Two node clusters recovery on a failure

Abstract

Systems and methods for high availability computing systems. Systems and methods include disaster recovery of two-node computing clusters. A method embodiment commences upon identifying a computing cluster having two nodes, the two nodes corresponding to a first node and a second node that each send and receive heartbeat indications periodically while performing storage I/O operations. One or both of the two nodes detect a heartbeat failure between the two nodes, and in response to detecting the heartbeat failure, one or both of the nodes temporarily cease storage I/O operations. A witness node is accessed in an on-demand basis as a result of detecting the heartbeat failure. The witness performs a leadership election operation to provide a leadership lock to only one requestor. The leader then resumes storage I/O operations and performs one or more disaster remediation operations. After remediation, the computing cluster is restored to a configuration having two nodes.

Description

FIELD

This disclosure relates to high availability computing systems, and more particularly to techniques for leadership election for recovery in two-node clusters.

BACKGROUND

Modern computing environments comprise distributed resources (e.g., compute facilities, data storage facilities, networking facilities, etc.) that are accessed by virtualized entities (e.g., virtual machines, executable containers, etc.) to perform various tasks. The foregoing resources of such distributed virtualization environments are managed by physical hardware units known as “nodes”. As an example, a computing node in a high-availability computing setting might correspond to a physical computing appliance having one or more CPUs, multiple solid state drives (SSDs), multiple hard disk drives (HDDs), and multiple network connection interfaces (e.g., for 1 GbE connections, for 10 GbE connections, etc.). Two or more of such computing nodes are often organized in clusters to facilitate ongoing system management (e.g., control, coordination, scaling, etc.). The size or scale of a particular cluster is often quantified by the number of nodes in a particular cluster. A large cluster, for example, might support over one hundred nodes that in turn support as many as several thousands (or more) autonomous virtualized entities (VEs).

Users of such clusters have a high expectation of consistency and availability pertaining to the VEs and/or the data associated with the clusters. As such, an efficient and reliable disaster recovery (DR) capability is desired by users in modern distributed virtualization environments. Such DR capabilities might be facilitated by implementing various data replication functions (e.g., snapshotting, cloning, deduplication, etc.) in a cluster to support certain DR operations and/or scenarios. Specifically, a particular replication policy that describes a replication factor (RF), such as “RF=4”, might be implemented in a cluster to ensure that four replicas of any changed data (e.g., metadata, user data, etc.) is distributed among various storage devices of the cluster to facilitate failover operations in case of some sort of a failure. Specifically, if a node in the cluster fails, the replicated data can be accessed by another available node in the cluster and failover can be accomplished. In some scenarios, DR capabilities implemented in the cluster might enable failback as well. For example, VEs and/or data from a failed node that had been the subject of a DR failover can be restored back to the same node when the failure event or events that precipitated the failover have been remediated.

Unfortunately, implementing such DR capabilities in a cluster that comprises only two nodes can present challenges. Specifically, DR capabilities are often managed at least in part by a leader node that is selected from a plurality of nodes of the cluster. However, in two-node computing clusters, situations can arise in which a leader needs to be elected. For example, if communication between the two nodes fails, each node may consider the other node a failed node and request to become leader so as to invoke certain DR operations (e.g., halt storage input/output operations, etc.) in response to the failure. When there is no other node available in the cluster to process leadership requests, all further operations at the cluster might be suspended until some human intervention is taken to address the failure. In such cases, not only are the users' expectations of high availability and consistency for the cluster jeopardized, but human intervention is also likely required to remedy the failures. What is needed is a technological solution for implementing disaster recovery, including leader selection in two-node computing clusters.

SUMMARY

The present disclosure describes techniques used in systems, methods, and in computer program products for systems and methods for disaster recovery of two-node computing clusters, which techniques advance the relevant technologies to address technological issues with legacy approaches. Certain embodiments are directed to technological solutions for implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters.

The disclosed embodiments modify and improve over legacy approaches. In particular, the herein-disclosed techniques provide technical solutions that address the technical problems attendant to implementing disaster recovery capabilities in two-node computing clusters. Such technical solutions relate to improvements in computer functionality. Various applications of the herein-disclosed improvements in computer functionality serve to reduce the demand for computer memory, reduce the demand for computer processing power, reduce network bandwidth use, and reduce the demand for inter-component communication. Some embodiments disclosed herein use techniques to improve the functioning of multiple systems within the disclosed environments, and some embodiments advance peripheral technical fields as well. As specific examples, use of the disclosed computer equipment, networking equipment, and constituent devices within the shown environments as described herein and as depicted in the figures provide advances in the technical field of computing cluster management as well as advances in various technical fields related to distributed storage systems.

Further details of aspects, objectives, and advantages of the technological embodiments are described herein and in the drawings and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described below are for illustration purposes only. The drawings are not intended to limit the scope of the present disclosure.

FIG. 1 illustrates a computing environment in which embodiments of the present disclosure can be implemented.

FIG. 2 depicts an arbiter-based disaster recovery technique as implemented in systems that facilitate disaster recovery of two-node computing clusters, according to an embodiment.

FIG. 3 presents a block diagram of a system for disaster recovery of two-node computing clusters, according to an embodiment.

FIG. 4 presents a leader selection protocol as implemented in systems that facilitate disaster recovery of two-node computing clusters, according to an embodiment.

FIG. 5A, FIG. 5B, FIG. 5C, and FIG. 5D depict disaster recovery scenarios as performed in systems that facilitate disaster recovery of two-node computing clusters, according to an embodiment.

FIG. 6A and FIG. 6B depict a set of system components as arrangements of computing modules that are interconnected so as to implement certain of the herein-disclosed embodiments.

FIG. 7A, FIG. 7B, and FIG. 7C depict virtualized controller architectures comprising collections of interconnected components suitable for implementing embodiments of the present disclosure and/or for use in the herein-described environments.

DETAILED DESCRIPTION

Embodiments in accordance with the present disclosure address the problem of implementing disaster recovery capabilities in two-node computing clusters. Some embodiments are directed to approaches for implementing a leader selection protocol for two-node computing clusters. The accompanying figures and discussions herein present example environments, systems, methods, and computer program products for disaster recovery of two-node computing clusters.

Overview

Disclosed herein are techniques for implementing a disaster recovery framework in a two-node cluster. The framework comprises a protocol for on-demand creation of an arbiter to facilitate leader selection and other DR operations in the event of some failure event. In certain embodiments, the nodes in the two-node cluster are configured in accordance with a set of disaster recovery configurations. For example, each node might be configured to comprise multiple physical storage facilities so as to facilitate a certain node-level and/or disk-level replication factor (e.g., RF=4) over the two nodes. A different cluster (e.g., a remote cluster) is capable of hosting an arbiter. The operations of such an arbiter are relatively simple and require relatively little CPU capacity, and thus, the capabilities of such an arbiter can be shared by many clusters (e.g., tens or hundreds of clusters or more). An access technique (e.g., a web service, an IP address, etc.) to communicate with the node of the cluster that hosts the arbiter is registered with the nodes of the two-node cluster. An in-memory copy of the registered access technique is maintained by each node of the two-node cluster. In the event of some failure being detected by one or both of the two nodes of the two-node cluster, the remote cluster can be accessed via the registered access technique to avail the services of the arbiter.

For detection of various types of failures, the two nodes of the two-node cluster monitor each other via periodic heartbeat indications. When a failure associated with one or both nodes is detected, the arbiter is accessed by any of the nodes that are then capable of accessing the arbiter. In cases where communication between the two nodes fails (e.g., a heartbeat is missed for more than one period) but the nodes are still functional, both nodes might access the arbiter. In other cases (e.g., when one node fails), merely one of the nodes might access the arbiter.

In any of the foregoing cases, upon detecting the failure, one or both nodes cease operations to the storage pool, and one or both nodes invoke a protocol to communicate with a witness node that hosts the aforementioned arbiter. The witness facilitates establishment of a single leader node even in the case that both nodes request leadership. In a two-node cluster, a quorum needed to elect a leader can be formed by one of the two nodes of the two-node cluster plus the witness. More specifically, the witness node in combination with at least one of the two nodes of the two-node cluster serves to form the quorum needed for the witness to vote with the at least one of the two nodes of the two-node cluster, while the other node of the two-node cluster is not elected to be the leader.

Responsive to the single leader node being selected, various disaster recovery operations are performed at the operational nodes of the cluster. For example, a node that is not the leader node might immediately halt operations, while a node that is selected as the leader node completes replication operations. The node that is selected as the leader node might initiate a failover (e.g., using replicated data) of the VEs at a failed node. In certain embodiments, the witness is created on demand upon request by a node of the two-node cluster. Such an on-demand arbiter is stopped and gracefully destroyed when the formation of the quorum has served its purpose (e.g., when leader selection protocol operations are completed).

