Device and method for quickly recovering faults of partial devices in optical data exchange or computing device cluster

Abstract

The invention relates to the technical field of data communication, and provides a device and a method for quickly recovering faults of partial equipment in an optical data exchange or computing equipment cluster. Wherein the management device confirms the identification number m of the second layer device sending the alarm message; according to a pre-established mapping relation between the identification number m of the second layer device and the optical path port Pm in each protection optical switch device, the management device controls the optical switches in each protection optical switch device in the protection optical switch array, so that the public port of the optical switches is conducted with the optical path port Pm; therefore, the optical signals led into the second layer of equipment m by the first layer of equipment library are re-etched into the second layer of redundant equipment through the optical switches and the optical couplers m in the protection optical switch equipment. According to the invention, when a part of equipment fails, the function of the equipment can be quickly switched to redundant equipment, so that the failure recovery time is saved, and cluster equipment resources are saved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data communication, in particular to a device and method for rapid recovery of partial device failure in optical data switching or computing equipment cluster. BACKGROUND

[0002] Optical data switching or computing equipment has several optical ports. The optical signal transmitted by the optical fiber received by the optical port is converted into an electrical signal by an optical-electric converter, processed internally by a digital signal, and forwarded to the corresponding target port, and then converted into an optical signal by an electric-optical converter, and sent out from the optical fiber. The whole process realizes high-speed and reliable data transmission and is widely used in data centers, supercomputers and other systems.

[0003] In a data center, optical data switching or computing equipment often appears in the form of a cluster, and several devices are divided into multiple functional levels, characterized by the same device function, the same number of ports, and similar connection methods in the same level.

[0004] Taking the topology of the leaf-spine Spine-Leaf architecture as an example, the data center network topology structure composed of two data switching layers of Spine and Leaf. The Leaf layer is composed of access switches that aggregate traffic from servers and are directly connected to the Spine or network core. The Spine switch interconnects all Leaf switches in the full mesh topology. Each level has several data switching devices with the same function, and they have the same model, the same number of ports, and the same connection method according to a certain rule, with high symmetry. The data switching device has a large number of ports, and the failure of the device itself or part of the ports will reduce the availability of the entire cluster and cause data switching congestion.

[0005] In the emerging AI supercomputer cluster, the computing power equipment is also often connected to the data switching equipment in the form of a cluster. Taking a 100-node DGX SuperPOD as an example, each superPOD includes 20 A100 GPUs, and each A100 GPU includes 8 IB Compute CoNNectioN, which is connected to the 8 Leaf computer switches of the SuperPOD. All GPUs in the same SuperPOD or even larger-scale GPUs across multiple SuperPODs complete computing tasks in parallel, and the failure of one GPU may cause the computing task of the entire cluster to be interrupted and rolled back, wasting a large amount of computing power.

[0006] Therefore, it is urgent to overcome the defects of the prior art in the technical field. SUMMARY

[0007] The technical problem to be solved by the present application is that the failure of individual devices in a multi-level optical data exchange or computing cluster has a greater impact.

[0008] The embodiment of the present application adopts the following technical solutions:

[0009] In a first aspect, the present application provides a rapid recovery device for partial device failure in an optical data exchange or computing device cluster, which includes one or more protection optical switch devices, comprising:

[0010] One end of the protection optical switch device has N optical path ports, respectively optical path port Q1, optical path port Q2, …, optical path port QN, and the other side has N+1 ports, respectively optical path port P1, optical path port P2, …, optical path port PN, and optical path port PC;

[0011] The optical path port Qi and the optical path port Pi are respectively connected in series at the common port of the optical coupler i and the first port of the optical coupler i, and the second port of the optical coupler i is connected through the 1×N optical switch between the optical path port PC via the optical switch selection conduction; wherein the i-th port of the 1×N optical switch is coupled with the second port of the optical coupler i, the common port of the 1×N optical switch is coupled with the optical path port PC, and i∈[1, N];

[0012] Wherein, when the 1×N optical switch switches to the k-th port of itself and the common port of itself conduction, the optical path of the optical path port Pk and the optical path of the optical path port PC are both connected with the optical path port Qk.

[0013] Preferably, an optical amplifier is further connected in series between the common port of the 1×N optical switch and the optical path port PC.

[0014] Preferably, the rapid recovery device includes one or more protection optical switch devices, specifically including protection optical switch device 1, protection optical switch device 2, …, protection optical switch device M, wherein the optical switch in each protection optical switch device is uniformly managed by a management device;

[0015] The output end of the first layer device library includes at least M sets of transmission ports, and each set of transmission ports includes at most N transmission ports;

[0016] Each set of transmission ports is carried by the optical path ports Q1-QN of a protection optical switch device, and the optical path ports P1-PN of the corresponding protection optical switch devices are respectively allocated to the second layer devices 1-N;

[0017] Wherein, the optical path port PC of each protection optical switch device is linked with a redundant device, and the redundant device has the functional attributes of each second layer device; each second layer device and the redundant device further establish a data control signal interaction link with the management device.

