SSD auto-recovery

By providing functions such as in-place restart, configurable timeout, automatic snapshot and automatic archiving in the cloud platform, the loss of SSD stored data in the cloud platform during failure or maintenance is solved, and data recovery control and availability are improved.

CN119096230APending Publication Date: 2024-12-06GOOGLE LLC
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202480002297.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-02-24
Filing Date
2024-01-25
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Data stored in local solid-state drives (SSDs) in cloud platforms are easily lost during hardware failure, instance termination or maintenance, and a single physical machine is a single point of failure, resulting in a large number of single-host failures that are difficult to avoid in large-scale cloud platforms.

Method used

Provides methods for managing SSD recovery on cloud platforms, including in-place restart maintenance mode, configurable timeout options, automatic snapshots and automatic archives, and extend stop/start virtual machine functionality to work with local SSD storage devices, and monitoring And manage the recovery process of virtual machines and SSDs.

Benefits of technology

Through these measures, users can better control the recovery of SSD storage devices, reduce the impact of hardware failure or maintenance on data, and ensure the safety and availability of data during maintenance or failure.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119096230A_ABST
    Figure CN119096230A_ABST
Patent Text Reader

Abstract

Aspects of the present disclosure relate to providing a user with more control of SSD storage device recovery, such as providing capabilities and configuration options for a cloud platform to manage SSD recovery. Aspects of the present disclosure may include providing an in-place restart maintenance mode, configurable timeout options for SSD recovery, automatic snapshot triggering, automatic archiving, and / or extension stop / start virtual machine functionality to work with a local SSD storage device.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is a continuation of U.S. patent application No. 18 / 173,853 filed on February 24, 2023, the disclosure of which is hereby incorporated by reference. Background Art

[0003] Cloud platforms can provide users with virtual machines that can access local solid-state drive (SSD) storage devices. For workloads such as databases, local SSDs can have low cost and low latency. However, the data stored on local SSDs is usually not replicated, so user data may be affected by hardware failures, user-initiated instance terminations and / or maintenance. A single physical machine may itself be a single point of failure, and in large-scale cloud platforms with millions of physical servers, a large number of single host failures may occur. If the problem can be detected early enough to migrate the virtual machine and SSD storage device to a different machine, technologies such as live migration can help minimize the impact of hardware failures or maintenance. However, this is not always possible because the physical machine may crash unexpectedly, potentially losing the data stored on the SSD. Summary of the invention

[0004] Various aspects of the present disclosure are directed to providing users with more control over SSD storage device recovery, such as providing capabilities and configuration options for cloud platforms to manage SSD recovery. Various aspects of the present disclosure may include providing a restart-in-place maintenance mode, configurable timeout options for SSD recovery, automatic snapshot triggering, automatic archiving, and / or extending stop / start virtual machine functionality to work with local SSD storage devices.

[0005] One aspect of the present disclosure provides a method for controlling recovery of a local solid-state drive (SSD) storage device, the method comprising: storing, by one or more processors, metadata indicating a physical machine that is hosting a virtual machine, the physical machine including a local SSD storage device; pausing or shutting down, by one or more processors, the virtual machine for maintenance; and restarting, by the one or more processors, the virtual machine on the physical machine based on the metadata after performing the maintenance.

[0006] In one example, the method further includes saving, by one or more processors, the content of the virtual machine on a local SSD storage device based on restarting the virtual machine on the physical machine. In another example, the method further includes performing maintenance on the virtual machine by one or more processors. In yet another example, restarting the virtual machine on the physical machine based on the metadata further includes matching an identifier of the SSD storage device of the physical machine with an instance identifier of the virtual machine.

[0007] In another example, the method further includes: monitoring, by one or more processors, the amount of time of a recovery process used to attempt to recover a local SSD storage device; determining, by one or more processors, that the amount of time has exceeded a predetermined maximum amount of time used to attempt to recover the local SSD storage device; and sending, by one or more processors, an instruction to stop the recovery process.

[0008] In yet another example, the method further includes: monitoring, by one or more processors, the impact on the virtual machine relative to one or more trigger thresholds; determining, by one or more processors, that one or more of the trigger thresholds are satisfied; snapshotting, by one or more processors, the contents of the local SSD storage device to generate a snapshot; and sending, by one or more processors, the snapshot to a persistent storage device. In yet another example, the one or more trigger thresholds include one or more of a speed of an input or output operation slowing down to below a threshold or a rate of data copying, importing, or exporting slowing down to below a threshold. In yet another example, the method further includes exporting, by one or more processors, the snapshot from the persistent storage device to a local SSD storage device or a different local SSD storage device.

[0009] In another example, the method also includes: monitoring, by one or more processors, the impact on the virtual machine relative to one or more trigger thresholds; determining, by one or more processors, that one or more of the trigger thresholds are met; snapshotting, by one or more processors, the contents of a local SSD storage device to generate a snapshot; and archiving, by one or more processors, the snapshot to a cold storage device.

[0010] Another aspect of the present disclosure provides a system, the system comprising: one or more processors; and one or more storage devices, the one or more storage devices being coupled to the one or more processors and storing instructions, the instructions, when executed by the one or more processors, causing the one or more processors to perform operations for controlling recovery of a local solid-state drive (SSD) storage device. The operations include: storing metadata indicating a physical machine that is hosting a virtual machine, the physical machine including a local SSD storage device; pausing or shutting down the virtual machine for maintenance; and restarting the virtual machine on the physical machine based on the metadata after performing the maintenance.