Definitions and Use of Figures

Some of the terms used in this description are defined below for easy reference. The presented terms and their respective definitions are not rigidly restricted to these definitions—a term may be further defined by the term's use within this disclosure. The term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects or designs. Rather, use of the word exemplary is intended to present concepts in a concrete fashion. As used in this application and the appended claims, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or”. That is, unless specified otherwise, or is clear from the context, “X employs A or B” is intended to mean any of the natural inclusive permutations. That is, if X employs A, X employs B, or X employs both A and B, then “X employs A or B” is satisfied under any of the foregoing instances. As used herein, at least one of A or B means at least one of A, or at least one of B, or at least one of both A and B. In other words, this phrase is disjunctive. The articles “a” and “an” as used in this application and the appended claims should generally be construed to mean “one or more” unless specified otherwise or is clear from the context to be directed to a singular form.

Various embodiments are described herein with reference to the figures. It should be noted that the figures are not necessarily drawn to scale and that elements of similar structures or functions are sometimes represented by like reference characters throughout the figures. It should also be noted that the figures are only intended to facilitate the description of the disclosed embodiments—they are not representative of an exhaustive treatment of all possible embodiments, and they are not intended to impute any limitation as to the scope of the claims. In addition, an illustrated embodiment need not portray all aspects or advantages of usage in any particular environment.

An aspect or an advantage described in conjunction with a particular embodiment is not necessarily limited to that embodiment and can be practiced in any other embodiments even if not so illustrated. References throughout this specification to “some embodiments” or “other embodiments” refer to a particular feature, structure, material or characteristic described in connection with the embodiments as being included in at least one embodiment. Thus, the appearance of the phrases “in some embodiments” or “in other embodiments” in various places throughout this specification are not necessarily referring to the same embodiment or embodiments. The disclosed embodiments are not intended to be limiting of the claims.

Descriptions of Example Embodiments

FIG. 1 illustrates a computing environment 100 in which embodiments of the present disclosure can be implemented. As an option, one or more variations of computing environment 100 or any aspect thereof may be implemented in the context of the architecture and functionality of the embodiments described herein.

FIG. 1 illustrates one aspect pertaining to implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters. Specifically, the figure presents a logical depiction of how the herein disclosed techniques can be implemented in a computing environment to display various system metrics associated with the computing environment to facility disaster recover capabilities in two-node computing clusters.

Referring to a steady state mode 152₁, the logical depiction of FIG. 1 illustrates a cluster 102 that comprises two computing nodes (e.g., node 104₁ and node 104₂). A two-node cluster might be implemented as the sole computing facility of a small enterprise, or as a special computing facility (e.g., for a remote office or branch office) of a larger enterprise that has other computing clusters. The nodes of cluster 102 can comprise various virtualized entities or VEs (e.g., user virtual machines or user VMs, virtualized disks or vDisks, virtualized network interface cards or vNICs, executable containers, etc.) that are implemented to execute certain tasks (operation 1) when in steady state mode 152₁ (e.g., normal operation without failure).

To facilitate the implementation of the user VMs and/or execution of the tasks, the nodes access a storage pool 106. The storage pool 106 can comprise storage facilities (e.g., SSDs, HDDs, etc.) local to the nodes and, in some cases, storage facilities (e.g., networked storage) external to the nodes and/or cluster 102. The storage facilities at the storage pool 106 store metadata (e.g., configuration specifications, state information, etc.) associated with the virtualized entities, user data (e.g., files, pictures, videos, etc.) managed by the virtualized entities, and/or other information. In this and other embodiments, the storage pool is a collection of a plurality of storage devices that includes both local storage (e.g., using node-internal storage devices) and any forms of network-accessible storage that are organized to form a contiguous address space that is shared by two or more nodes.

As shown in steady state mode 152₁, node 104₁ and node 104₂ might monitor one another by sending “heartbeat” signals from time to time (e.g., every two seconds) that indicate the node sending the signal is operating as expected.

As earlier mentioned and as shown in steady state mode 152₁, users of computing clusters, including two-node computing clusters (e.g., cluster 102), often have a high expectation of consistency and availability pertaining to the VEs and/or the data associated with the clusters. As such, an efficient and reliable disaster recovery (DR) capability is desired. However, implementing such DR capabilities in a cluster that comprises only two nodes can present challenges. The herein disclosed techniques address such challenges by implementing a disaster recovery framework in the two-node cluster.

As indicated in FIG. 1, the disaster recovery framework at least in part comprises techniques for multi-level data replication 110 at cluster 102. Multi-level data replication is a type of data replication that is performed over various levels (e.g., a node level, a disk level, etc.) of hierarchically related physical components. For example, a node-level data replication will generate and store an instance of node-level replicated data 112 at each node in cluster 102. A node-level data replication might be implemented to replicate the user data of cluster 102 according a node-level replication factor of two (e.g., RF=2) so as to facilitate certain disaster recovery operations at the cluster. As another example, a disk-level data replication will generate and store instances of disk-level replicated data 114 at selected “disks” (e.g., SSDs, HDDs, etc.) associated with the nodes of cluster 102. A disk-level data replication might be implemented to replicate the VE metadata of cluster 102 according to a disk-level replication factor of four (e.g., RF=4) to facilitate certain disaster recovery operations at the cluster.

The aforementioned disaster recovery framework further comprises a protocol for accessing an arbiter to facilitate selection of a leader node that in turn performs DR operations in the event of a failure at the two-node cluster. As such, certain arbiter access information 116 (e.g., is made available by a registration process) to node 104₁ and node 104₂ of cluster 102. As can be observed, an arbiter 122 might be a service managed by a virtual machine (e.g., witness VM 120) operating at a node 104₃. In this case, the arbiter access information 116 might comprise a URI or IP address of the witness VM 120, a username and password for access authentication, and/or other information for accessing the arbiter 122. As shown, node 104₃ as well as the hosted witness VM and its arbiter instance are external to cluster 102. A particular witness VM and arbiter might serve multiple (e.g., 50, 500, 5000, etc.) instances of computing clusters. In some embodiments, the witness VM is accessible via the Internet. The witness VM and its arbiter service can be hosted in a different cluster that is geographically proximal or geographically distal from cluster 102.

Referring to a disaster recovery mode 154₁ of FIG. 1, when a failure associated with one or both nodes of cluster 102 is detected, the arbiter 122 is accessed by any of the nodes that are then capable of accessing the arbiter. For example, node 104₁ and node 104₂ might both detect a failure (operation 2) in response to a heartbeat signal not being received at either node after a certain period of time (e.g., 10 seconds or five consecutive unsuccessful heartbeat pings). This situation might occur when the intra-cluster communication between nodes of cluster 102 has failed. In this case, node 104₁ and node 104₂ will invoke the protocol with the arbiter 122 by issuing a respective leadership request over an external network 130 (e.g., the Internet) to the arbiter 122 (operation 3 and operation 4). In other cases (e.g., one node fails), merely one of the nodes of cluster 102 might access the arbiter 122.

Upon receiving the one or more leadership requests, an atomic operation 124 is executed at arbiter 122 to select a leader node (operation 5). For example, a compare-and-swap operation and/or a semaphore might be implemented at arbiter 122 to select a single leader node in the presence of multiple contending leadership requests. In this case, the leader node is selected based at least in part on at least one outcome of at least one atomic leadership election operation (e.g., atomic operation 124) performed by arbiter 122. Using this approach, the selected leader node might be the node associated with the first received leadership request. The node that is not selected as leader (e.g., node 104₁) then receives a leadership request rejection (operation 6) and halts operations (operation 7) to, for example, avoid data access (e.g., write) conflicts at storage pool 106. In some cases, the node not selected as leader might receive no response, which lack of response serves to indicate that the node was not selected as the leader node.

The selected leader node (e.g., node 104₂) receives an indication of a leadership lock (e.g., leadership request confirmation) (operation 8) which can, for example, enable the leader node to invoke certain DR operations (operation 9). Examples of such DR operations include node operating mode transition operations (e.g., from a steady state mode to a single-node mode), failover operations (e.g., into a failover configuration at the remote cluster or into a failover configuration within the two-node cluster), and failback operations (e.g., into a failback configuration within the two-node cluster). As used herein, a failback configuration is one where at least the same number of nodes of the cluster are again operational after a failover operation. More specifically, a failback configuration is one where the two nodes of the cluster are again operational after one of the two nodes triggered a failover operation.