[0018] Preferably, the management device is configured to acquire the alarm message from any one of the second layer devices 1-N, and further comprises:

[0019] The management device confirms the identification number m of the second layer device sending the alarm message;

[0020] According to the pre-established mapping relationship between the identification number m of the second layer device and the optical path port Pm in each protection optical switch device, the management device controls the optical switch in each protection optical switch device, so that the common port is connected with the optical path port Pm, thereby copying the optical signal of the first layer device pool into the second layer redundant device through the optical switch in each protection optical switch device and the optical coupler m;

[0021] The second layer redundant device replaces the role of the second layer device m and continues to interact with the first layer device pool.

[0022] Preferably, the management device further comprises:

[0023] Any one of the second layer devices 1-N, when confirming that the communication between itself and the first layer device pool is in a temporary followable state, sends an optical path detection request to the management device;

[0024] The management device confirms whether the current second layer redundant device is in a replacement state, and if the second layer redundant device is in a replacement state, returns a response message to the second layer device x that the second layer redundant device is in a replacement state;

[0025] If the second layer redundant device is in an idle state, a response message is returned that the second layer redundant device enters a temporary followable state, wherein at the same time of sending the response message, the common port in each protection optical switch device is controlled to be connected with the corresponding optical path port Qx in the optical switch device allocated to the second layer device x, thereby confirming the state stability of the second layer device x by the management device through comparing the optical signals exchanged between the second layer redundant device and the second layer device x; wherein x belongs to [1, N].

[0026] Preferably, the temporary followable state specifically comprises:

[0027] The first layer device pool and the second layer device are in a heartbeat message maintaining link state; or,

[0028] The first layer device pool and the second layer device are in a continuous data sending state in a non-encrypted state; or,

[0029] The first layer device library and the second layer device are in an inherent script execution state, wherein the execution instruction of the script is previously known by the management device.

[0030] Preferably, the state stability of the second layer device x is confirmed by the management device through comparing the optical signals exchanged between the second layer redundant device and the second layer device x, and specifically further comprising:

[0031] The management device controls the working power of the optical amplifier in each protection optical switch device, and confirms whether the adjustment of the working power of the optical amplifier in each protection optical switch device is within a preset range, with the same optical power accepted by the second layer device x and the second layer redundant device as a target, so as to confirm the stability of the working state of each optical path port.

[0032] Preferably, the fast recovery device comprises one or more protection optical switch devices, and specifically comprises a northward protection optical switch array composed of U protection optical switch devices and a southward protection optical switch array composed of V protection optical switch devices; N third layer devices and a third layer redundant device are connected in series between the northward protection optical switch array and the southward protection optical switch array, wherein the number of optical interfaces of each third layer device towards the southward protection optical switch array is V, and the number of optical interfaces of each third layer device towards the northward protection optical switch array is U.

[0033] The other side optical interfaces of the northward protection optical switch array and the southward protection optical switch array, except the optical interfaces coupled with the third layer devices, are respectively used for connecting a northward device library and a southward device library.

[0034] Preferably, the optical switch in each protection optical switch device is uniformly managed by the management device, and each third layer device and the third layer redundant device also establish a data interaction channel with the management device.

[0035] Preferably, the second layer device is specifically computer rack 1, computer rack 2,..., computer rack 8; the second layer redundant device is specifically a redundant computer rack; the first layer device library is specifically 8 leaf computer switches; and 32 backup optical switches are matched, wherein N is 8, and the fast recovery device comprises:

[0036] The j'th port of the k'th H100 of the redundant computer rack is connected to the PC port of the 4k'+j'-4'th backup optical switch; wherein each H100 comprises a group of 8 optical interfaces numbered 1 to 8.

[0037] The j'th port of the k'th H100 of the i'th computer rack is connected to the Pi' port of the 4k'+j'-4'th backup optical switch.

[0038] The port Q1 of the 4th backup optical switch is connected to the port 4i+1 of the j'th leaf computing switch; wherein k' and j' are natural numbers.

[0039] In a second aspect, the present application further provides a method for fast recovery of partial equipment failure in an optical data switching or computing device cluster, the fast recovery device comprising a protection optical switching device 1, a protection optical switching device 2,..., and a protection optical switching device M, wherein the optical switches in each protection optical switching device are uniformly managed by a management device; the output end of the first layer device library comprises at least M sets of transmission ports, each set of transmission ports comprising at most N transmission ports; each set of transmission ports is carried by the optical path ports Q1-QN of one protection optical switching device, while the optical path ports P1-PN of the corresponding respective protection optical switching devices are respectively assigned to the second layer devices 1-2; the method comprises:

[0040] The management device is configured to acquire an alarm message originating from any one of the second layer devices 1-2;

[0041] The management device confirms the identification number m of the second layer device sending the alarm message;

[0042] According to a pre-established mapping relationship between the identification number m of the second layer device and the optical path port Pm in each protection optical switching device, the management device controls the optical switch in each protection optical switching device so that its common port is connected to the optical path port Pm, thereby copying the optical signal from the first layer device library to the second layer redundant device via the optical switch and the optical coupler m in each protection optical switching device;

[0043] The second layer redundant device replaces the role of the second layer device m and continues to interact with the first layer device library.

[0044] Preferably, the method further comprises:

[0045] Any one of the second layer devices 1-2, when confirming that the communication between itself and the first layer device library is in a temporarily followable state, sends an optical path detection request to the management device;

[0046] The management device confirms whether the current second layer redundant device is in a replacement state, and if the second layer redundant device is in a replacement state, returns a response message to the second layer device x that the second layer redundant device is in a replacement state;

[0047] If the second layer redundant device is in an idle state, a response message is returned, in which the common port in each protection optical switch device is turned on with the optical path port Qx allocated to the second layer device x in the optical switch device at the same time of sending the response message, so that the state stability of the second layer device x is confirmed by the management device through comparison of the optical signals exchanged between the second layer redundant device and the second layer device x.