[0011] In one example, restarting the virtual machine on the physical machine based on the metadata further includes matching an identifier of an SSD storage device of the physical machine with an identifier of the virtual machine.

[0012] In another example, the operation further includes: monitoring an amount of time for a recovery process used to attempt to recover the local SSD storage device; determining that the amount of time has exceeded a predetermined maximum amount of time for attempting to recover the local SSD storage device; and sending an instruction to stop the recovery process.

[0013] In yet another example, the operations further include: monitoring the impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the trigger thresholds are satisfied; snapshotting the contents of the local SSD storage device to generate a snapshot; and sending the snapshot to the persistent storage device. In yet another example, the operations further include exporting the snapshot from the persistent storage device to the local SSD storage device or a different local SSD storage device.

[0014] In yet another example, the operations further include: monitoring the impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the trigger thresholds are met; snapshotting the contents of the local SSD storage device to generate a snapshot; and archiving the snapshot to a cold storage device.

[0015] Yet another aspect of the present disclosure provides a non-transitory computer-readable medium for storing instructions, which, when executed by one or more processors, cause the one or more processors to perform operations for controlling recovery of a local solid-state drive (SSD) storage device. The operations include: storing metadata indicating a physical machine that is hosting a virtual machine, the physical machine including a local SSD storage device; pausing or shutting down the virtual machine for maintenance; and restarting the virtual machine on the physical machine based on the metadata after performing the maintenance.

[0016] In one example, restarting the virtual machine on the physical machine based on the metadata further includes matching an identifier of an SSD storage device of the physical machine with an identifier of the virtual machine.

[0017] In another example, the operation further includes: monitoring an amount of time for a recovery process used to attempt to recover the local SSD storage device; determining that the amount of time has exceeded a predetermined maximum amount of time for attempting to recover the local SSD storage device; and sending an instruction to stop the recovery process.

[0018] In yet another example, the operations further include: monitoring the impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the trigger thresholds are satisfied; snapshotting the contents of the local SSD storage device to generate a snapshot; sending the snapshot to a persistent storage device; and importing the snapshot back to the local SSD storage device or a different local SSD storage device.

[0019] In yet another example, the operations further include: monitoring the impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the trigger thresholds are met; snapshotting the contents of the local SSD storage device to generate a snapshot; and archiving the snapshot to a cold storage device. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Depicted is a block diagram of an example SSD auto-recovery system for a cloud-based platform in accordance with aspects of the present disclosure.

[0021] Figure 2 Depicted is a block diagram of an example environment for implementing an SSD auto-recovery system in accordance with aspects of the present disclosure.

[0022] Figure 3 Depicted is a flow diagram of an example process for implementing an in-place restart for one or more virtual machines in accordance with aspects of the present disclosure.

[0023] Figure 4 Depicted is a flow diagram of an example process for implementing a configurable SSD recovery timeout for one or more virtual machines in accordance with aspects of the present disclosure.

[0024] Figure 5 Depicted is a flow diagram of an example process for implementing automatic snapshotting of contents of one or more virtual machines in accordance with aspects of the present disclosure.

[0025] Figure 6 Depicted is a block diagram of an example cloud-based platform to implement an SSD recovery system in accordance with aspects of the present disclosure. DETAILED DESCRIPTION

[0026] Generally disclosed herein is an implementation of a method for controlling recovery of a local solid-state drive (SSD) storage device in a cloud platform. A cloud platform may provide a virtual machine that can access a local SSD storage device on a physical machine hosting the virtual machine. However, data stored on the local SSD may not be replicated, and therefore the data may be affected by hardware failures, instance terminations, and / or maintenance. Therefore, the method allows for more control over local SSD recovery and provides more options for handling SSD data based on planned and / or unplanned maintenance.

[0027] The method includes providing a "restart in place" maintenance mode together with an instance migration / termination option for planned maintenance. If "restart in place" is selected, the virtual machine can be restarted on the same physical machine that hosted the virtual machine before being restarted. The same physical machine can save the contents of its local SSD without data loss. The underlying virtual machine scheduling and / or fleet management system can check the metadata to ensure that the virtual machine is pinned to the physical machine so that the virtual machine is restarted on the same physical machine. The virtual machine scheduling and fleet management system can also prevent other virtual machines from being scheduled on the same physical machine to ensure that the physical machine is not overbooked.

[0028] Users may be provided advance notification and / or signals / APIs to control the timing of maintenance and coordinate application-level fail-over, recovery, and / or rehydration. Users may be notified that a virtual machine will be scheduled to go offline for maintenance. For example, users may be notified of scheduled maintenance a certain amount of days in advance. Based on that notification, users may choose to drain their applications. Additionally, users may be provided a "trigger now" API to immediately trigger scheduled maintenance. Users may be notified once maintenance is complete. Based on that notification, users may rehydrate their applications as needed.

[0029] The method also includes providing a configurable timeout to wait for the local SSD to recover. The configurable timeout can be applicable to both planned and unplanned maintenance. The user can provide the amount of time the user is willing to wait for the SSD to recover before determining that data in the local SSD is lost.

