difference of the individual traffic parameter and compare the difference of the individual traffic parameter with the expected deviation of the unique stream profile ΔΠi,jm to compare, determines that the error Fi,jm In response to determining that the difference of the individual traffic parameter is greater than the expected deviation of the unique stream profile ΔΠi,jm the connection switch 34 determines that the error Fi,jm According to the present example, the expected deviation of the unique stream profile is Δ∏i,jm equal to the specified deviation value that the individual traffic parameters of the specific stream Si,jm without realizing that an error Fi,jm occurred. Determining the error Fi,jm is expressed in equation 1 as: |W−w>ΔW| where W is the value of the individual traffic parameter of the specific stream determined by the unique offline stream profile Πi,jm represents the value of the individual traffic parameter of the specific stream determined by the unique online stream profile πi,jm , ΔW represents the predetermined deviation value that the individual traffic parameters of the specific stream Si,jm without realizing that an error Fi,jm occurred, and W and w are both elements of the service instance identifier Ri,jm the application Ai,jm (or W∈Ri,jm,w∈Ri,jm, where Ri,jm represents a series of real numbers).

[0041] If the specific stream Si,jm, during the specified time interval ΔTi,jm from the application Ai,jm generated contains several traffic parameters, then the connection switch 34 determines the error Fi,jm by equating the expected deviation of the unique stream profile ΔΠi,jm with a correlation factor of a first vector that represents the unique offline stream profile Πi,jm and a second vector representing the unique online stream profile πi,jm Specifically, in one example, the first vector is expressed as a multidimensional vector [W, L], where W represents a first traffic parameter and L represents a second traffic parameter, and Πi,jm=[W,L], and [W,L]∈Ri,jm2 and Ri,jm2 represents a series of real numbers. The second vector is expressed as a multidimensional vector [w,l], where w represents the first traffic parameter, l represents a second traffic parameter, and πi,jm=[w,l], where [w,l]∈Ri,jm2 According to one embodiment, the correlation factor is equal to cos(π / 4), but it should be recognized that the correlation factor can range in value from 0 to 1. According to one embodiment, the terminal switch 34 determines that the error Fi,jm occurred if equation 2 is true, where equation 2 can be expressed as follows: |1−Ww+LlW2+L2w2+l2|>Correlation factor

[0042] According to one embodiment, the connection switch 34 determines the error Fi,jm, if equation 3 is true, where equation 3 can be expressed as follows: |1−dot(Πi,jm,πi,jm)|Πi,jm||πi,jm||>Correlation factor

[0043] Fig. 3 shows a process flow diagram illustrating a method 300 for detecting an error Fi,jm within one of the communication links 22 or one of the nodes 30 that are part of the distributed computing system 10. With general reference to the Fig. 1, Fig. 2A, Fig. 2B and Fig. 3, the method 300 may begin at decision block 302. In decision block 302, the port switch 34 of one of the communication links 22 continues to monitor the communication link 22 until the distributed computing system 10 is online. Once the port switch 34 determines that the distributed computing system is online, the method 300 may proceed to block 304.

[0044] In block 304, in response to determining that the distributed computing system 10 is online, the port switch 34 identifies the unique online stream profile πi,jm for the specific stream Si,jm, where the specific stream Si,jm an application Ai,jm performed by one of the nodes 30 during the specified time interval in electronic communication with the connection switch 34. The method 300 may continue with block 306.

[0045] In block 306, the connection switch determines the expected unique deviation of the stream profile ΔΠi,jm during the specified time interval ΔTi,jm. The method may then continue with block 308.

[0046] In block 308, the terminal switch 34 determines at the end of the specified time interval ΔTi,jm a difference between a unique offline stream profile Πi,jm and the unique online stream profile πi,jm, where the unique offline stream profile Πi,jm the behavior of the specific stream Si,jm when the distributed computing system 10 is offline. The method 300 may then proceed to decision block 310.

[0047] In decision block 310, the connection switch 34 compares an absolute difference value with the expected deviation of the unique stream profile ΔΠi,jm. In response to determining that the difference is less than or equal to the expected deviation of the unique stream profile ΔΠi,jm is, the connection switch 34 determines that no errors have occurred and the method 300 can be terminated.

[0048] In response to determining that the difference is greater than the expected deviation of the unique stream profile ΔΠi,jm the method 300 may continue to block 312. In block 312, the connection switch determines that the error Fi,jm occurred within the communication link 22. The method 300 may then proceed to block 314.

[0049] In block 314, the connection switch 34 transmits a notification via the distributed computing system to each of the computing devices 32, the notification indicating that the error Fi,jm has occurred. The method 300 can then be terminated.

[0050] With general reference to the figures, the disclosed distributed computing system provides various technical effects and advantages. Specifically, the distributed computing system provides a method for using existing per-stream filtering and monitoring techniques to detect errors within a communication link. It should be appreciated that since the disclosed method uses standard per-stream filtering and monitoring already present in the system to detect the error, no additional hardware is required. The disclosed method may result in improved detection times and bandwidth compared to existing methods because no additional packet transmission is required to detect the error.

[0051] The connection switches and computing devices may refer to or be part of an electronic circuit, a combinational logic circuit, a field-programmable gate array (FPGA), a processor (shared, dedicated, or group) that executes code, or a combination of some or all of the above, such as in a system on a chip. Additionally, the controllers may be microprocessor-based, such as a computer having at least one processor, memory (RAM and / or ROM), and associated input and output buses. The processor may operate under the control of an operating system located in memory. The operating system may manage computer resources such that computer program code embodied as one or more computer software applications, such as an application located in memory, may execute instructions by the processor.In an alternative embodiment, the processor may execute the application directly, in which case the operating system may be omitted.