[0048] Preferably, the temporarily followable state specifically includes:

[0049] The first layer device library and the second layer device are in a heartbeat message maintaining link state; or,

[0050] The first layer device library and the second layer device are in a continuous data sending state in a non-encrypted state; or,

[0051] The first layer device library and the second layer device are in an inherent script execution state, wherein the execution instruction of the script is known in advance by the management device.

[0052] Preferably, the state stability of the second layer device x is confirmed by the management device through comparison of the optical signals exchanged between the second layer redundant device and the second layer device x, and specifically further includes:

[0053] The management device controls the working power of the optical amplifier in each protection optical switch device, and in the case of taking the same optical power accepted by the second layer device x and the second layer redundant device as a target, it is confirmed whether the adjustment of the working power of the optical amplifier in each protection optical switch device is within a preset range, so as to confirm the stability of the working state of each optical path port.

[0054] Compared with the prior art, the embodiment of the present application has the beneficial effects that:

[0055] The present application realizes that when part of the device fails, the function of the device can be quickly switched to the redundant device through N-to-1 backup of the optical data exchange or computing cluster and the same level device, saves the fault recovery time, and saves the cluster device resources. Compared with the prior art, when the optical data exchange or computing cluster fails, the manual replacement of fiber connection is generally used to replace the failed device or port, which not only reduces the utilization efficiency of the device itself, but also consumes human resources and has huge time cost. BRIEF DESCRIPTION OF DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 is a structure schematic diagram of a backup optical opening device in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0058] Figure 2 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0059] Figure 3 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0060] Figure 4 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0061] Figure 5 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0062] Figure 6 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0063] Figure 7 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0064] Figure 8 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application;

[0065] Figure 9 is a structure schematic diagram of a backup optical opening device with an optical amplifier in a fast recovery device for partial device failure in an optical data exchange or computing device cluster provided by an embodiment of the present application; DETAILED DESCRIPTION

[0066] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0067] In the description of the present application, the terms "inner", "outer", "longitudinal", "transverse", "upper", "lower", "top", "bottom" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and do not require the present application to be constructed and operated in a particular orientation, therefore should not be understood as a limitation on the present application.

[0068] In the present application, the terms "first", "second" and the like are only used for descriptive purposes and should not be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" and the like can include one or more of the features explicitly or implicitly. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0069] In the embodiments of the present application, the "first layer" and "second layer" applied can respectively represent the "upper layer" and "lower layer" concepts in a specific architecture, or respectively represent the "lower layer" and "upper layer" concepts (the meaning expressed here is that the position of the redundant device in the embodiments of the present application can be set to the opposite layer in a similar mirroring manner); and in other specific architectures, they can also respectively represent the "north" and "south" concepts, or respectively represent the "south" and "north" concepts. Based on the core of the technical concept of the present application, other architecture description methods can also be applied, which are not limited here.

[0070] In the present application, unless otherwise specified and limited, the term "connection" should be understood broadly, for example, "connection" can be fixed connection, or detachable connection, or integral; can be directly connected, or indirectly connected through intermediate medium. In addition, the term "coupling" can be an electrically connected way for signal transmission.

[0071] The application proposes N-to-1 backup for optical data exchange or computing cluster level devices, that is, one redundant device of the same type is added to N parallel optical data exchange or computing devices, the optical paths connected by the ports of the same position in each device are combined through a beam splitter, and the backup paths connected to the ports of the same position of the N parallel devices are connected to the optical paths of the corresponding position ports of the redundant device through a 1:N optical switch. When a device, such as the Nth device, in the N parallel optical data exchange or computing devices fails, the management device detects the failure of the device, and then transfers the state and data of the device to the redundant device, disables the ports of the failed device, and switches the state of the backup optical switch connected to all the ports to the Nth port, so that the functions originally required by the Nth optical data exchange or computing device are transferred to the redundant optical data exchange or computing device to be implemented.

[0072] Furthermore, the technical features involved in each embodiment of the application described below can be combined with each other as long as there is no conflict between them.

[0073] Embodiment 1:

[0074] Embodiment 1 of the application provides a rapid recovery device for partial device failure in an optical data exchange or computing device cluster, which comprises one or more protection optical switch devices, such as Figure 1 as shown, comprising:

[0075] One end of the protection optical switch device has N optical path ports, which are optical path port Q1, optical path port Q2, …, and optical path port QN, respectively, and the other side has N+1 ports, which are optical path port P1, optical path port P2, …, optical path port PN, and optical path port PC.

[0076] The optical path port Qi and the optical path port Pi are respectively connected in series at the common port of the optical coupler i and the first port of the optical coupler i, and the second port of the optical coupler i is connected between the optical path port PC through a 1×N optical switch; wherein the i-th port of the 1×N optical switch is coupled with the second port of the optical coupler i, and the common port of the 1×N optical switch is coupled with the optical path port PC, i∈[1,N].