[0030] The method also includes providing automatic snapshots for persistent storage devices. Persistent storage devices (such as persistent data disks) can be associated with virtual machines. Configurable thresholds associated with the impact on virtual machines can be selected to trigger snapshots. Configurable thresholds can include application-level thresholds that must be met by the infrastructure of the virtual machine, such as input / output operations slowing down to below a threshold or transaction rates of data copy / import / export meeting a threshold. A signal that the virtual machine is about to terminate can also trigger a snapshot, such as a notification that maintenance is scheduled to occur. Once a threshold is reached or a signal is received, a snapshot of the contents of a local SSD storage device can be initiated for storage in a persistent storage device. A snapshot of the contents of a local SSD can be sent to a persistent storage device. The virtual machine can continue to run and service traffic, or the virtual machine may be removed from a service pool during the snapshot process due to potential application impacts. After a predetermined period of time or once the threshold in question is no longer met, the data stored in the persistent storage device can be (such as by an application) directed back to the same local SSD or a different local SSD for future virtual machines.

[0031] The method may further include archiving the contents of the local SSD when the virtual machine is to be terminated. When the virtual machine is to be terminated, the contents of the local SSD may be copied to a cold storage device. The capacity of the terminated virtual machine cannot be used again until the archiving is completed.

[0032] The method may also include extending the stop / start virtual machine functionality to work with the local SSD storage device. The stop / start virtual machine functionality may correspond to allowing a user to shut down an instance of a virtual machine for a certain period of time and then resume the instance. The local SSD storage device may be preserved by the "in-place restart" described above, taking automatic snapshots and / or automatic archiving.

[0033] Figure 1 A block diagram of an example SSD automatic recovery system 100 for a cloud-based platform is depicted. A cloud-based platform can provide services that allow provisioning or maintenance of computing resources and / or applications, such as a data center, a cloud environment, and / or a container framework. For example, a cloud-based platform can be used as a service to provide software applications, such as accounting, word processing, inventory tracking, etc. As another example, the platform's infrastructure can be divided in the form of virtual machines or containers on which software applications run.

[0034] The SSD recovery system 100 may be configured to receive input data 102 via a user interface. For example, the SSD recovery system 100 may receive input data 102 as part of a call to an API that exposes the SSD recovery system 100. The SSD recovery system 100 may be implemented on one or more computing devices. Input to the SSD recovery system 100 may also be provided via a storage medium (including a remote storage device connected to one or more computing devices via a network), or as input through a user interface on a client computing device coupled to the SSD recovery system 100.

[0035] The input data 102 may include data associated with a user selection of an in-place restart mode, a configurable timeout, and / or automatic snapshots. The data associated with the user selection may indicate that the user of the cloud platform wants these configurations to be applied to their SSD storage devices.

[0036] The input data 102 may also include data associated with the restart-in-place mode, such as metadata to ensure that the virtual machine is restarted on the same physical machine that hosted the virtual machine before being restarted. For example, the metadata may include options around the maintenance mode, such as restart-in-place, terminate, migrate, timeout selection and / or length, and / or the number of SSD disks and their identifiers to ensure that they are exposed as before. The metadata may also include CPU, memory, IP address, instance identifier, or any other metadata associated with the virtual machine.

[0037] The input data may also include data associated with a configurable timeout, such as the amount of time to attempt local SSD recovery due to planned or unplanned maintenance. For example, the amount of time may range from 0 to multiple days. The amount of time may be different depending on whether the maintenance is planned (such as a scheduled update) or unplanned (such as a machine crash).

[0038] The input data may also include data associated with taking automatic snapshots, such as thresholds for triggering snapshots. For example, the thresholds may include thresholds associated with the speed of input / output operations of the virtual machine and / or the transaction rate of data copy / import / export. The data associated with taking automatic snapshots may also include a signal that the virtual machine is about to terminate, such as a notification about scheduled maintenance. The data associated with taking automatic snapshots may also include whether the snapshot may be stored in a persistent storage device (such as a persistent data disk) or archived in a cold storage device (such as a storage device for infrequently used data).

[0039] Based on the input data 102, the SSD recovery system 100 can be configured to output one or more results related to SSD recovery generated as output data 104. The output data 104 may include instructions associated with an in-place restart mode, a configurable timeout, and / or taking an automatic snapshot. For example, the output data 104 may include instructions for a physical machine to host the same virtual machine when the virtual machine is restarted. As another example, the output data 104 may include instructions for how long to attempt local SSD recovery before stopping recovery. As yet another example, the output data 104 may include instructions for when to trigger taking an automatic snapshot for a virtual machine and where to store the snapshot (such as in a persistent storage device or a cold storage device).

[0040] The SSD recovery system 100 can be configured to provide output data 104 as a set of computer-readable instructions, such as one or more computer programs. The computer program can be written in any type of programming language and according to any programming paradigm (e.g., declarative, procedural, assembled, object-oriented, data-oriented, functional, or imperative). The computer program can be written to perform one or more different functions and operate within a computing environment, for example, on a physical device, on a virtual machine, or across multiple devices. The computer program can also implement the functionality described herein, for example, as performed by a system, engine, module, or model.

[0041] The SSD recovery system 100 may also be configured to forward the output data 104 to one or more other devices configured to translate the output data 104 into an executable program written in a computer programming language and optionally as part of a framework for recovering SSD data. The SSD recovery system 100 may also be configured to send the output data 104 to a storage device for storage and later retrieval. The SSD recovery system 100 may also be configured to send the output data 104 for display, such as on a user display.