The disaster recovery capabilities facilitated by the herein disclosed techniques result in improvements in computer functionality that serve to reduce the demand for computer processing power, reduce the demand for computer memory and data storage, reduce network bandwidth use, and reduce the demand for inter-component communication, in two-node computing clusters. Specifically, by providing disaster recovery capabilities in computing clusters having merely two nodes, applications of the herein disclosed techniques satisfy the data consistency and availability expectations of users without requiring additional computing resources—such as additional nodes, each with compute, storage and networking facilities—that could be used to support such DR capabilities, but that might not be otherwise used (e.g., for tasks unrelated to DR).

One embodiment of techniques for arbiter-based disaster recovery is disclosed in further detail as follows.

FIG. 2 depicts an arbiter-based disaster recovery technique 200 as implemented in systems that facilitate disaster recovery of two-node computing clusters. As an option, one or more variations of arbiter-based disaster recovery technique 200 or any aspect thereof may be implemented in the context of the architecture and functionality of the embodiments described herein. The arbiter-based disaster recovery technique 200 or any aspect thereof may be implemented in any environment.

FIG. 2 illustrates one aspect pertaining to implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters. Specifically, the figure presents one embodiment of certain steps and/or operations that facilitate accessing an arbiter to select a single leader node from the nodes of a two-node cluster so as to perform various disaster recovery operations to remediate failures at the cluster. As shown, portions of the steps and/or operations can correspond to a steady state mode 152₂ or a disaster recovery mode 154₂.

The arbiter-based disaster recovery technique 200 can commence by configuring a computing cluster that comprises two nodes to operate in a distributed virtualized environment (step 210). The configuration of the nodes and/or cluster might specify certain parameters related to the DR capabilities and/or other aspects of the cluster, such as parameters pertaining to a node-level replication factor (e.g., RF=2), a disk-level replication factor (e.g., RF=4), a resource usage level (e.g., a maximum usage level of 40%), a recovery point objective (e.g., 6 hours), a network latency limit (e.g., 500 ms), a number of SSDs (e.g., two per node), and/or a number of HDDs (e.g., two per node). In steady state mode 152₂, various computing tasks are continuously performed at virtualized entities operating at the nodes (step 220). Such computing tasks can include data replication tasks (e.g., in accordance with a data replication policy) to facilitate certain DR operations and/or other tasks.

Detecting one or more failures (e.g., failure event 202) associated with one or both nodes in the computing cluster can initiate the disaster recovery mode 154₂ (step 230). For example, failure event 202 might be associated with a node failure, an intra-cluster communication failure, an external communication failure, an arbiter (or witness VM) access failure, and/or any other failure and/or combination thereof. In response to detecting the one or more failures, a protocol is invoked to communicate with the arbiter to establish (e.g., select) a leader node from one of the two nodes of the cluster (step 240). As earlier mentioned, an arbiter service managed by a witness VM might execute one or more atomic operations to select a leader node in the presence of multiple contending leadership requests from the nodes.

When the leader node is selected, various disaster recovery operations based at least in part on the detected failures are performed (step 250). The particular disaster recovery operations performed are selected, for example, based at least in part on the cluster and/or node configuration, the policies associated with the cluster and/or users of the cluster, the then-current state and/or operating mode of the nodes, and/or other information. Until the failure is remediated (see “No” path of decision 260), the disaster recovery operations (e.g., listening for recovery signal) will continue. When the failure is remediated (see “Yes” path of decision 260), the cluster returns the steady state mode 152₂, where computing tasks are executed at the virtualized entities of the nodes.

One embodiment of a system for implementing the arbiter-based disaster recovery technique 200 and/or other herein disclosed techniques is disclosed as follows.

FIG. 3 presents a block diagram of a system 300 for disaster recovery of two-node computing clusters. As an option, one or more variations of system 300 or any aspect thereof may be implemented in the context of the architecture and functionality of the embodiments described herein. The system 300 or any aspect thereof may be implemented in any environment.

FIG. 3 illustrates one aspect pertaining to implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters. Specifically, the figure is being presented to show one embodiment of certain representative components and associated data flows that describe how the herein disclosed techniques might be implemented in a modern computing system. The components and data flows shown in FIG. 3 present one partitioning and associated data manipulation approach. The specific example shown is purely exemplary, and other subsystems, data structures, and/or partitioning are reasonable.

As shown in FIG. 3, the system 300 comprises a cluster 102 comprising two nodes (e.g., node 104₁ and node 104₂) that have multiple tiers of storage in a storage pool 106. Each node can be associated with one server, multiple servers, or portions of a server. The nodes can be associated (e.g., logically and/or physically) with the clusters. As shown, the multiple tiers of storage include instances of local storage (e.g., local storage 310₁ and local storage 310₂). For example, the local storage can be within or directly attached to a server and/or appliance associated with the nodes. Such local storage can include solid state drives (SSDs), hard disk drives (HDDs), and/or other storage devices. In some embodiments, the multiple tiers of storage can include storage that is accessible through an external network 130, such as a networked storage (e.g., a storage area network or SAN, network attached storage or NAS, etc.).

According to the disaster recovery framework of the herein disclosed techniques, various node-level replicated data (e.g., user data) and disk-level replicated data (e.g., metadata) are distributed over the storage pool 106 to facilitate certain disaster recovery operations. A node 104₃ external to cluster 102 and having an associated instance of local storage 310₃ is also shown in system 300. Furthermore, cluster 102 is associated with a failure domain 320₁ whereas node 104₃ is associated with a failure domain 320₂. As used herein, a failure domain or availability domain is a logical collection of hardware components (e.g., nodes, switches, racks, etc.) that are affected by failures within the collection. As an example, a failure domain might comprise a single physical node appliance or a rack of node appliances. The separate failure domains of cluster 102 and node 104₃ indicate that a failure endemic to cluster 102 will not affect node 104₃.

As shown, any of the nodes of system 300 can implement one or more virtualized entities, such as virtual machines (e.g., VM 306₁₁, VM 306_1M, VM 306₂₁, . . . , VM 306_2N, witness VM 120) and/or executable containers. The VMs can be characterized as software-based computing “machines” implemented in a hypervisor-assisted virtualization environment that emulates the underlying hardware resources (e.g., CPU, memory, etc.) of the nodes. For example, multiple VMs can operate on one physical machine (e.g., node host computer) running a single host operating system, while the VMs run multiple applications on various respective guest operating systems. Such flexibility can be facilitated at least in part by one of a variety of heterogeneous hypervisors (e.g., hypervisor-E 307, hypervisor-A 308, or hypervisor-X 309), which hypervisors are logically located between the various guest operating systems of the VMs and the host operating system of the physical infrastructure (e.g., the nodes). Any controller VM can communicate with any hypervisor of any hypervisor type over a respective hypervisor-agnostic communication channel (e.g., hypervisor-agnostic communication channel 311₁, hypervisor-agnostic communication channel 311₂). Such a hypervisor-agnostic communication channel can be used to facilitate halting of a VM and/or for invoking other DR operations. As used herein, a hypervisor-agnostic communication channel is a mechanism whereby a controller VM can issue a generic command (e.g., “halt”) that is translated into a hypervisor-specific command. The communication channel can operate by invocation of a hypervisor-specific call interface such as an application programming interface, and/or can operate over a packet switching network.

As an alternative to and/or in addition to VMs being implemented in the nodes, executable containers may be implemented in the nodes. The executable containers comprise groups of processes and/or resources (e.g., memory, CPU, disk, etc.) that are isolated from the node host computer and/or other containers. Such executable containers directly interface with the kernel of the host operating system without, in most cases, a hypervisor layer. This lightweight implementation can facilitate efficient distribution of certain software components, such as applications or services (e.g., micro-services). Any node of system 300 can implement both a hypervisor-assisted virtualization environment and a container virtualization environment for various purposes.

Furthermore, any node in system 300 can implement a virtualized controller to facilitate, at least in part, access to storage facilities (e.g., storage pool 106, local storage 310₃, networked storage, etc.) by the VMs and/or the executable containers operating at the node. As used in these embodiments, a virtualized controller is a collection of software instructions that serve to abstract details of underlying hardware or software components from one or more higher-level processing entities. A virtualized controller can be implemented as a virtual machine, as an executable container (e.g., a Docker container), or within a layer (e.g., such as a layer in a hypervisor). As can be observed in system 300, an instance of a virtual machine (e.g., controller VM 304₁, controller VM 304₂, and controller VM 304₃) at each node is used as a virtualized controller to, at least in part, manage storage and I/O (input/output or IO) activities at the node. Controller VM 304₁ of node 104₁ and controller VM 304₂ of node 104₂ also facilitate the intra-cluster communications 332 (e.g., heartbeat signaling).