[0077] The splitting ratio of the optical coupler is arbitrary. Because there is an additional optical switch introduced on the optical path from PC to QN, in order to ensure that the insertion loss of the optical path from PN to QN is similar to that of the optical path from PC to QN, an asymmetric splitting ratio is usually used, for example, PC path is divided by 60%, PN path is divided by 40%, and the insertion loss of the optical path from PC to QN after the introduction of the optical switch is similar to that of the optical path from PN to QN.

[0078] When the 1xN optical switch is switched to the kth port of itself and the common port of itself, the optical path of the optical path port Pk and the optical path of the optical path port Pc are both connected to the optical path port Qk.

[0079] The patent embodiment provides a protection optical switch device, which can realize quick switching of the function of the device to the redundant device when part of the device fails by N-to-1 backup of the optical data exchange or computing cluster same-level device, saves the fault recovery time, and saves the cluster device resources.

[0080] Compared with the prior art, when the optical data exchange or computing cluster fails, the manual replacement of the fiber connection is generally used to replace the failed device or port, which not only reduces the utilization efficiency of the device itself, but also consumes human resources and has huge time cost.

[0081] The protection optical switch device has an optical coupler inserted in the optical path, which introduces an insertion loss of at least 3 dB, which may cause the optical path insertion loss to exceed the power budget of the existing optical module. The splitting ratio of the optical coupler of the PN-to-QN optical path can be increased, and an optical amplifier is added at the PC port to compensate for the power loss of the PC-to-QN port. For example, the splitting ratio of the optical coupler of the PN-to-QN optical path is asymmetrically distributed, so that the loss between PN and QN is small, and the large insertion loss between PC and QN is compensated by the optical amplifier. An optical amplifier is added at the PC port to compensate for the additional loss caused by the low splitting ratio of the optical coupler and the additional loss caused by the optical switch. In this way, the insertion of the backup optical switch does not significantly affect the power budget of the optical module interconnecting different levels of optical data exchange and computing devices.

[0082] As shown in Figure 2 Another improved backup optical switch device is provided in combination with the embodiment of the application, and an optical amplifier is connected in series between the common port of the 1xN optical switch and the optical path port PC. The optical amplifier here does not simply amplify the optical signal from one of the optical path ports Q1-QN split by the optical coupler, and the key use method of the optical amplifier in the improved structure shown in Figure 2 will be further emphasized in the method part of the subsequent extension embodiment of the application.

[0083] As shown in Figure 3 The fast recovery device includes one or more protection optical switch devices, specifically including protection optical switch device 1, protection optical switch device 2,..., and protection optical switch device M, wherein the optical switch in each protection optical switch device is uniformly managed by a management device.

[0084] The output end of the first-level device library includes at least M sets of transmission ports, and each set of transmission ports includes at most N transmission ports.

[0085] Each set of transmission ports is carried by an optical path port Q1-QN of a protection optical switch device, while the optical path ports P1-PN of the corresponding protection optical switch device are respectively assigned to the second layer devices 1-N.

[0086] Each optical path port PC of each protection optical switch device is linked to a redundant device, which has the functional attribute of each second layer device; each second layer device and the redundant device also establish a data control signal interaction link with the management device.

[0087] For a topology level of the same network topology level of N identical optical data exchange or computing devices, a same redundant optical data exchange or computing device can be added, and a set of N:1 backup optical switches are used for redundant backup. Assuming that the type of device (i.e., the second layer device in Figure 3 ), there are M optical ports, M N-port protection optical switch devices or no less than N:1 backup optical switches are required. The mth optical port of the redundant device of the N parallel devices is connected to the PC port of the mth N:1 backup optical switch; the Mth optical port (where m is a natural number from 1 to M) of the Nth device (where N is a natural number from 1 to N) is connected to the PN port of the mth N:1 backup optical switch; the QN port of the mth N:1 backup optical switch is connected to the kth optical port (where k is a natural number from 1 to K, K=MxN) of the other level device that should originally be connected to the mth port of the Nth device. It needs to be emphasized that the K optical ports of the other level device do not necessarily belong to the same device. Here, only the total optical ports of the other level device that are connected to the K=MxN optical ports of the N devices to be protected are represented, and the specific allocation manner is irrelevant to the N devices and the M N:1 backup optical switches of the current level.

[0088] When the N devices are working normally, the redundant device ports are closed, and the switch states of all the backup optical switches are in any state. If one of the N devices fails, the management device transfers the data and state of the failed device to the redundant device. Assuming that the failed device is the Nth device, the switch states of all the M N:1 backup optical switches are switched to N, and the optical ports of the Nth failed device are closed, so as to realize the replacement of the Nth failed device by the redundant device.

[0089] In a most basic use process, reference is made to the fast recovery device architecture shown in Figure 2 . It needs to be emphasized that the following method process is not limited to the fast recovery device architecture shown in Figure 2 . In feasible scheme scenarios, it is also applicable to the following Figure 4 and Figure 6In the quick recovery device architecture shown), the management device is configured to acquire the alarm message originated from any one of the second layer devices 1-second layer device N, and further comprises:

[0090] The management device confirms the identification number m of the second layer device sending the alarm message, and according to the pre-established mapping relationship between the identification number m of the second layer device and the optical path port Pm in each protection optical switch device, the management device controls the optical switch in each protection optical switch device, so that the common port is connected with the optical path port Pm, thereby the optical signal of the first layer device bank is copied to the second layer redundant device through the optical switch in each protection optical switch device and the optical coupler m, and the second layer redundant device replaces the role of the second layer device m and continues to interact with the first layer device bank.