[0042] The SSD recovery system 100 may include an in-place restart engine 106. The in-place restart engine 106 may be implemented as one or more computer programs, a specially configured electronic circuit system, or any combination of the foregoing. The in-place restart engine 106 may be configured to restart the virtual machine on the same physical machine on which the virtual machine was hosted before the restart. The in-place restart engine 106 may pause or shut down the virtual machine for planned or unplanned maintenance. Before being paused or shut down, the in-place restart engine 106 may receive metadata for the virtual machine to indicate which physical machine is hosting the virtual machine. Once the maintenance is completed, the in-place restart engine 106 may restart the virtual machine on the same physical machine based on the metadata. Since the same physical machine may include a local SSD storage device, the virtual machine content may be saved.

[0043] The SSD recovery system 100 may also include a timeout engine 108. The timeout engine 108 may be implemented as one or more computer programs, a specially configured electronic circuit system, or any combination of the foregoing. The timeout engine 108 may be configured to stop the local SSD recovery for the physical machine hosting the virtual machine if a predetermined amount of time is exceeded. For example, the predetermined amount of time may be in the range of from 0 (indicating the termination of the virtual machine without performing local SSD recovery) to several days of attempting to recover the data stored on the local SSD associated with the virtual machine. The timeout engine 108 may receive the predetermined amount of time. In response to planned or unplanned maintenance, local SSD recovery may be performed. The timeout engine 108 may monitor how long the local SSD recovery is taking place. If the local SSD recovery exceeds the predetermined amount of time, the timeout engine 108 may stop the local SSD recovery. Then, the virtual machine may be restarted on a different physical machine that may not have access to the local SSD data of the previous physical machine. If the predetermined amount of time is 0, the timeout engine 108 may be configured to restart the virtual machine on a different physical machine without attempting local SSD recovery.

[0044] The SSD recovery system 100 may also include a snapshot engine 110. The snapshot engine 110 may be implemented as one or more computer programs, a specially configured electronic circuit system, or any combination of the foregoing. The snapshot engine 110 may be configured to automatically snapshot a local SSD storage device in response to one or more thresholds. The thresholds that trigger the automatic snapshot may be associated with an impact on the virtual machine, such as an application-level threshold that the infrastructure of the virtual machine must meet. If one or more of the thresholds are met, the snapshot engine 110 may be configured to snapshot the contents of the local SSD storage device. The snapshot engine 110 may be configured to store the snapshot in a persistent storage device or to archive the snapshot in a cold storage device. For snapshots stored in a persistent storage device, after a predetermined amount of time or if the threshold in question is no longer met, the snapshot engine 110 may import the snapshot back to the same local SSD or a different local SSD.

[0045] Figure 2A block diagram of an example environment 200 for implementing an SSD auto-recovery system 218 is depicted. The SSD auto-recovery system 218 may be implemented on one or more devices having one or more processors in one or more locations, such as in a server computing device 202. The client computing devices 204 and the server computing devices 202 may be communicatively coupled to one or more storage devices 206 via a network 208. The storage devices 206 may be a combination of volatile and non-volatile memory and may be located in the same or different physical locations as the computing devices 202, 204. For example, the storage devices 206 may include any type of non-transitory computer-readable medium capable of storing information, such as a hard drive, a solid-state drive, a tape drive, an optical storage device, a memory card, a ROM, a RAM, a DVD, a CD-ROM, a writable memory, and a read-only memory.

[0046] The server computing device 202 may include one or more processors 210 and memory 212. The memory 212 may store information accessible by the processor 210, including instructions 214 that may be executed by the processor 210. The memory 212 may also include data 216 that may be retrieved, manipulated, or stored by the processor 210. The memory 212 may be a type of non-transitory computer-readable medium capable of storing information accessible by the processor 210, such as volatile memory and non-volatile memory. The processor 210 may include one or more central processing units (CPUs), graphics processing units (GPUs), field programmable gate arrays (FPGAs), and / or application specific integrated circuits (ASICs), such as tensor processing units (TPUs).

[0047] Instructions 214 may include one or more instructions that, when executed by processor 210, cause the processor or processors to perform actions defined by instructions 214. Instructions 214 may be stored in an object code format for direct processing by processor 210, or in other formats including interpretable scripts or independent source code modules that are interpreted on demand or compiled in advance. Instructions 214 may include instructions for implementing functions that may be used with Figure 1 The instructions of the SSD automatic recovery system 218 corresponding to the SSD recovery system 100 can be executed using the processor 210 and / or using other processors located away from the server computing device 202.

[0048] Data 216 may be retrieved, stored, or modified by processor 210 in accordance with instructions 214. Data 216 may be stored in a computer register, in a relational or non-relational database as a table with a plurality of different fields and records, or as a JSON, YAML, proto, or XML document. Data 216 may also be formatted in a computer-readable format, such as, but not limited to, binary values, ASCII, or Unicode. In addition, data 216 may include information sufficient to identify the relevant information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other storage (including other network locations), or information used by a function to calculate the relevant data.

[0049] The client computing device 204 may also be configured similarly to the server computing device 202, with one or more processors 220, memory 222, instructions 224, and data 226. The client computing device 204 may also include user input 228 and user output 230. The user input 228 may include any suitable mechanism or technology for receiving input from a user, such as a keyboard, mouse, mechanical actuator, soft actuator, touch screen, microphone, and sensor.

[0050] The server computing device 202 may be configured to transmit data to the client computing device 204, and the client computing device 204 may be configured to display at least a portion of the received data on a display implemented as part of the user output 230. The user output 230 may also be used to display an interface between the client computing device 204 and the server computing device 202. The user output 230 may alternatively or additionally include one or more speakers, transducers or other audio outputs, haptic interfaces, or other tactile feedback that provides non-visual and non-auditory information to a platform user of the client computing device 204.