Furthermore, the controller VMs of the nodes in system 300 interact using external communications (e.g., external communications 334₁ and external communications 334₂) over external network 130. For example, such external communications might be used to issue leadership requests from the nodes of cluster 102 to an arbiter 122 managed by witness VM 120 operating at node 104₃, and receive leadership request responses (e.g., a leadership lock) issued from arbiter 122 to node 104₁ and/or node 104₂ of cluster 102.

The virtualized entities at the nodes of system 300 can interface with the controller VM of the node through a respective hypervisor. In such cases, the controller VM is not formed as part of specific implementations of a given hypervisor. Instead, the controller VM can run as a virtual machine above the hypervisor at the various nodes. When the controller VMs run above the hypervisors, varying virtual machine architectures and/or hypervisors can operate with the system 300. For example, a hypervisor (e.g., hypervisor-E 307) at one node might correspond to software from a first vendor (e.g., VMware), and a hypervisor (e.g., hypervisor-A 308) at another node might correspond to a second software vendor (e.g., Nutanix). As another virtualized controller implementation example, executable containers (e.g., Docker containers) can be used to implement a virtualized controller in an operating system virtualization environment at a given node. In this case, for example, the virtualized entities at a particular node can interface with a controller container through a hypervisor and/or the kernel of the host operating system of the node.

As earlier mentioned, arbiter 122 can be implemented in witness VM 120 operating at node 104₃. The arbiter can comprise a set of programming objects and/or instructions that collectively (e.g., as a service, web service, micro-service, etc.) provide certain functionality to facilitate the herein disclosed techniques. Specifically, for example, the arbiter 122 might execute one or more atomic operations (e.g., atomic operation 124) to select a leader node in the presence of multiple leadership requests. Such leadership requests might be issued by node 104₁ and/or node 104₂ of cluster 102 in response to a failure (e.g., failure of intra-cluster communications 332) detected at the cluster. Access to the arbiter 122 by the nodes of cluster 102 might be facilitated by arbiter access information 116 (e.g., IP address, username and password, etc.) stored at storage pool 106. As can be observed, certain information stored in local storage 310₃ in accordance with a leader selection data structure 324 can be accessed by arbiter 122 to select and/or identify the leader node. For example, the arbiter might perform a compare-and-swap operation on the string value in a “nodeIPaddress” field to set the value to the IP address of the selected leader node. More specifically, the value will be set if and only if the “nodeIPaddress” field is empty. The string value set in the “nodeIPaddress” field and/or other information (e.g., a cluster identifier stored in a “clusterID” field) can be used to issue a leadership lock to a node uniquely identified by the combination of “nodeIPaddress” and “clusterID”.

The foregoing discussions include techniques for invoking a protocol to communicate with the arbiter 122 to establish a leader node (e.g., step 240 of FIG. 2), which techniques are disclosed in further detail as follows.

FIG. 4 presents a leader selection protocol 400 as implemented in systems that facilitate disaster recovery of two-node computing clusters. As an option, one or more variations of leader selection protocol 400 or any aspect thereof may be implemented in the context of the architecture and functionality of the embodiments described herein. The leader selection protocol 400 or any aspect thereof may be implemented in any environment.

FIG. 4 illustrates one aspect pertaining to implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters. Specifically, the figure is being presented to depict a set of high order interactions (e.g., operations, messages, etc.) exhibited by various computing components earlier described that in part comprise one embodiment of a protocol for selecting a leader node from the nodes of a two-node cluster. The particular computing components shown in FIG. 4 are the controller VMs (e.g., controller VM 304₁ and controller VM 304₂) of the two nodes (e.g., node 104₁ and node 104₂) of cluster 102, and the controller VM 304₃ associated with the witness VM 120 of node 104₃. The arbiter 122 earlier mentioned is shown as being managed at witness VM 120. As also shown, a portion of the interactions of leader selection protocol 400 correspond to step 240 as discussed pertaining to FIG. 2.

The leader selection protocol 400 can commence with receiving the credentials (e.g., URL, IP address, access authorization and/or authentication credentials, etc.) of a target witness VM at cluster 102 (message 402). Any controller VM (e.g., controller VM 304₂) associated with the nodes of the cluster can receive the credentials on behalf of the cluster. In some embodiments, as is shown in FIG. 4, the witness VM 120 is not created at the time the credentials are received at cluster 102. Rather, the target witness VM credentials are merely received from controller VM 304₃ on behalf of witness VM 120. The received credentials and/or other information are stored to register the witness VM with the cluster for later use in requesting creation of a witness (operation 404).

At some later point in time, a failure is detected by one or both nodes at cluster 102 (operation 406₁ and operation 406₂). The controller VM 304₁, for example, of node 104₁ issues a first leadership request to the target witness VM which is intercepted at controller VM 304₃ (message 408₁). In response to receiving the first leadership request, the witness VM 120 is created (command message 410) and the request is forwarded to the witness VM 120 (message 412). Using the information (e.g., cluster identifier, node identifier, node IP address, etc.) received in the first leadership request, a first atomic operation to identify a leader node is invoked at arbiter 122 of witness VM 120 (operation 414). A second leadership request is also issued from controller VM 304₂ of node 104₂ (message 408₂) which, in turn, invokes a second atomic operation at arbiter 122 of witness VM 120 (operation 416).

At a later moment in time, the first atomic operation invoked at arbiter 122 is completed (operation 418). For example, the outcome of the first atomic operation indicates that node 104₁ is selected as the leader node. Due to the atomic nature of the selection operation at arbiter 122, node 104₁ was selected without any conflict arising from the second leadership request issued from node 104₂. As a result of the outcome of the first atomic operation, a confirmation of the first leadership request (e.g., a leadership lock) is issued to controller VM 304₁ of node 104₁ (message 420). The outcome resulting from completion of the second atomic operation executed at arbiter 122 (operation 422) precipitates an issuance of a rejection of the second leadership request to controller VM 304₂ of node 104₂ (message 424). In some cases, the witness VM 120 can be destroyed when a leader node is successfully selected (command message 426).

As such, the depicted protocol of FIG. 4, in particular the acts responsive to command message 410 and command message 426, serve to implement an ephemeral witness. An ephemeral witness is a process that is capable of performing atomic operations (e.g., using a semaphore or a compare and swap data value), which process is generated at the direction of another process (e.g., a controller VM). The ephemeral witness is able to self-destroy (operation 423) upon a command or message.

FIG. 5A, FIG. 5B, FIG. 5C, and FIG. 5D depict a set of disaster recovery scenarios as performed in systems that facilitate disaster recovery of two-node computing clusters. As an option, one or more variations of disaster recovery scenarios or any aspect thereof may be implemented in the context of the architecture and functionality of the embodiments described herein. The disaster recovery scenarios or any aspect thereof may be implemented in any environment.

FIG. 5A, FIG. 5B, FIG. 5C, and FIG. 5D illustrate one aspect pertaining to implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters. Specifically, the figures are being presented to illustrate an embodiment of the herein disclosed techniques that facilitates the following scenarios pertaining to a two-node cluster: a steady state scenario 582, a node failure scenario 584, a remote cluster failover scenario 586, a remote cluster failback scenario 588, a node communication failure scenario 590, a node operating mode transition scenario 592, an intra-cluster failover scenario 594, and an intra-cluster failback scenario 596.

As shown in steady state scenario 582 of FIG. 5A, various computing tasks are performed at cluster 102, which is a two-node cluster (operation A). The data (e.g., user data, metadata, etc.) associated with cluster 102 are replicated according to one or more replication policies associated with the cluster (operation B). At least some of the replicated data pertains to a user VM 306₁₁ running at a node of cluster 102 and operated by a user 504. As shown, a witness virtual machine 120 is implemented at remote cluster 502, however the witness could be implemented in any network-accessible computing system.

As shown in node failure scenario 584 of FIG. 5A, a failure is detected at cluster 102 (operation C). As can be observed, the failure in node failure scenario 584 pertains to a failed node 512. Access to any virtualized entities (e.g., user VM 306₁₁) operating on failed node 512 is suspended. The active node 514 in cluster 102 issues a leadership request to the witness VM 120 in accordance with the herein disclosed techniques (operation D). The active node 514 is selected as the leader node (e.g., by an arbiter) by witness VM 120 (operation E) and a leadership lock is received at the active node 514 (operation F). The active node 514 transitions to a single-node mode (operation G). In the single-node mode, node-level replication is disabled, but disk-level replication over the storage facilities available at the active node 514 is maintained. If a sufficient number of storage facilities at the active node 514 is not achieved (e.g. within 60 seconds), the active node 514 may transition to a read-only mode in which data writes are suspended. For example, if one of two SSDs at active node 514 becomes unavailable while the active node is in single-node mode with a disk-level replication factor of two, the active node 514 will transition to a read-only mode until a second SSD becomes available.