[0091] The following process is based on the above-mentioned Figure 2 The protection optical switch device comprising the optical amplifier improved structure, when further applied in the quick recovery device architecture similar Figure 3 The process instance content further includes:

[0092] Any one of the second layer devices 1-second layer device N, when confirming that the communication between itself and the first layer device bank is in a temporary followable state, the second layer device x sends an optical path detection request to the management device. Wherein, the temporary followable state specifically includes: the first layer device bank and the second layer device are in a heartbeat message maintaining link state; or, the first layer device bank and the second layer device are in a continuous data sending state in a non-encrypted state; or, the first layer device bank and the second layer device are in an inherent script execution state, wherein the execution instruction of the script is pre-known by the management device. The meaning expressed by the temporary followable state here is that the communication content between the current first layer device and the second layer device can be predicted in advance, so that the corresponding second layer redundant device can perform matching verification with the second layer device x even when it is temporarily connected to the first layer device.

[0093] The management device confirms whether the current Layer 2 redundant device is in a replacement state. If the Layer 2 redundant device is in a replacement state, it returns a response message indicating that the Layer 2 redundant device is in a replacement state to the Layer 2 device x. The replacement state indicates that a Layer 2 device in the current system architecture has failed, or that the optical path connection between the corresponding Layer 2 device and the Layer 1 device library has failed, and that the management device has controlled the associated protective optical switch to switch the optical path link between the failed Layer 2 device and the Layer 1 device library to the optical path link between the Layer 2 redundant device and the Layer 1 device library. Therefore, the replacement state here can also be understood as the Layer 2 redundant device replacing a failed Layer 2 device.

[0094] If the second-layer redundant device is in an idle state, a response message for the second-layer redundant device to enter temporary follow is returned. At the same time as sending the response message, the common port in each protection optical switch device is controlled to be turned on with the optical path port Qx assigned to the second-layer device x in the optical switch device. Thus, by comparing the optical signals interacted between the second-layer redundant device and the second-layer device x, the management device confirms the stability of the state of the second-layer device x. Here, x belongs to [1, N].

[0095] The aforementioned method of comparing the optical signals of the second-layer redundant device with those of the second-layer device x, allowing the management device to verify the stability of the inherent optical path of the second-layer device x, can be implemented in the following way:

[0096] The management equipment controls the operating power of the optical amplifiers in each protective optical switch. With the goal of ensuring the same optical power received by the second-layer device x and the second-layer redundant device, it verifies whether the adjustment of the operating power of the optical amplifiers in each protective optical switch is within a preset range, thereby confirming the stability of the operating status of each optical path port. In practical implementation, the above method is not solely for confirming the stability of the operating status of each optical path port; it also includes handling various faults, such as those requiring a switch to backup. As an optional solution, the implementation can also incorporate error detection and alarm mechanisms, going beyond just detecting operating power.

[0097] As above Figure 3 The fast recovery device architecture shown in the present invention also provides, as in the embodiments of the present invention, the following: Figure 4The fast recovery device architecture example includes one or more protective optical switch devices, specifically a north-facing protective optical switch array composed of U protective optical switch devices and a south-facing protective optical switch array composed of V protective optical switch devices. N third-layer devices and one third-layer redundant device are connected in series between the north-facing and south-facing protective optical switch arrays. Each third-layer device has V optical interfaces facing the south-facing protective optical switch array and U optical interfaces facing the north-facing protective optical switch array. Here, the first-layer device in this embodiment is also described as the north-facing device of the third-layer device, and the corresponding second-layer device is also described as the south-facing device of the third-layer device; the protective optical switch array of the first-layer device is also described as N:1 north-facing protective switch devices × U of the third-layer device; the protective optical switch array of the second-layer device is also described as N:1 south-facing protective switch devices × V of the third-layer device.

[0098] The northbound and southbound protection optical switch arrays each have an optical interface on one side, in addition to the optical interface coupled to the Layer 3 devices, used to connect to the northbound and southbound device libraries, respectively. The optical switches in each protection optical switch device are managed uniformly by the Layer 3 device management equipment, and each Layer 3 device and its redundant components also establish data exchange channels with the management equipment. In practical implementation, the terms "northbound" and "southbound" are more often used to refer to the northbound or southbound direction of a specific Layer 3 device.

[0099] In optical data switching or computing equipment clusters, devices at a certain layer often have both northbound and southbound connection ports. When introducing redundant devices via backup optical switches, backup optical switches for both the northbound and southbound connection ports need to be configured simultaneously. Figure 4 As shown, this layer has N optical data switching or computing devices, each with U northbound optical ports and V southbound optical ports. A redundant device is configured, with U backup optical switches (at least N:1) for the northbound direction and V backup optical switches (at least N:1) for the southbound direction. When the Nth device fails, the N:1 backup optical switches for both the northbound and southbound directions simultaneously switch to the Nth port to replace the failed device.