[0051] although Figure 2Processors 210, 220 and memories 212, 222 are shown as being within computing devices 202, 204, but the components described herein may include multiple processors and memories that can operate at different physical locations rather than within the same computing device. For example, some of the instructions 214, 224 and data 216, 226 may be stored on a removable SD card, while other instructions and data may be stored within a read-only computer chip. Some or all of the instructions and data may be stored in a location physically away from the processors 210, 220 but still accessible to the processors. Similarly, processors 210, 220 may include a collection of processors that can perform concurrent and / or sequential operations. Computing devices 202, 204 may each include one or more internal clocks that provide timing information, which may be used for time measurement of operations and programs run by computing devices 202, 204.

[0052] The server computing device 202 may be configured to receive a request for processing data from the client computing device 204. For example, the environment 200 may be part of a computing platform configured to provide a variety of services to a user through various user interfaces and / or application programming interfaces (APIs) that expose platform services. The various services may include techniques for automatic SSD recovery. The client computing device 204 may send input data associated with an in-place restart for a virtual machine, a configurable timeout for an SSD recovery attempt, and / or automatic snapshots of the contents of the SSD storage device. The server computing device 202 may receive the input data and, in response, generate output data including instructions associated with an in-place restart, a configurable timeout, and / or automatic snapshots.

[0053] Devices 202, 204 may be capable of communicating directly and indirectly over network 208. For example, using network sockets, a client computing device 204 may connect to a service operating in a data center over the Internet Protocol. Devices 202, 204 may set up listening sockets that may accept initiated connections for sending and receiving information. Network 208 itself may include a variety of configurations and protocols, including the Internet, the World Wide Web, an intranet, a virtual private network, a wide area network, a local area network, and a private network using a proprietary communication protocol of one or more companies. Network 208 may support a variety of short-range and long-range connections. Short-range and long-range connections may be made over different bandwidths, such as those typically associated with a wireless network. 2.402 GHz to 2.480 GHz, usually associated with the 2.4 GHz and 5 GHz associated with communication protocols; or using multiple communication standards, such as for wireless broadband communication Additionally or alternatively, the network 208 may also support wired connections between the devices 202, 204, including through various types of Ethernet connections.

[0054] although Figure 2 A single server computing device 202, client computing device 204, and storage medium 206 are shown, but it should be understood that various aspects of the present disclosure may be implemented according to a variety of different configurations and numbers of computing devices (including in examples of sequential or parallel processing), or through a distributed network of multiple devices.

[0055] Figure 3 A flow chart of an example process 300 for implementing an in-place restart for one or more virtual machines is depicted. The example process 300 may be performed in a system of one or more processors at one or more locations, such as Figure 1 The automatic recovery system 100 is executed on the SSD.

[0056] As shown in block 310, the in-place restart engine 106 may receive data associated with an in-place restart. The data may include data associated with a user selection of an in-place restart mode to indicate that the user wants the in-place restart configuration to be applied to a physical machine having an SSD storage device. The data may also include metadata to match the virtual machine with the physical machine on which the virtual machine is being hosted. For example, the metadata may include the number of SSD disks on the physical machine and an identifier for each SSD disk. As another example, the metadata may include an IP address or an instance identifier of the virtual machine.

[0057] As shown in block 320, the restart-in-place engine 106 may store metadata associating the virtual machine with the physical machine. The metadata may be stored in any storage medium, which may be a combination of volatile and non-volatile memory and may be in the same or different physical location as the restart-in-place engine 106.

[0058] As shown in block 330, the in-place restart engine 106 may pause or shut down the virtual machine for maintenance. The in-place restart engine 106 may also prevent other virtual machines from being scheduled on the same physical machine to ensure that the physical machine is not oversubscribed. Maintenance may include planned maintenance (such as scheduled updates) or unplanned maintenance (such as a host crash). If it is planned maintenance, a notification may be provided to the user in advance that the virtual machine will be paused or shut down to allow the user to drain their applications. Planned maintenance may also be triggered by the user, such as through an API, to start maintenance immediately.

[0059] Maintenance may be performed on the virtual machine, as shown in block 340. The restart-in-place engine 106 or another engine, such as a maintenance engine (not shown), may perform maintenance, such as providing updates to the virtual machine. For example, the restart-in-place engine 106 or another engine may update a kernel or operating system on a physical machine, or update virtualization software that enables the virtual machine to run.

[0060] As shown in box 350, after maintenance is completed, the in-place restart engine 106 can restart the virtual machine on the same physical machine based on metadata. For example, the in-place restart engine 106 can match the identifier of the SSD disk with the instance identifier of the virtual machine. The in-place restart engine 106 can also match other identifiers associated with the virtual machine (such as a name or networking settings). The identifier of the virtual machine can be an internal unique identifier. The identifier of the SSD disk can include an index of the local SSD, a PCI topology, and / or an SSD name. A notification of maintenance completion can be provided to the user to allow the user to rehydrate their applications on demand. Since the same physical machine includes the same local SSD storage device, the content of the virtual machine can be saved, thereby providing consistency across virtual machine restarts.

[0061] Figure 4 A flow chart of an example process 400 for implementing a configurable SSD recovery timeout for one or more virtual machines is depicted. The example process 400 may be performed in a system of one or more processors at one or more locations, such as Figure 1 The automatic recovery system 100 is executed on the SSD.