As shown in remote cluster failover scenario 586 of FIG. 5B, the then-current leader node (e.g., leader node 516₁) as selected by witness VM 120 can invoke a failover migration of the virtualized entities (e.g., VMs) of the failed node 512 to the remote cluster 502 (operation H). For example, a failover migration of a VM hosted on failed node 512 might be performed to bring up a replica VM 506₁₁ at the remote cluster 502 so that user 504 can continue working.

As shown in remote cluster failback scenario 588 of FIG. 5B, the recovery of the failed node might be detected (e.g., by leader node 516₁) at some moment in time (operation I). With the earlier failure being remediated, a failback from the remote cluster 502 to the recovered node is performed (operation J) and the nodes in cluster 102 return to a steady state operation (operation K). As an example, user 504 can again access an instance of a VM (e.g., user VM 306₁₂) operating at the earlier failed node of cluster 102.

As shown in node communication failure scenario 590 of FIG. 5C illustrates certain operations pertaining to detecting an intra-cluster communication failure (e.g., node-to-node communication failure) at cluster 102 (operation L). As shown, user 504 might be interacting with user VM 306₁₂ at the time of the failure event. In the case of an intra-cluster communication failure, the nodes at cluster 102 might not know the state of the other node so that each will issue a leadership request to witness VM 120 at remote cluster 502 (operation M). According to the herein disclosed techniques, a single leader node is selected at the witness VM 120 even in the presence of the multiple leadership requests (operation N). The selected leader node receives the leadership lock from the witness VM 120 (operation O).

As shown in node operating mode transition scenario 592 of FIG. 5C, the leader node 516₂ at cluster 102, as selected by witness VM 120, transitions to a single-node mode (operation P). The node that did not receive the leadership lock (e.g., the non-leader node) will halt operations and services (operation Q). In this case, for example, user 504 will not have access to user VM 306₁₂.

In some embodiments such as is depicted in FIG. 3, a first node (e.g., node 104₁) hosts a hypervisor of a first type (e.g., hypervisor-E 307) and a second node (e.g., node 104₂) hosts a hypervisor of a second type (e.g., hypervisor-A 308). In such configurations, acts pertaining to halting (e.g., by a node that did not receive the leadership lock), or acts pertaining to invoking DR operations (e.g., initiating a failover) in response to the failure, can be facilitated in a hypervisor-agnostic manner by the controller VM running on the respective node. Specifically, the controller VM on the node that did not receive the leadership lock can issue a hypervisor-agnostic halt command to the hypervisor of the first type, and the controller VM on the node that did receive the leadership lock can issue a hypervisor-agnostic command to the hypervisor of the second type.

As shown in intra-cluster failover scenario 594 of FIG. 5D, the nodes of cluster 102 might be configured to facilitate a failover migration of the virtualized entities (e.g., VMs) of the non-leader node to the leader node 516₂ (operation R). For example, if the nodes are configured to limit node resource usage to a maximum of 40% and maintain a node-level replication factor of at least two, an intra-cluster failover is possible (e.g., the remote cluster 502 comprising the witness VM 120 is not required for the failover). In this case, user 504 can continue working on a replica VM 506₁₂ at the leader node 516₂ that was failed over from the non-leader node.

As shown in intra-cluster failback scenario 596 of FIG. 5D, when the recovery from the intra-cluster communication failure is detected (operation S), a failback from the leader node 516₂ to the non-leader node is performed (operation T) and the nodes in cluster 102 return to a steady state operation (operation U). As an example, user 504 can again access an instance of a VM (e.g., user VM 306₁₃) operating at the earlier halted node of cluster 102.

Other disaster recovery scenarios in two-node computing clusters as facilitated by the herein disclosed techniques are possible. For example, if the external communication between the witness VM and one of the nodes of the two-node cluster fails, an alert is issued to the node unable to access the witness VM. In this case, the cluster is otherwise unaffected, and no user intervention is required. If the witness VM fails, or the external communications from the witness VM to both nodes of the cluster fail, an alert is generated but the cluster is otherwise unaffected. When a complete network failure occurs (e.g., no intra-cluster or external communications), the overall cluster becomes unavailable and user intervention may be necessary to return the cluster to steady state operation.

Additional Embodiments of the Disclosure

Additional Practical Application Examples

FIG. 6A depicts a system 6A00 as an arrangement of computing modules that are interconnected so as to operate cooperatively to implement certain of the herein-disclosed embodiments. This and other embodiments present particular arrangements of elements that, individually and/or as combined, serve to form improved technological processes that address implementing disaster recovery capabilities in two-node computing clusters. The partitioning of system 6A00 is merely illustrative and other partitions are possible. As an option, the system 6A00 may be implemented in the context of the architecture and functionality of the embodiments described herein. Of course, however, the system 6A00 or any operation therein may be carried out in any desired environment.

The system 6A00 comprises at least one processor and at least one memory, the memory serving to store program instructions corresponding to the operations of the system. As shown, an operation can be implemented in whole or in part using program instructions accessible by a module. The modules are connected to a communication path 6A05, and any operation can communicate with other operations over communication path 6A05. The modules of the system can, individually or in combination, perform method operations within system 6A00. Any operations performed within system 6A00 may be performed in any order unless as may be specified in the claims.

The shown embodiment implements a portion of a computer system, presented as system 6A00, comprising one or more computer processors to execute a set of program code instructions (module 6A10) and modules for accessing memory to hold program code instructions to perform: identifying a computing cluster that comprises a first node and a second node (module 6A20); detecting at least one failure associated with the computing cluster (module 6A30); accessing an ephemeral witness in response to detecting the at least one failure, the ephemeral witness to select a leader node, the leader node being selected from the first node or the second node (module 6A40); and performing one or more disaster recovery operations at the computing cluster, the one or more disaster recovery operations being based at least in part on the at least one failure (module 6A50).

Variations of the foregoing may include more or fewer of the shown modules. Certain variations may perform more or fewer (or different) steps and/or certain variations may use data elements in more or in fewer (or different) operations.

Still further, some embodiments include variations in the operations performed, and some embodiments include variations of aspects of the data elements used in the operations.

FIG. 6B depicts a system 6B00 as an arrangement of computing modules that are interconnected so as to operate cooperatively to implement certain of the herein-disclosed embodiments. The partitioning of system 6B00 is merely illustrative and other partitions are possible. As an option, the system 6B00 may be implemented in the context of the architecture and functionality of the embodiments described herein. Of course, however, the system 6B00 or any operation therein may be carried out in any desired environment.

The system 6B00 comprises at least one processor and at least one memory, the memory serving to store program instructions corresponding to the operations of the system. As shown, an operation can be implemented in whole or in part using program instructions accessible by a module. The modules are connected to a communication path 6B05, and any operation can communicate with other operations over communication path 6B05. The modules of the system can, individually or in combination, perform method operations within system 6B00. Any operations performed within system 6B00 may be performed in any order unless as may be specified in the claims.

The shown embodiment implements a portion of a computer system, presented as system 6B00, comprising one or more computer processors to execute a set of program code instructions (module 6B10) and modules for accessing memory to hold program code instructions to perform: identifying a computing cluster that is initially configured as having only two nodes, the two nodes consisting of a first node and a second node that each send and receive heartbeat indications periodically, while performing storage I/O operations, to a storage pool having a contiguous address space that is shared by the two nodes (module 6B20); detecting, by one or both of the two nodes, at least one failure associated with the computing cluster that prevents the heartbeat indications from being received by at least one of the two nodes (module 6B30); in response to detecting the at least one failure, ceasing operations to the storage pool, and initiating a protocol by at least one of the two nodes, the protocol including an atomic leadership election operation that provides a leadership lock to only a single requestor (module 6B40); initiating, by the single requestor having the leadership lock, acts to resume operations with the storage pool to perform one or more disaster recovery operations (module 6B50); and reconfiguring the computing cluster to a failback configuration having only the two nodes (module 6B60).