[0100] Apart from Figure 4 The fast recovery device architecture example shown is an optical data exchange or computing equipment cluster in which devices at different layers can independently choose whether to configure redundant devices and their backup optical switches. Figure 5The first layer has N devices, each device has V south-facing optical ports, the second layer has L devices, each device has U north-facing ports, wherein N*V=L*U. The first layer N devices are configured with a redundant device, and the V port number is not less than N:1 south backup optical switch, which provides a failure protection mechanism for the first layer N devices. The second layer L devices are configured with a redundant device, and the U port number is not less than L:1 north backup optical switch, which provides a failure protection mechanism for the second layer L devices. The failure protection mechanisms of two adjacent layers can coexist independently, or can exist independently according to the needs.

[0101] As shown in the rapid recovery device architecture shown above Figure 3 , the embodiment of the application also provides a rapid recovery device architecture example (i.e. 256 card DGX H100 Super POD 8:1 backup scheme) as Figure 6 indicated, the second layer device is specifically computer rack 1, computer rack 2,..., computer rack 8; the second layer redundant device is specifically redundant computer rack; the first layer device library is specifically 8 leaf computing switches; and 32 backup optical switches are matched, wherein N is 8, and the rapid recovery device comprises:

[0102] The j'th port of the k'th H100 of the redundant computer rack is connected to the PC port of the 4k'+j'-4'th backup optical switch; wherein each H100 comprises a group of 8 optical interfaces, and is numbered from 1 to 8;

[0103] The j'th port of the k'th H100 of the i'th computer rack is connected to the Pi' port of the 4k'+j'-4'th backup optical switch;

[0104] The Qi' port of the 4k'+j'-4'th backup optical switch is connected to the 4i'+k-4'th port of the j'th leaf computing switch; wherein k' and j' are natural numbers.

[0105] As shown in Figure 7 , for the 1:N backup optical opening device based on embodiment 1 of the embodiment of the application, a further improved scheme is proposed, that is, the function diagram of the N:S backup optical switch of the multi-path backup.

[0106] In some clusters, N:1 backup cannot meet the failure rate requirement, and multi-path backup can also be used, for example, S backup ports backup N ports. As shown in Figure 7 , an NxS optical switch can be used to realize S-way optical port backup for N optical ports, wherein the S ports of the optical switch can be independently configured to connect any N ports.

[0107] Embodiment 2:

[0108] The application also provides a method for fast recovery of partial device failure in an optical data exchange or computing device cluster, and the fast recovery device comprises a protection optical switch device 1, a protection optical switch device 2,..., and a protection optical switch device M, wherein the optical switch in each protection optical switch device is uniformly managed by a management device; the output end of a first layer device library comprises at least M sets of transmission ports, each set of transmission ports comprises at most N transmission ports; each set of transmission ports is carried by the optical path ports Q1-QN of a protection optical switch device, and the optical path ports P1-PN of the corresponding protection optical switch device are respectively allocated to second layer devices 1-2. Figure 8 The method comprises the following steps:

[0109] In step 201, the management device is used to acquire an alarm message originating from any one of the second layer devices 1-2.

[0110] In step 202, the management device confirms the identification number m of the second layer device sending the alarm message, wherein the second layer device corresponding to the identification number m of the second layer device is referred to as the second layer device m.

[0111] In step 203, according to a pre-established mapping relationship between the identification number m of the second layer device and the optical path port Pm in each protection optical switch device, the management device controls the optical switch in each protection optical switch device, so that the common port is in conduction with the optical path port Pm, thereby copying the optical signal of the first layer device library to the second layer redundant device via the optical switch in each protection optical switch device and the optical coupler m.

[0112] In step 204, the second layer redundant device replaces the role of the second layer device m and continues to interact with the first layer device library.

[0113] The present application realizes that when partial device failure occurs, the function of the device can be quickly switched to the redundant device by performing N-to-1 backup on the same level devices in the optical data exchange or computing cluster, thereby saving the fault recovery time and saving the cluster device resources. Compared with the prior art, when the optical data exchange or computing cluster fails, the manual replacement of fiber connection is generally used to replace the failed device or port, which not only reduces the utilization efficiency of the device itself, but also consumes human resources and has a huge time cost.

[0114] As shown in Figure 9 There is also a preferred extended implementation scheme in combination with the embodiments of the application, and the method comprises the following steps:

[0115] In step 301, any second layer device x in the second layer device 1-second layer device N sends a light path detection request to the management device when confirming that the communication between itself and the first layer device library is in a temporarily followable state.

[0116] The temporarily followable state here means that the current communication content between the first layer device and the second layer device can be predicted in advance, so that the corresponding second layer redundant device can be matched and verified with the second layer device x through the corresponding parsed content even when temporarily accessing the first layer device.

[0117] In step 302, the management device confirms whether the current second layer redundant device is in a replacement state, and returns a response message that the second layer redundant device is in a replacement state to the second layer device x if the second layer redundant device is in a replacement state.

[0118] The replacement state indicates that there is a second layer device failure in the current system architecture, or the optical path connection between the corresponding second layer device and the first layer device library has a problem, and the management device has controlled the supporting protection optical switch device to switch the optical path link between the second layer device and the first layer device library to the optical path link between the second layer redundant device and the first layer device library; therefore, the replacement state here can also be understood as the state of the second layer redundant device replacing a second layer device that has failed.

[0119] The temporarily followable state specifically includes that the first layer device library and the second layer device are in a heartbeat packet link maintenance state; or the first layer device library and the second layer device are in a continuous data sending state in a non-encrypted state; or the first layer device library and the second layer device are in an inherent script execution state, wherein the execution instruction of the script is known in advance by the management device.

