Virtualization-based transaction handling in an on-demand network code execution system

Abstract

Systems and methods are described for providing virtualized transaction retries in an on-demand network code execution system. A user may generate a task on the system by submitting code. The task may include a transaction that can succeed or fail depending on external factors, such as accessing a resource that other processes are also trying to access. The system may obtain success criteria for evaluating whether the transaction was successful, and may configure a virtual machine instance to execute the submitted code. The system may generate a “snapshot” or otherwise capture information regarding the state of the virtual machine instance prior to executing the code, and may then execute the code or a portion of it to attempt the transaction. The system may then apply the success criteria to determine whether the transaction was successful, and if not may retry the transaction by restoring the snapshot and re-executing the code.

Description

BACKGROUND

Computing devices can utilize communication networks to exchange data. Companies and organizations operate computer networks that interconnect a number of computing devices to support operations or provide services to third parties. The computing systems can be located in a single geographic location or located in multiple, distinct geographic locations (e.g., interconnected via private or public communication networks). Specifically, hosted computing environments or data processing centers, generally referred to herein as “data centers,” may include a number of interconnected computing systems to provide computing resources to users of the data center. The data centers may be private data centers operated on behalf of an organization, or public data centers operated on behalf, or for the benefit of, the general public.

To facilitate increased utilization of data center resources, virtualization technologies allow a single physical computing device to host one or more instances of virtual machines that appear and operate as independent computing devices to users of a data center. With virtualization, the single physical computing device can create, maintain, delete, or otherwise manage virtual machines in a dynamic manner. In turn, users can request computing resources from a data center, such as single computing devices or a configuration of networked computing devices, and be provided with varying numbers of virtual machine resources.

In some scenarios, a user can request that a data center provide computing resources to execute a particular task. The task may correspond to a set of computer-executable instructions, which the data center may then execute on behalf of the user. The data center may thus further facilitate increased utilization of data center resources.

BRIEF DESCRIPTION OF THE DRAWINGS

Throughout the drawings, reference numbers may be re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate example embodiments described herein and are not intended to limit the scope of the disclosure.

FIG. 1 is a block diagram depicting an illustrative environment in which an on-demand code execution system can execute tasks corresponding to code, which may be submitted by users of the on-demand code execution system, and can retry execution of submitted code in accordance with aspects of the present disclosure.

FIG. 2 depicts a general architecture of a computing device providing a transaction execution system that is configured to determine whether to retry code execution in the on-demand code execution system of FIG. 1.

FIGS. 3A, 3B, and 3C are flow diagrams depicting illustrative interactions for submitting code corresponding to a task to the on-demand code execution system of FIG. 1, and for the on-demand code execution system to determine whether a task has succeeded and to retry execution of tasks that did not complete successfully.

FIG. 4 is a flow chart depicting an illustrative routine for retrying execution of submitted code based on a measure of execution success in the on-demand code execution system of FIG. 1.

DETAILED DESCRIPTION

Generally described, aspects of the present disclosure relate to an on-demand code execution system. The on-demand code execution system enables rapid execution of code, which may be supplied by users of the on-demand code execution system. More specifically, embodiments of the present disclosure relate to improving the performance of an on-demand code execution system that is implemented using various computing resources. As described in detail herein, the on-demand code execution system may provide a network-accessible service enabling users to submit or designate computer-executable code to be executed by virtual machine instances on the on-demand code execution system. Each set of code on the on-demand code execution system may define a “task,” and implement specific functionality corresponding to that task when executed on a virtual machine instance of the on-demand code execution system. Individual implementations of the task on the on-demand code execution system may be referred to as an “execution” of the task (or a “task execution”). The on-demand code execution system can further enable users to trigger execution of a task based on a variety of potential events, such as detecting new data at a network-based storage system, transmission of an application programming interface (“API”) call to the on-demand code execution system, or transmission of a specially formatted hypertext transport protocol (“HTTP”) packet to the on-demand code execution system. Thus, users may utilize the on-demand code execution system to execute any specified executable code “on-demand,” without requiring configuration or maintenance of the underlying hardware or infrastructure on which the code is executed. Further, the on-demand code execution system may be configured to execute tasks in a rapid manner (e.g., in under 100 milliseconds [ms]), thus enabling execution of tasks in “real-time” (e.g., with little or no perceptible delay to an end user).