[0062] As shown in box 410, the timeout engine 108 can receive data associated with a configurable timeout. The data can include data associated with a user selection of a configurable timeout to indicate that the user wants the configurable timeout to apply to SSD storage device recovery. The data can also include a threshold associated with a maximum amount of time to attempt local SSD recovery for planned or unplanned maintenance. For example, the maximum amount of time can range from 0 to multiple hours or days, and the maximum amount of time may be different if the maintenance is planned or unplanned. "0" can indicate that local SSD recovery should not be attempted.

[0063] As shown in block 420, the timeout engine 108 may pause or shut down the virtual machine for maintenance. The timeout engine 108 may also prevent other virtual machines from being scheduled on the same physical machine to ensure that the physical machine is not oversubscribed.

[0064] As shown in box 430, maintenance can be performed on the virtual machine, and as shown in box 440, local SSD recovery can be performed for the virtual machine. Although shown as separate boxes, performing maintenance on the virtual machine can also include performing local SSD recovery. The timeout engine 108 or another engine (such as a maintenance engine (not shown)) can perform maintenance and / or local SSD recovery. Local SSD recovery can be performed by confirming that the data on the physical machine is not damaged and then allowing the virtual machine to access the data if it is not damaged. As shown in box 450, the timeout engine 108 can monitor the amount of time it takes for the local SSD recovery to determine whether the local SSD recovery exceeds a threshold.

[0065] As shown in block 460, if the local SSD recovery exceeds a threshold amount of time, the timeout engine 108 may send an instruction to stop the local SSD recovery or stop the local SSD recovery. The virtual machine may be restarted on a different physical machine that may not have access to data on the local SSD of the previous physical machine hosting the virtual machine. If the threshold is 0, the timeout engine 108 may send an instruction to immediately restart the virtual machine on a different physical machine without attempting a local SSD recovery or monitoring the amount of time it takes for the local SSD recovery to take place.

[0066] Figure 5 A flow chart of an example process 500 for implementing automatic snapshotting of the contents of one or more virtual machines is depicted. The example process 500 may be performed in a system of one or more processors at one or more locations, such as Figure 1 The automatic recovery system 100 is executed on the SSD.

[0067] As shown in box 510, the snapshot engine 110 can receive data associated with taking an automatic snapshot. The data can include data associated with a user selection to take an automatic snapshot to indicate that the user wants the automatic snapshot configuration to be applied to a physical machine with an SSD storage device. The data can also include one or more trigger thresholds associated with the impact on the virtual machine to trigger the snapshot. The trigger thresholds can include application-level thresholds that the infrastructure of the virtual machine must meet, such as input / output operations slowing down to below a threshold or the transaction rate of data copy / import / export meeting a threshold. The trigger threshold can also include an indication that the virtual machine is about to terminate, such as a notification that maintenance is scheduled to occur. The data can also include whether the snapshot can be stored in a persistent storage device or archived in a cold storage device.

[0068] As shown in block 520, snapshot engine 110 may monitor aspects of one or more trigger thresholds to determine whether the thresholds are met. For example, snapshot engine 110 may monitor the speed of input / output operations of the virtual machine and / or the rate of transactions for data. Snapshot engine 110 may also monitor for signals that the virtual machine is about to terminate.

[0069] If one or more trigger thresholds are met, the snapshot engine 110 may automatically snapshot the contents of the local SSD, as shown in block 530. The virtual machine may continue to run and serve traffic, or the virtual machine may be removed from the service pool during the snapshot process due to potential application impact.

[0070] As shown in block 540, the snapshot engine 110 may send the snapshot to a persistent storage device (such as a persistent disk) or a cold storage device (such as an infrequently used storage device). The infrequently used storage device may include any storage device that can be externalized, such as a hard disk drive (HDD) or an external SSD. The snapshot engine 110 may also determine whether it is possible to archive the snapshot in the cold storage device, such as via a maximum amount of time that the snapshot may take and / or a maximum size of the snapshot. If the time and / or size are exceeded, the snapshot engine 110 may send an instruction to compress the snapshot or compress the snapshot, or may distribute the snapshot to other cloud regions.

[0071] As shown in block 550, if the snapshot was sent to a persistent storage device, the snapshot engine 110 may export the snapshot back. For example, the snapshot engine 110 may export the snapshot after a predetermined period of time or when one or more trigger thresholds are no longer met. The snapshot engine 110 may export the snapshot back to the same local SSD or a different local SSD for use with a future virtual machine.

[0072] Figure 6 Describes a method for implementing an SSD recovery system (such as Figure 1 1 is a block diagram of an example cloud-based platform 600 of an SSD recovery system 100. The platform 600 may include a collection 602 of host machines or physical machines 604 (e.g., hardware resources) that support or execute a virtual computing environment 650. The virtual computing environment 650 may include a virtual machine manager or hypervisor 652 and a virtual machine layer 654 that runs one or more virtual machines 656 that are configured to execute instances 658 of one or more applications 660.

[0073] Each physical machine 604 may include one or more physical processors 606 (e.g., data processing hardware) and associated physical memory 608 (e.g., memory hardware). Although each physical machine 604 is shown as having a single physical processor 606, the physical machine 604 may include multiple physical processors 606. The physical machine 604 may also include physical memory 608, which may be divided into virtual memory by the host operating system (OS) 610 and assigned for use by the virtual machine 656, the hypervisor 652, or the host OS 610. The physical memory 608 may include random access memory (RAM) and / or a disk storage device, such as an SSD storage device.