System Architecture Overview

Additional System Architecture Examples

FIG. 7A depicts a virtualized controller as implemented by the shown virtual machine architecture 7A00. The heretofore-disclosed embodiments, including variations of any virtualized controllers, can be implemented in distributed systems where a plurality of networked-connected devices communicate and coordinate actions using inter-component messaging. Distributed systems are systems of interconnected components that are designed for, or dedicated to, storage operations as well as being designed for, or dedicated to, computing and/or networking operations. Interconnected components in a distributed system can operate cooperatively to achieve a particular objective, such as to provide high performance computing, high performance networking capabilities, and/or high performance storage and/or high capacity storage capabilities. For example, a first set of components of a distributed computing system can coordinate to efficiently use a set of computational or compute resources, while a second set of components of the same distributed storage system can coordinate to efficiently use a set of data storage facilities.

A hyperconverged system coordinates the efficient use of compute and storage resources by and between the components of the distributed system. Adding a hyperconverged unit to a hyperconverged system expands the system in multiple dimensions. As an example, adding a hyperconverged unit to a hyperconverged system can expand the system in the dimension of storage capacity while concurrently expanding the system in the dimension of computing capacity and also in the dimension of networking bandwidth. Components of any of the foregoing distributed systems can comprise physically and/or logically distributed autonomous entities.

Physical and/or logical collections of such autonomous entities can sometimes be referred to as nodes. In some hyperconverged systems, compute and storage resources can be integrated into a unit of a node. Multiple nodes can be interrelated into an array of nodes, which nodes can be grouped into physical groupings (e.g., arrays) and/or into logical groupings or topologies of nodes (e.g., spoke-and-wheel topologies, rings, etc.). Some hyperconverged systems implement certain aspects of virtualization. For example, in a hypervisor-assisted virtualization environment, certain of the autonomous entities of a distributed system can be implemented as virtual machines. As another example, in some virtualization environments, autonomous entities of a distributed system can be implemented as executable containers. In some systems and/or environments, hypervisor-assisted virtualization techniques and operating system virtualization techniques are combined.

As shown, virtual machine architecture 7A00 comprises a collection of interconnected components suitable for implementing embodiments of the present disclosure and/or for use in the herein-described environments. Moreover, virtual machine architecture 7A00 includes a virtual machine instance in configuration 751 that is further described as pertaining to controller virtual machine instance 730. Configuration 751 supports virtual machine instances that are deployed as user virtual machines, or controller virtual machines or both. Such virtual machines interface with a hypervisor (as shown). Some virtual machines include processing of storage I/O (input/output or IO) as received from any or every source within the computing platform. An example implementation of such a virtual machine that processes storage I/O is depicted as 730.

In this and other configurations, a controller virtual machine instance receives block I/O (input/output or IO) storage requests as network file system (NFS) requests in the form of NFS requests 702, and/or internet small computer storage interface (iSCSI) block IO requests in the form of iSCSI requests 703, and/or Samba file system (SMB) requests in the form of SMB requests 704. The controller virtual machine (CVM) instance publishes and responds to an internet protocol (IP) address (e.g., CVM IP address 710). Various forms of input and output (I/O or IO) can be handled by one or more IO control handler functions (e.g., IOCTL handler functions 708) that interface to other functions such as data IO manager functions 714 and/or metadata manager functions 722. As shown, the data IO manager functions can include communication with virtual disk configuration manager 712 and/or can include direct or indirect communication with any of various block IO functions (e.g., NFS IO, iSCSI IO, SMB IO, etc.).

In addition to block IO functions, configuration 751 supports IO of any form (e.g., block IO, streaming IO, packet-based IO, HTTP traffic, etc.) through either or both of a user interface (UI) handler such as UI IO handler 740 and/or through any of a range of application programming interfaces (APIs), possibly through API IO manager 745.

Communications link 715 can be configured to transmit (e.g., send, receive, signal, etc.) any type of communications packets comprising any organization of data items. The data items can comprise a payload data, a destination address (e.g., a destination IP address) and a source address (e.g., a source IP address), and can include various packet processing techniques (e.g., tunneling), encodings (e.g., encryption), and/or formatting of bit fields into fixed-length blocks or into variable length fields used to populate the payload. In some cases, packet characteristics include a version identifier, a packet or payload length, a traffic class, a flow label, etc. In some cases, the payload comprises a data structure that is encoded and/or formatted to fit into byte or word boundaries of the packet.

In some embodiments, hard-wired circuitry may be used in place of, or in combination with, software instructions to implement aspects of the disclosure. Thus, embodiments of the disclosure are not limited to any specific combination of hardware circuitry and/or software. In embodiments, the term “logic” shall mean any combination of software or hardware that is used to implement all or part of the disclosure.

The term “computer readable medium” or “computer usable medium” as used herein refers to any medium that participates in providing instructions to a data processor for execution. Such a medium may take many forms including, but not limited to, non-volatile media and volatile media. Non-volatile media includes any non-volatile storage medium, for example, solid state storage devices (SSDs) or optical or magnetic disks such as hard disk drives (HDDs) or hybrid disk drives, or persistent random access memories (RAPMs) or optical or magnetic media drives such as paper tape or magnetic tape drives. Volatile media includes dynamic memory such as random access memory. As shown, controller virtual machine instance 730 includes content cache manager facility 716 that accesses storage locations, possibly including local dynamic random access memory (DRAM) (e.g., through local memory device access block 718) and/or possibly including accesses to local solid state storage (e.g., through local SSD device access block 720).

Common forms of computer readable media include any non-transitory computer readable medium, for example, floppy disk, flexible disk, hard disk, magnetic tape, or any other magnetic medium; CD-ROM or any other optical medium; punch cards, paper tape, or any other physical medium with patterns of holes; or any RAM, PROM, EPROM, FLASH-EPROM, or any other memory chip or cartridge. Any data can be stored, for example, in any form of external data repository 731, which in turn can be formatted into any one or more storage areas, and which can comprise parameterized storage accessible by a key (e.g., a filename, a table name, a block address, an offset address, etc.). External data repository 731 can store any forms of data, and may comprise a storage area dedicated to storage of metadata pertaining to the stored forms of data. In some cases, metadata can be divided into portions. Such portions and/or cache copies can be stored in the external storage data repository and/or in a local storage area (e.g., in local DRAM areas and/or in local SSD areas). Such local storage can be accessed using functions provided by local metadata storage access block 724. External data repository 731 can be configured using CVM virtual disk controller 726, which can in turn manage any number or any configuration of virtual disks.

Execution of the sequences of instructions to practice certain embodiments of the disclosure are performed by one or more instances of a software instruction processor, or a processing element such as a data processor, or such as a central processing unit (e.g., CPU1, CPU2, . . . , CPUN). According to certain embodiments of the disclosure, two or more instances of configuration 751 can be coupled by communications link 715 (e.g., backplane, LAN, PSTN, wired or wireless network, etc.) and each instance may perform respective portions of sequences of instructions as may be required to practice embodiments of the disclosure.

The shown computing platform 706 is interconnected to the Internet 748 through one or more network interface ports (e.g., network interface port 723₁ and network interface port 723₂). Configuration 751 can be addressed through one or more network interface ports using an IP address. Any operational element within computing platform 706 can perform sending and receiving operations using any of a range of network protocols, possibly including network protocols that send and receive packets (e.g., network protocol packet 721₁ and network protocol packet 721₂).

Computing platform 706 may transmit and receive messages that can be composed of configuration data and/or any other forms of data and/or instructions organized into a data structure (e.g., communications packets). In some cases, the data structure includes program code instructions (e.g., application code) communicated through the Internet 748 and/or through any one or more instances of communications link 715. Received program code may be processed and/or executed by a CPU as it is received and/or program code may be stored in any volatile or non-volatile storage for later execution. Program code can be transmitted via an upload (e.g., an upload from an access device over the Internet 748 to computing platform 706). Further, program code and/or the results of executing program code can be delivered to a particular user via a download (e.g., a download from computing platform 706 over the Internet 748 to an access device).

Configuration 751 is merely one sample configuration. Other configurations or partitions can include further data processors, and/or multiple communications interfaces, and/or multiple storage devices, etc. within a partition. For example, a partition can bound a multi-core processor (e.g., possibly including embedded or collocated memory), or a partition can bound a computing cluster having a plurality of computing elements, any of which computing elements are connected directly or indirectly to a communications link. A first partition can be configured to communicate to a second partition. A particular first partition and a particular second partition can be congruent (e.g., in a processing element array) or can be different (e.g., comprising disjoint sets of components).