[0052] The description of the present disclosure is merely exemplary in nature, and variations that do not depart from the gist of the present disclosure are intended to be within the scope of the present disclosure. Such variations should not be considered a departure from the spirit and scope of the present disclosure. legend

[0053] In the drawings, N stands for No and Y stands for Yes.

Claims

[1] Distributed computing system that includes: one or more computing devices; and one or more communication links in electronic communication with the one or more computing devices, each communication link comprising one or more nodes and a connection switch, and wherein the one or more nodes are in electronic communication with each other through a communication link, and the communication link is connected to the distributed computing system through the connection switch, the connection switch executing instructions to: determine that the distributed computing system is online; in response to determining that the distributed computing system is online, identify a unique online stream profile for a specific stream, the specific stream corresponding to an application executed by one of the nodes in electronic communication with the connection switch during a specified time interval; determine an expected deviation of the unique stream profile during a specified time interval; to determine, at the end of the specified time interval, a difference between a unique offline stream profile and the unique online stream profile, where the unique offline stream profile represents the behavior of the specific stream when the distributed computing system is offline. compare an absolute difference value with the expected deviation of the unique stream profile; and in response to determining that the difference is greater than the expected deviation of the unique stream profile, determine that an error has occurred within the communication link. [2] The distributed computing system of claim 1, wherein the port switch executes instructions to: in response to determining that the error occurred within the communication link, send a notification via the distributed computing system to each of the computing devices. [3] The distributed computing system of claim 1, wherein the fault affects one of the nodes and / or the communication link. [4] The distributed computing system of claim 1, wherein the specific stream generated by the application during the specified time interval includes a single traffic parameter. [5] The distributed computing system of claim 4, wherein the connection switch determines an occurrence of the fault by: subtracting a value of the individual traffic parameter of the specific stream represented by the unique offline stream profile from a value of the individual traffic parameter of the specific stream represented by the unique online stream profile to determine a difference of the individual traffic parameter; compares the difference of the individual traffic parameter with the expected deviation of the unique stream profile; and in response to determining that the difference of the individual traffic parameter is greater than the expected deviation of the unique stream profile, determines that the error has occurred. [6] The distributed computing system of claim 1, wherein the specific stream generated by the application during the specified time interval comprises more than one traffic parameter. [7] The distributed computing system of claim 6, wherein the port switch determines the error by equating the expected deviation of the unique stream profile with a correlation factor of a first vector representing the unique offline stream profile and a second vector representing the unique online stream profile. [8] The distributed computing system of claim 1, wherein the expected deviation of the unique stream profile represents a predetermined deviation value that one or more traffic parameters of the specific stream allow without detecting the error within the communication link. [9] A distributed computing system according to claim 8, wherein the traffic parameters include the packet transmission frequency and / or the packet length and / or the burst size and / or the packet priority. [10] The distributed computing system of claim 1, wherein the port switch performs per-stream filtering and monitoring based on the IEEE (Institute of Electrical and Electronics Engineers) 802.1Qci standard. [11] The distributed computing system of claim 1, further comprising a switched network connecting the communication links to the one or more computing devices. [12] The distributed computing system of claim 1, wherein the communication link is a bus implemented based on the 10Base-T1S network communication protocol. [13] A method for detecting an error by a distributed computing system, the method comprising: Determining by a connection switch that the distributed computing system is online, wherein the distributed computing system includes one or more communication links and each of the communication links includes one or more nodes and a connection switch, wherein the one or more nodes are in electronic communication with each other through the communication link and the communication link is connected to the distributed computing system through the connection switch; in response to determining that the distributed computing system is online, identifying, by the port switch, a unique online stream profile for a specific stream, the specific stream corresponding to an application executed by one of the nodes during a specified time interval in electronic communication with the port switch; Determining an expected deviation of the unique stream profile during a specified time interval by the connection switch; determining, by the connection switch, a difference between a unique offline stream profile and the unique online stream profile at the end of the predetermined time interval, wherein the unique offline stream profile represents the behavior of the specific stream when the distributed computing system is offline; Comparing an absolute difference value with the expected deviation of the unique stream profile by the connection switch; and Determining, in response to determining that the difference is greater than the expected deviation of the unique stream profile, that an error has occurred within the communication link. [14] The method of claim 13, further comprising: Transmitting, in response to determining that the error occurred within the communication link, a notification via a distributed computing system to each of the computing devices by the connection switch. [15] Distributed computing system that includes: one or more computing devices; and one or more communication links in electronic communication with the one or more computing devices, each communication link comprising one or more nodes and a connection switch, the one or more nodes being in electronic communication with each other through a communication link, and the communication link being connected to the distributed computing system through the connection switch, the connection switch performing per-stream filtering and monitoring based on IEEE standard 802.1Qci, and the connection switch executing instructions to: determine that the distributed computing system is online; in response to determining that the distributed computing system is online, identify a unique online stream profile for a specific stream, the specific stream corresponding to an application executed by one of the nodes during a specified time interval in electronic communication with the connection switch. determine an expected deviation of the unique stream profile during a specified time interval; at the end of the specified time interval, determine a difference between a unique offline stream profile and the unique online stream profile, wherein the unique offline stream profile represents the behavior of the specific stream when the distributed computing system is offline; compare an absolute difference value with the expected unique deviation of the stream profile; in response to determining that the difference is greater than the expected deviation of the unique stream profile, determine that an error has occurred within the communication link; and in response to determining that the error occurred within the communication link, send a notification via the distributed computing system to each of the computing devices.

Images