[0074] The host OS 610 may be executed on a given one of the physical machines 604 or may be configured to operate across multiple host machines 604. In addition, while the host OS 610 is shown as part of the virtual computing environment 650, each physical machine 604 may be equipped with its own OS 612. The OS 612 on each physical machine 604 may be managed as a common OS 610 for the hypervisor 652 and the virtual machine layer 654.

[0075] The hypervisor 652 may correspond to a computing engine that includes at least one of software, firmware, or hardware configured to create, instantiate / deploy, and execute virtual machines 656. The hypervisor 652 may include an SSD automatic recovery system 666. The SSD automatic recovery system 666 may be associated with Figure 1 The SSD automatic recovery system 100 corresponds to. Figure 6 654 , but in other implementations, the SSD automatic recovery system 666 may be part of the VM layer 654 .

[0076] Each virtual machine 656 may be referred to as a guest machine. The hypervisor 652 may be configured to provide each virtual machine 656 with a corresponding guest OS 662 having a virtual operating platform and manage the execution of the corresponding guest OS 662 on the virtual machine 656. In some examples, multiple virtual machines 656 having multiple guest OS 662 may share virtualized resources. For example, virtual machines with different operating systems may all run on a single physical host machine.

[0077] The host OS 610 can virtualize the underlying host hardware and manage the concurrent execution of the guest OS 662 on one or more virtual machines 656. For example, the host OS 610 can manage the virtual machines 656 to include emulated versions of the underlying host hardware or different computer architectures. The emulated version of the hardware associated with each virtual machine 656 can be referred to as virtual hardware 664.

[0078] The virtual hardware 664 may include one or more virtual processors, such as virtual central processing units (vCPUs), that emulate one or more physical processors 606 of the physical machine 604. The virtual processors may be interchangeably referred to as computing resources associated with the virtual machine 656. The computing resources may include a target computing resource level required to execute a corresponding individual service instance 658 of the application 660.

[0079] The virtual hardware 664 may also include a virtual memory that communicates with the virtual processor and stores guest instructions that can be executed by the virtual processor to perform operations. The virtual memory may be interchangeably referred to as a memory resource associated with the virtual machine 656. The memory resources may include a target memory resource level required to execute the corresponding individual service instance 658.

[0080] The virtual hardware 664 may also include at least one virtual storage device that provides runtime capacity for services on the host machine 604. The at least one virtual storage device may be referred to as a storage resource associated with the virtual machine 656. The storage resource may include a target storage resource level required to execute the corresponding individual service instance 658.

[0081] The virtual processor may execute instructions from the virtual memory that cause the virtual processor to execute a corresponding separate service instance 658 of the software application 660. The separate service instance 658 may be referred to as a guest instance that cannot determine whether it is executed by the virtual hardware 664 or the physical machine 604. The processor 606 of the host machine 604 may enable the virtual hardware 664 to efficiently execute the software instance 658 of the application 660 by allowing the guest software instructions to be executed directly on the processor 606 of the host machine 604 without code rewriting, recompiling, or instruction emulation.

[0082] The guest OS 662 executing on each virtual machine 656 can include software that controls the execution of the corresponding individual service instance 658 of the application 660 by the virtual machine 656. The guest OS executing on the virtual machine can be the same as or different from other guest OSs executing on other virtual machines. The guest OS 662 executing on each virtual machine 656 can also assign network boundaries, for example, assign network addresses through which the corresponding guest software can communicate with other processes reachable through the infrastructure (such as an internal network). The network boundary can be referred to as a network resource associated with the virtual machine 656.

[0083] Aspects of the present disclosure may be implemented in digital circuits, in computer-readable storage media, as one or more computer programs, or a combination of one or more of the foregoing. The computer-readable storage medium may be non-transitory, for example, as one or more instructions executable by a cloud computing platform and stored on a tangible storage device.

[0084] The phrase "configured to" is used in different contexts relating to portions of computer systems, hardware, or computer programs. When a system is said to be configured to perform one or more operations, this means that the system has appropriate software, firmware, and / or hardware installed on the system that, when in operation, causes the system to perform the one or more operations. When certain hardware is said to be configured to perform one or more operations, this means that the hardware includes one or more circuits that, when in operation, receive inputs and generate outputs based on the inputs and corresponding to the one or more operations. When a computer program is said to be configured to perform one or more operations, this means that the computer program includes one or more program instructions that, when executed by one or more computers, cause the one or more computers to perform the one or more operations.

[0085] Unless otherwise stated, the aforementioned alternative examples are not mutually exclusive, but may be implemented in various combinations to achieve unique advantages. Since these and other variations and combinations of the features discussed above may be utilized without departing from the subject matter defined by the claims, the foregoing description of the embodiments should be made in an illustrative manner rather than in a manner that limits the subject matter defined by the claims. In addition, the examples described herein and clauses expressed as "such as", "including", etc. should not be interpreted as limiting the subject matter of the claims to specific examples; rather, the examples are intended to illustrate only one of many possible embodiments. In addition, the same reference numerals in different figures may identify the same or similar elements.