[0120] In step 303, if the second layer redundant device is in an idle state, a response message is returned that the second layer redundant device enters a temporary follow state, wherein the common port in each protection optical switch device is controlled to be conductive with the optical switch device corresponding to the optical path port Qx allocated to the second layer device x at the same time of sending the response message, so as to confirm the state stability of the second layer device x by the management device through comparison of the optical signals exchanged between the second layer redundant device and the second layer device x.

[0121] The above step 303 involves confirming the state stability of the second layer device x by the management device through comparison of the optical signals exchanged between the second layer redundant device and the second layer device x, and the following detailed implementation possibilities are provided in the embodiment of the present application:

[0122] The management device controls the working power of the optical amplifier in each protection optical switch device, and confirms whether the adjustment of the working power of the optical amplifier in each protection optical switch device is within a preset range, so as to confirm the stability of the working state of each optical path port, in the case that the same optical power accepted by the second layer device x and the second layer redundant device is taken as a target. Figure 2 The protection optical switch device shown in the above embodiment and the method of controlling the working power of the optical amplifier in the protection optical switch device form a closed loop at the method level. In the specific implementation process, the above method is not necessarily simply for confirming the stability of the working state of each optical path port, but also includes various faults, and if it is necessary to switch to a backup, it also needs to be processed in this way. As an optional solution, the implementation process can also be implemented in cooperation with error codes, alarms and other methods, and is not limited to the detection of the working power.

[0123] It should be noted that the information interaction, execution process and the like between the modules and units in the above device and system are based on the same concept as the processing method embodiments of the present application, and the specific content can be referred to the description in the method embodiments of the present application, which will not be described here.

[0124] Those skilled in the art can understand that all or part of the steps in the various methods of the embodiments can be completed by a program instructing the relevant hardware, and the program can be stored in a computer readable storage medium, which can include a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0125] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A rapid recovery device for partial equipment failure in an optical data exchange or computing equipment cluster, characterized in that, The rapid recovery device comprises one or more protection optical switch devices, which comprise: The protection optical switch device has N optical path ports at one end, respectively optical path port Q1, optical path port Q2, …, optical path port QN, and N+1 ports at the other end, respectively optical path port P1, optical path port P2, …, optical path port PN, and optical path port PC; The optical path port Qi and the optical path port Pi are respectively connected in series at the common port of the optical coupler i and the first port of the optical coupler i, and the second port of the optical coupler i is connected to the optical path port PC through a 1×N optical switch; wherein the i-th port of the 1×N optical switch is coupled to the second port of the optical coupler i, the common port of the 1×N optical switch is coupled to the optical path port PC, and i∈[1, N]; When the 1×N optical switch switches to the k-th port of itself and the common port of itself, the optical path of the optical path port Pk and the optical path of the optical path port PC are both connected to the optical path port Qk.

2. The device for fast recovery from partial failure of a cluster of optical data switching or computing devices according to claim 1, wherein, An optical amplifier is further connected in series between the common port of the 1×N optical switch and the optical path port PC.

3. The device for fast recovery of partial device failure in a cluster of optical data switching or computing devices according to claim 1, wherein, The rapid recovery device comprises one or more protection optical switch devices, which comprise a protection optical switch array composed of a protection optical switch device 1, a protection optical switch device 2, …, and a protection optical switch device M, wherein the optical switches in each protection optical switch device are uniformly managed by a management device; The output end of the first layer device library comprises at least M sets of transmission ports, and each set of transmission ports comprises at most N transmission ports; Each set of transmission ports is carried by the optical path ports Q1-QN of one protection optical switch device, and the optical path ports P1-PN of the corresponding protection optical switch devices are respectively allocated to the second layer devices 1-2; Wherein, the optical path port PC of each protection optical switch device is linked to a redundant device, and the redundant device has the functional attributes of each second layer device; each second layer device and the redundant device further establish a data control signal interaction link with the management device.

4. The apparatus for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 3, wherein, The management device is used for obtaining an alarm message from any one of the second layer devices 1-2, and further comprises: The management device confirms the identification number m of the second layer device sending the alarm message; According to the pre-established mapping relationship between the identification number m of the second layer device and the optical path port Pm in each protection optical switch device, the management device controls the optical switch in each protection optical switch device to make the common port and the optical path port Pm conductive, so that the optical signal of the first layer device library is copied into the second layer redundant device through the optical switch and the optical coupler m in each protection optical switch device; The second layer redundant device replaces the role of the second layer device m and continues to interact with the first layer device library.

5. The apparatus for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 3, wherein, Further comprising: Any second layer device x in the second layer devices 1-2 sends an optical path detection request to the management device when confirming that the communication between itself and the first layer device library is in a temporary followable state. The management device confirms whether the current second layer redundancy device is in a replacement state, and if the second layer redundancy device is in the replacement state, returns a response message that the second layer redundancy device is in the replacement state to the second layer device x; If the second layer redundancy device is in an idle state, a response message that the second layer redundancy device enters a temporary following state is returned, wherein, at the same time of sending the response message, the common port in each protection optical switch device is controlled to be conductive with the optical switch port Qx allocated to the second layer device x in the optical switch device, so that the state stability of the second layer device x is confirmed by the management device by comparing the optical signals exchanged between the second layer redundancy device and the second layer device x; wherein x belongs to [1, N].