A cluster is often embodied as a collection of computing nodes that can communicate between each other through a local area network (e.g., LAN or virtual LAN (VLAN)) or a backplane. Some clusters are characterized by assignment of a particular set of the aforementioned computing nodes to access a shared storage facility that is also configured to communicate over the local area network or backplane. In many cases, the physical bounds of a cluster are defined by a mechanical structure such as a cabinet or such as a chassis or rack that hosts a finite number of mounted-in computing units. A computing unit in a rack can take on a role as a server, or as a storage unit, or as a networking unit, or any combination therefrom. In some cases, a unit in a rack is dedicated to provisioning of power to other units. In some cases, a unit in a rack is dedicated to environmental conditioning functions such as filtering and movement of air through the rack and/or temperature control for the rack. Racks can be combined to form larger clusters. For example, the LAN of a first rack having a quantity of 32 computing nodes can be interfaced with the LAN of a second rack having 16 nodes to form a two-rack cluster of 48 nodes. The former two LANs can be configured as subnets, or can be configured as one VLAN. Multiple clusters can communicate between one module to another over a WAN (e.g., when geographically distal) or a LAN (e.g., when geographically proximal).

A module as used herein can be implemented using any mix of any portions of memory and any extent of hard-wired circuitry including hard-wired circuitry embodied as a data processor. Some embodiments of a module include one or more special-purpose hardware components (e.g., power control, logic, sensors, transducers, etc.). A data processor can be organized to execute a processing entity that is configured to execute as a single process or configured to execute using multiple concurrent processes to perform work. A processing entity can be hardware-based (e.g., involving one or more cores) or software-based, and/or can be formed using a combination of hardware and software that implements logic, and/or can carry out computations and/or processing steps using one or more processes and/or one or more tasks and/or one or more threads or any combination thereof.

Some embodiments of a module include instructions that are stored in a memory for execution so as to facilitate operational and/or performance characteristics pertaining to disaster recovery of two-node computing clusters. In some embodiments, a module may include one or more state machines and/or combinational logic used to implement or facilitate the operational and/or performance characteristics pertaining to disaster recovery of two-node computing clusters.

Various implementations of the data repository comprise storage media organized to hold a series of records or files such that individual records or files are accessed using a name or key (e.g., a primary key or a combination of keys and/or query clauses). Such files or records can be organized into one or more data structures (e.g., data structures used to implement or facilitate aspects of disaster recovery of two-node computing clusters). Such files or records can be brought into and/or stored in volatile or non-volatile memory. More specifically, the occurrence and organization of the foregoing files, records, and data structures improve the way that the computer stores and retrieves data in memory, for example, to improve the way data is accessed when the computer is performing operations pertaining to disaster recovery of two-node computing clusters, and/or for improving the way data is manipulated when performing computerized operations pertaining to implementing a leader selection protocol in a disaster recovery framework for two-node computing clusters.

Further details regarding general approaches to managing data repositories are described in U.S. Pat. No. 8,601,473 titled “ARCHITECTURE FOR MANAGING I/O AND STORAGE FOR A VIRTUALIZATION ENVIRONMENT”, issued on Dec. 3, 2013, which is hereby incorporated by reference in its entirety.

Further details regarding general approaches to managing and maintaining data in data repositories are described in U.S. Pat. No. 8,549,518 titled “METHOD AND SYSTEM FOR IMPLEMENTING A MAINTENANCE SERVICE FOR MANAGING I/O AND STORAGE FOR A VIRTUALIZATION ENVIRONMENT”, issued on Oct. 1, 2013, which is hereby incorporated by reference in its entirety.

FIG. 7B depicts a virtualized controller implemented by containerized architecture 7B00. The containerized architecture comprises a collection of interconnected components suitable for implementing embodiments of the present disclosure and/or for use in the herein-described environments. Moreover, the shown containerized architecture 7B00 includes an executable container instance in configuration 752 that is further described as pertaining to executable container instance 750. Configuration 752 includes an operating system layer (as shown) that performs addressing functions such as providing access to external requestors via an IP address (e.g., “P.Q.R.S”, as shown). Providing access to external requestors can include implementing all or portions of a protocol specification (e.g., “http:”) and possibly handling port-specific functions.

The operating system layer can perform port forwarding to any executable container (e.g., executable container instance 750). An executable container instance can be executed by a processor. Runnable portions of an executable container instance sometimes derive from an executable container image, which in turn might include all, or portions of any of, a Java archive repository (JAR) and/or its contents, and/or a script or scripts and/or a directory of scripts, and/or a virtual machine configuration, and may include any dependencies therefrom. In some cases, a configuration within an executable container might include an image comprising a minimum set of runnable code. Contents of larger libraries and/or code or data that would not be accessed during runtime of the executable container instance can be omitted from the larger library to form a smaller library composed of only the code or data that would be accessed during runtime of the executable container instance. In some cases, start-up time for an executable container instance can be much faster than start-up time for a virtual machine instance, at least inasmuch as the executable container image might be much smaller than a respective virtual machine instance. Furthermore, start-up time for an executable container instance can be much faster than start-up time for a virtual machine instance, at least inasmuch as the executable container image might have many fewer code and/or data initialization steps to perform than a respective virtual machine instance.

An executable container instance (e.g., a Docker container instance) can serve as an instance of an application container or as a controller executable container. Any executable container of any sort can be rooted in a directory system, and can be configured to be accessed by file system commands (e.g., “ls” or “ls -a”, etc.). The executable container might optionally include operating system components 778, however such a separate set of operating system components need not be provided. As an alternative, an executable container can include runnable instance 758, which is built (e.g., through compilation and linking, or just-in-time compilation, etc.) to include all of the library and OS-like functions needed for execution of the runnable instance. In some cases, a runnable instance can be built with a virtual disk configuration manager, any of a variety of data IO management functions, etc. In some cases, a runnable instance includes code for, and access to, container virtual disk controller 776. Such a container virtual disk controller can perform any of the functions that the aforementioned CVM virtual disk controller 726 can perform, yet such a container virtual disk controller does not rely on a hypervisor or any particular operating system so as to perform its range of functions.

In some environments, multiple executable containers can be collocated and/or can share one or more contexts. For example, multiple executable containers that share access to a virtual disk can be assembled into a pod (e.g., a Kubernetes pod). Pods provide sharing mechanisms (e.g., when multiple executable containers are amalgamated into the scope of a pod) as well as isolation mechanisms (e.g., such that the namespace scope of one pod does not share the namespace scope of another pod).

FIG. 7C depicts a virtualized controller implemented by a daemon-assisted containerized architecture 7C00. The containerized architecture comprises a collection of interconnected components suitable for implementing embodiments of the present disclosure and/or for use in the herein-described environments. Moreover, the shown daemon-assisted containerized architecture includes a user executable container instance in configuration 753 that is further described as pertaining to user executable container instance 780. Configuration 753 includes a daemon layer (as shown) that performs certain functions of an operating system.

User executable container instance 780 comprises any number of user containerized functions (e.g., user containerized function1, user containerized function2, . . . , user containerized functionN). Such user containerized functions can execute autonomously, or can be interfaced with or wrapped in a runnable object to create a runnable instance (e.g., runnable instance 758). In some cases, the shown operating system components 778 comprise portions of an operating system, which portions are interfaced with or included in the runnable instance and/or any user containerized functions. In this embodiment of a daemon-assisted containerized architecture, the computing platform 706 might or might not host operating system components other than operating system components 778. More specifically, the shown daemon might or might not host operating system components other than operating system components 778 of user executable container instance 780.

The virtual machine architecture 7A00 of FIG. 7A and/or the containerized architecture 7B00 of FIG. 7B and/or the daemon-assisted containerized architecture 7C00 of FIG. 7C can be used in any combination to implement a distributed platform that contains multiple servers and/or nodes that manage multiple tiers of storage where the tiers of storage might be formed using the shown external data repository 731 and/or any forms of network accessible storage. As such, the multiple tiers of storage may include storage that is accessible over communications link 715. Such network accessible storage may include cloud storage or networked storage (e.g., a SAN or “storage area network”). Unlike prior approaches, the presently-discussed embodiments permit local storage that is within or directly attached to the server or node to be managed as part of a storage pool. Such local storage can include any combinations of the aforementioned SSDs and/or HDDs and/or RAPMs and/or hybrid disk drives. The address spaces of a plurality of storage devices, including both local storage (e.g., using node-internal storage devices) and any forms of network-accessible storage, are collected to form a storage pool having a contiguous address space.

Significant performance advantages can be gained by allowing the virtualization system to access and utilize local (e.g., node-internal) storage. This is because I/O performance is typically much faster when performing access to local storage as compared to performing access to networked storage or cloud storage. This faster performance for locally attached storage can be increased even further by using certain types of optimized local storage devices, such as SSDs or RAPMs, or hybrid HDDs or other types of high-performance storage devices.