Claims

1. A method for controlling recovery of a local solid state drive (SSD) storage device, characterized in that include: storing, by the one or more processors, metadata indicating a physical machine that is hosting the virtual machine, the physical machine including a local solid-state drive storage device; pausing or shutting down, by the one or more processors, the virtual machine for maintenance; as well as The one or more processors restart the virtual machine on the physical machine based on the metadata after performing maintenance.

2. The method according to claim 1, characterized in that Also included is saving, by the one or more processors, content of the virtual machine on the local solid-state drive storage device based on restarting the virtual machine on the physical machine.

3. The method according to claim 1, characterized in that Also included is performing, by the one or more processors, the maintaining of the virtual machine.

4. The method according to claim 1, characterized in that Restarting the virtual machine on the physical machine based on the metadata further includes matching an identifier of the SSD storage device of the physical machine with an identifier of the virtual machine.

5. The method according to claim 1, characterized in that Also includes: monitoring, by the one or more processors, an amount of time for a recovery process to attempt to recover the local solid-state drive storage device; determining, by the one or more processors, that the amount of time has exceeded a predetermined maximum amount of time for attempting to recover the local solid-state drive storage device; as well as An instruction is sent by the one or more processors to stop the recovery process.

6. The method according to claim 1, characterized in that Also includes: monitoring, by the one or more processors, an impact on the virtual machine relative to one or more trigger thresholds; determining, by the one or more processors, that one or more of the triggering thresholds are satisfied; snapshotting, by the one or more processors, the contents of the local solid-state drive storage device to generate a snapshot; as well as The snapshot is sent, by the one or more processors, to a persistent storage device.

7. The method according to claim 6, characterized in that The one or more triggering thresholds include one or more of a speed of an input or output operation slowing down below a threshold or a rate of data copying, importing or exporting slowing down below a threshold.

8. The method according to claim 6, characterized in that Also included is exporting, by the one or more processors, the snapshot from the persistent storage device to the local SSD storage device or a different local solid-state drive storage device.

9. The method according to claim 1, characterized in that Also includes: monitoring, by the one or more processors, an impact on the virtual machine relative to one or more trigger thresholds; determining, by the one or more processors, that one or more of the triggering thresholds are satisfied; snapshotting, by the one or more processors, the contents of the local solid-state drive storage device to generate a snapshot; as well as The snapshot is archived, by the one or more processors, to a cold storage device.

10. A system, characterized in that: include: one or more processors; as well as one or more storage devices coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations for controlling recovery of a local solid-state drive (SSD) storage device, the operations comprising: storing metadata indicating a physical machine that is hosting the virtual machine, the physical machine including a local solid-state drive storage device; Pausing or shutting down the virtual machine for maintenance; and The virtual machine is restarted on the physical machine based on the metadata after maintenance is performed.

11. The system according to claim 10, characterized in that Restarting the virtual machine on the physical machine based on the metadata further includes matching an identifier of the solid-state drive storage device of the physical machine with an identifier of the virtual machine.

12. The system according to claim 10, characterized in that The operations also include: monitoring an amount of time for a recovery process that attempts to recover the local solid-state drive storage device; determining that the amount of time has exceeded a predetermined maximum amount of time for attempting to recover the local solid-state drive storage device; and Sending an instruction to stop the recovery process.

13. The system according to claim 10, characterized in that The operations also include: monitoring an impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the triggering thresholds are satisfied; snapshotting the contents of the local solid-state drive storage device to generate a snapshot; and The snapshot is sent to a persistent storage device.

14. The system of claim 13, wherein: The operations also include exporting the snapshot from the persistent storage device to the local solid-state drive storage device or a different local solid-state drive storage device.

15. The system of claim 10, wherein: The operations also include: monitoring an impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the triggering thresholds are satisfied; snapshotting the contents of the local solid-state drive storage device to generate a snapshot; and The snapshot is archived to a cold storage device.

16. A non-transitory computer-readable medium for storing instructions, characterized in that: The instructions, when executed by one or more processors, cause the one or more processors to perform operations for controlling recovery of a local solid-state drive (SSD) storage device, the operations comprising: storing metadata indicating a physical machine that is hosting the virtual machine, the physical machine including a local solid-state drive storage device; Pausing or shutting down the virtual machine for maintenance; and The virtual machine is restarted on the physical machine based on the metadata after maintenance is performed.

17. The non-transitory computer readable medium of claim 16, wherein: Restarting the virtual machine on the physical machine based on the metadata further includes matching an identifier of the solid-state drive storage device of the physical machine with an identifier of the virtual machine.

18. The non-transitory computer readable medium of claim 16, wherein: The operations also include: monitoring an amount of time for a recovery process that attempts to recover the local solid-state drive storage device; determining that the amount of time has exceeded a predetermined maximum amount of time for attempting to recover the local solid-state drive storage device; and Sending an instruction to stop the recovery process.

19. The non-transitory computer readable medium of claim 16, wherein: The operations also include: monitoring an impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the triggering thresholds are satisfied; snapshotting the contents of the local solid-state drive storage device to generate a snapshot; sending the snapshot to a persistent storage device; and The snapshot is imported back to the local solid-state drive storage device or a different local solid-state drive storage device.

20. The non-transitory computer readable medium of claim 16, wherein: The operations also include: monitoring an impact on the virtual machine relative to one or more trigger thresholds; determining that one or more of the triggering thresholds are satisfied; snapshotting the contents of the local solid-state drive storage device to generate a snapshot; and The snapshot is archived to a cold storage device.

Citation Information

Patent Citations

  • Automated SSD Recovery

    US20240289027A1