6. The apparatus for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 5, wherein, The temporary following state specifically includes: The first layer device library and the second layer device are in a heartbeat message maintaining link state; or, The first layer device library and the second layer device are in a continuous data sending state in a non-encrypted state; or, The first layer device library and the second layer device are in an inherent script execution state, wherein the execution instruction of the script is known in advance by the management device.

7. The apparatus for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 5, wherein, The state stability of the second layer device x is confirmed by the management device by comparing the optical signals exchanged between the second layer redundancy device and the second layer device x, and specifically further includes: The management device controls the working power of the optical amplifier in each protection optical switch device, and in the case that the same optical power accepted by the second layer device x and the second layer redundancy device is taken as a target, confirms whether the adjustment of the working power of the optical amplifier in each protection optical switch device is within a preset range, so as to confirm the stability of the working state of each optical path port.

8. The apparatus for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 1, wherein, The fast recovery device includes one or more protection optical switch devices, specifically including a northward protection optical switch array composed of U protection optical switch devices and a southward protection optical switch array composed of V protection optical switch devices; N third layer devices and a third layer redundancy device are connected in series between the northward protection optical switch array and the southward protection optical switch array, wherein the number of optical interfaces of each third layer device towards the southward protection optical switch array side is V, and the number of optical interfaces of each third layer device towards the northward protection optical switch array side is U; The other side optical interfaces of the northward protection optical switch array and the southward protection optical switch array respectively for connecting a northward device library and a southward device library, except for the optical interfaces coupled with the third layer devices; Wherein, the optical switch in each protection optical switch device is uniformly managed by the management device, and each third layer device and the third layer redundancy device also establish a data interaction channel with the management device.

9. The apparatus for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 1, wherein, The second layer device is specifically a computer rack 1, a computer rack 2,..., a computer rack 8; the second layer redundancy device is specifically a redundant computer rack; the first layer device library is specifically 8 leaf computer switches; and 32 backup optical switches are matched, wherein N is 8, and the fast recovery device includes: The j'th port of the k'th H100 of the redundant computer rack is connected to the 4k'+j'-4'th PC port of the backup optical switch; wherein each H100 comprises a set of 8 optical interfaces numbered 1 to 8; The j'th port of the k'th H100 of the i'th computer rack is connected to the 4k'+j'-4'th Pi' port of the backup optical switch; The 4k'+j'-4'th Qi' port of the backup optical switch is connected to the 4i'+k-4'th port of the j'th leaf switch; wherein k' and j' are natural numbers.

10. A method for fast recovery from partial equipment failure in a cluster of optical data switching or computing devices, characterized in that, The fast recovery device comprises a protection optical switch device 1, a protection optical switch device 2,..., and a protection optical switch device M, wherein the optical switches in each protection optical switch device are uniformly managed by a management device; the output of the first layer device library comprises at least M sets of transmission ports, each set of transmission ports comprises at most N transmission ports; each set of transmission ports is carried by the optical path ports Q1-QN of a protection optical switch device, and the optical path ports P1-PN of the corresponding protection optical switch device are respectively allocated to the second layer devices 1-2; the method comprises: The management device is configured to acquire an alarm message from any one of the second layer devices 1-2; The management device confirms the identification number m of the second layer device sending the alarm message; According to a pre-established mapping relationship between the identification number m of the second layer device and the optical path port Pm in each protection optical switch device, the management device controls the optical switch in each protection optical switch device, so that the common port is in conduction with the optical path port Pm, thereby copying the optical signal from the first layer device library to the second layer redundant device through the optical switch in each protection optical switch device and the optical coupler m; The second layer redundant device replaces the role of the second layer device m and continues to interact with the first layer device library.

11. The method for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 10, wherein, The method further comprises: Any one of the second layer devices 1-2, when confirming that the communication between itself and the first layer device library is in a temporary followable state, sends an optical path detection request to the management device; The management device confirms whether the current second layer redundant device is in a replacement state, and if the second layer redundant device is in a replacement state, returns a response message to the second layer device x that the second layer redundant device is in a replacement state; If the second layer redundant device is in an idle state, a response message is returned that the second layer redundant device enters a temporary followable state, wherein at the same time of sending the response message, the common port in each protection optical switch device is controlled to be in conduction with the corresponding optical path port Qx allocated to the second layer device x in the optical switch device, so that the state stability of the second layer device x is confirmed by the management device by comparing the optical signals exchanged between the second layer redundant device and the second layer device x.

12. The method for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 11, wherein, The temporary followable state specifically comprises: The first layer device library and the second layer device are in a heartbeat message maintaining link state; or, The first layer device library and the second layer device are in a non-encrypted continuous data transmission state; or The first layer device library and the second layer device are in an inherent script execution state, wherein the execution instructions of the script are previously known by the management device.

13. The method for fast recovery from partial equipment failure in a cluster of optical data switching or computing equipment according to claim 11, wherein, The state stability of the second layer device x is confirmed by the management device through comparing the optical signals exchanged between the second layer redundant device and the second layer device x, and specifically further comprising: The management device controls the working power of the optical amplifier in each protection optical switch device, and confirms whether the adjustment of the working power of the optical amplifier in each protection optical switch device is within a preset range under the condition that the same optical power accepted by the second layer device x and the second layer redundant device is taken as a target, so as to confirm the stability of the working state of each optical path port.

Images