In example embodiments, each storage controller exports one or more block devices or NFS or iSCSI targets that appear as disks to user virtual machines or user executable containers. These disks are virtual since they are implemented by the software running inside the storage controllers. Thus, to the user virtual machines or user executable containers, the storage controllers appear to be exporting a clustered storage appliance that contains some disks. User data (including operating system components) in the user virtual machines resides on these virtual disks.

Any one or more of the aforementioned virtual disks (or “vDisks”) can be structured from any one or more of the storage devices in the storage pool. As used herein, the term vDisk refers to a storage abstraction that is exposed by a controller virtual machine or container to be used by another virtual machine or container. In some embodiments, the vDisk is exposed by operation of a storage protocol such as iSCSI or NFS or SMB. In some embodiments, a vDisk is mountable. In some embodiments, a vDisk is mounted as a virtual storage device.

In example embodiments, some or all of the servers or nodes run virtualization software. Such virtualization software might include a hypervisor (e.g., as shown in configuration 751 of FIG. 7A) to manage the interactions between the underlying hardware and user virtual machines or containers that run client software.

Distinct from user virtual machines or user executable containers, a special controller virtual machine (e.g., as depicted by controller virtual machine instance 730) or as a special controller executable container is used to manage certain storage and I/O activities. Such a special controller virtual machine is referred to as a “CVM”, or as a controller executable container, or as a service virtual machine “SVM”, or as a service executable container, or as a “storage controller”. In some embodiments, multiple storage controllers are hosted by multiple nodes. Such storage controllers coordinate within a computing system to form a computing cluster.

The storage controllers are not formed as part of specific implementations of hypervisors. Instead, the storage controllers run above hypervisors on the various nodes and work together to form a distributed system that manages all of the storage resources, including the locally attached storage, the networked storage, and the cloud storage. In example embodiments, the storage controllers run as special virtual machines—above the hypervisors—thus, the approach of using such special virtual machines can be used and implemented within any virtual machine architecture. Furthermore, the storage controllers can be used in conjunction with any hypervisor from any virtualization vendor and/or implemented using any combinations or variations of the aforementioned executable containers in conjunction with any host operating system components.

In the foregoing specification, the disclosure has been described with reference to specific embodiments thereof. It will however be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the disclosure. For example, the above-described process flows are described with reference to a particular ordering of process actions. However, the ordering of many of the described process actions may be changed without affecting the scope or operation of the disclosure. The specification and drawings are to be regarded in an illustrative sense rather than in a restrictive sense.

Claims

1. A method comprising: detecting a failure condition between a first node and a second node in a computing cluster that comprises a storage resource that is accessed by the first and second nodes;electing a new leader using a witness on an on-demand basis to elect a new leader by: issuing a leadership request,creating an instance of the witness in response to receipt of the leadership request, andaccessing the instance of the witness to perform an atomic leadership election operation to form a quorum that includes the instance of the witness, wherein the atomic leadership election operation provides a leadership lock to only a single requestor, the single requestor corresponding to only one of the first or second nodes, and the instance of the witness is destroyed after said atomic leadership election operation has completed; andinitiating, by the single requestor having the leadership lock, a disaster recovery operation for the computing cluster.

2. The method of claim 1, wherein the leadership request is received at an arbiter in the instance of the witness, and the atomic leadership election operation is performed over a data value to establish the leadership lock.

3. The method of claim 1, wherein the atomic leadership election operation is associated with at least one of, a compare-and-swap operation, or a semaphore.

4. The method of claim 1, wherein the witness comprises a witness virtual machine.

5. The method of claim 1, wherein the instance of the witness is in a different computing cluster from the computing cluster for the first and second nodes, and the computing cluster operates within a first failure domain and the instance of the witness operates within a second failure domain.

6. The method of claim 1, wherein the disaster recovery operation comprises at least one of, a node operating mode transition operation, a failover operation, or a failback operation, and the disaster recovery operation specifies at least one of, a node-level replication factor, a disk-level replication factor, a resource usage level, a recovery point objective, a network latency limit, an SSD, or an HDD.

7. The method of claim 6, wherein the failover operation corresponds to a failover to at least one of, a remote cluster, or an active node in the computing cluster and the disaster recovery operation is invoked over a hypervisor-agnostic communication channel.

8. The method of claim 1, wherein the instance of the witness comprises a witness virtual machine and the instance of the witness is in a different computing cluster from the computing cluster for the first and second nodes.

9. A non-transitory computer readable medium having stored thereon a sequence of instructions which, when executed by a processor causes the processor to perform a process comprising: detecting a failure condition between a first node and a second node in a computing cluster that comprises a storage resource that is accessed by the first and second nodes;electing a new leader using a witness created on an on-demand basis by: issuing a leadership request,creating an instance of the witness in response to receipt of the leadership request, andaccessing the instance of the witness to perform an atomic leadership election operation to form a quorum that includes the instance of the witness, wherein the atomic leadership election operation provides a leadership lock to only a single requestor, the single requestor corresponding to only one of the first or second nodes, and the instance of the witness is destroyed after said atomic leadership election operation has completed; andinitiating, by the single requestor having the leadership lock, a disaster recovery operation for the computing cluster.

10. The computer readable medium of claim 9, wherein the leadership request is received at an arbiter in the instance of the witness, and the atomic leadership election operation is performed over a data value to establish the leadership lock.

11. The computer readable medium of claim 9, wherein the atomic leadership election operation is associated with at least one of, a compare-and-swap operation, or a semaphore.

12. The computer readable medium of claim 9, wherein the witness comprises a witness virtual machine.

13. The computer readable medium of claim 9, wherein the instance of the witness comprises a witness virtual machine and the instance of the witness is in a different computing cluster from the computing cluster for the first and second nodes.

14. The computer readable medium of claim 9, wherein the instance of the witness is in a different computing cluster from the computing cluster for the first and second nodes, and the computing cluster operates within a first failure domain and the instance of the witness operates within a second failure domain.

15. The computer readable medium of claim 9, wherein the disaster recovery operation comprises at least one of, a node operating mode transition operation, a failover operation, or a failback operation, and the disaster recovery operation specifies at least one of, a node-level replication factor, a disk-level replication factor, a resource usage level, a recovery point objective, a network latency limit, an SSD, or an HDD.

16. The computer readable medium of claim 15, wherein the failover operation corresponds to a failover to at least one of, a remote cluster, or an active node in the computing cluster and the disaster recovery operation is invoked over a hypervisor-agnostic communication channel.

17. A system for disaster recovery of two-node computing clusters, the system comprising: a storage medium having stored thereon a sequence of instructions; anda processor that executes the sequence of instructions to cause the processor to perform a process comprising, detecting a failure condition between a first node and a second node in a computing cluster that comprises a storage resource that is accessed by the first and second nodes;electing a new leader using a witness on an on-demand basis by: issuing a leadership request,creating an instance of the witness in response to receipt of the leadership request, andaccessing the instance of the witness to perform an atomic leadership election operation is to form a quorum that includes the instance of the witness, wherein the atomic leadership election operation provides a leadership lock to only a single requestor, the single requestor corresponding to only one of the first or second nodes, and the instance of the witness is destroyed after said atomic leadership election operation has completed; andinitiating, by the single requestor having the leadership lock, a disaster recovery operation for the computing cluster.

18. The system of claim 17, wherein the witness comprises a witness virtual machine.

19. The system of claim 17, wherein the leadership request is received at an arbiter in the instance of the witness, and the atomic leadership election operation is performed over a data value to establish the leadership lock.

20. The system of claim 17, wherein the atomic leadership election operation is associated with at least one of, a compare-and-swap operation, or a semaphore.

21. The system of claim 17, wherein the instance of the witness comprises a witness virtual machine and the instance of the witness is in a different computing cluster from the computing cluster for the first and second nodes.

22. The system of claim 17, wherein the instance of the witness is in a different computing cluster from the computing cluster for the first and second nodes, and the computing cluster operates within a first failure domain and the instance of the witness operates within a second failure domain.

23. The system of claim 17, wherein the disaster recovery operation comprises at least one of, a node operating mode transition operation, a failover operation, or a failback operation, and the disaster recovery operation specifies at least one of, a node-level replication factor, a disk-level replication factor, a resource usage level, a recovery point objective, a network latency limit, an SSD, or an HDD.

24. The system of claim 23, wherein the failover operation corresponds to a failover to at least one of, a remote cluster, or an active node in the computing cluster and the disaster recovery operation is invoked over a hypervisor-agnostic communication channel.