The on-demand code-execution system may thus allow users to execute code in a serverless environment (e.g., one in which the underlying server is not under user control). The term “serverless environment,” as used herein, is intended to refer to an environment in which responsibility for managing generation, configuration, and state of an underlying execution environment is abstracted away from a user, such that the user need not, for example, create the execution environment, install an operating system within the execution environment, or manage a state of the environment in order to execute desired code in the environment. Similarly, the term “server-based environment” is intended to refer to an environment in which a user is at least partly responsible for managing generation, configuration, or state of an underlying execution environment in addition to executing desired code in the environment. One skilled in the art will thus appreciate that “serverless” and “server-based” may indicate the degree of user control over execution environments in which code is executed, rather than the actual absence or presence of a server.

To further facilitate rapid execution of code, the on-demand code execution system may implement a transaction execution system as described herein. The transaction execution system may enable virtualization of transactions that have a possibility of succeeding or failing due to conditions that are external to the executing code, such as network congestion, contention for shared resources, intermittent hardware or software failure, maintenance activities, or other conditions that may cause a transaction to succeed at some times and fail at other times. The transaction execution system may illustratively handle code that performs such transactions by taking a “snapshot” of an execution environment prior to executing the transaction, executing the transaction and obtaining a result, and then restoring the snapshot and re-executing the code if the transaction is unsuccessful. The user-submitted code may thus be simplified by removing any error handling routines or retry mechanisms that manually implement recovering from a failed transaction attempt: Instead, the retries are automatically performed by the on-demand code execution system. It thus appears from the perspective of the user-submitted code that every transaction succeeds on the first try, since any transactions that fail are effectively “rewound” to a point in time before the failure and re-attempted. As a result, the user may not have to worry about implementing transaction tracking paradigms into the user-submitted code.

As will be appreciated by one of skill in the art in light of the present disclosure, the embodiments disclosed herein improve the ability of computing systems, such as on-demand code execution systems, to execute code in an efficient manner. Moreover, the presently disclosed embodiments address technical problems inherent within computing systems; specifically, the occurrence of intermittent failures in computing systems, and the difficulty of programmatically handling such failures when they occur. These technical problems are addressed by the various technical solutions described herein, including the provisioning of a transaction execution system within an on-demand code execution system that utilizes state-saving techniques (such as snapshots) to retry transactions externally to code that attempts the transaction, potentially providing to the code an appearance that transactions always succeed. Thus, the present disclosure represents an improvement on existing data processing systems and computing systems in general.

As described in more detail below, the on-demand code execution system may include a worker manager configured to receive user code (threads, programs, etc., composed in any of a variety of programming languages) and execute the code in a highly scalable, low latency manner, without requiring user configuration of a virtual machine instance. Specifically, the worker manager can, prior to receiving the user code and prior to receiving any information from a user regarding any particular virtual machine instance configuration, create and configure virtual machine instances according to a predetermined set of configurations, each corresponding to any one or more of a variety of run-time environments. Thereafter, the worker manager receives user-initiated requests to execute code, and identifies a pre-configured virtual machine instance on which to execute the code based on configuration information associated with the request. The worker manager can further allocate the identified virtual machine instance to execute the user's code at least partly by creating and configuring containers inside the allocated virtual machine instance, and provisioning the containers with code of the task as well as any dependency code objects. Various embodiments for implementing a worker manager and executing user code on virtual machine instances is described in more detail in U.S. Pat. No. 9,323,556, entitled “PROGRAMMATIC EVENT DETECTION AND MESSAGE GENERATION FOR REQUESTS TO EXECUTE PROGRAM CODE,” and filed Sep. 30, 2014 (the “'556 patent”), the entirety of which is hereby incorporated by reference.

As used herein, the term “virtual machine instance” is intended to refer to an execution of software or other executable code that emulates hardware to provide an environment or platform on which software may execute (an “execution environment”). Virtual machine instances are generally executed by hardware devices, which may differ from the physical hardware emulated by the virtual machine instance. For example, a virtual machine may emulate a first type of processor and memory while being executed on a second type of processor and memory. Thus, virtual machines can be utilized to execute software intended for a first execution environment (e.g., a first operating system) on a physical device that is executing a second execution environment (e.g., a second operating system). In some instances, hardware emulated by a virtual machine instance may be the same or similar to hardware of an underlying device. For example, a device with a first type of processor may implement a plurality of virtual machine instances, each emulating an instance of that first type of processor. Thus, virtual machine instances can be used to divide a device into a number of logical sub-devices (each referred to as a “virtual machine instance”). While virtual machine instances can generally provide a level of abstraction away from the hardware of an underlying physical device, this abstraction is not required. For example, assume a device implements a plurality of virtual machine instances, each of which emulate hardware identical to that provided by the device. Under such a scenario, each virtual machine instance may allow a software application to execute code on the underlying hardware without translation, while maintaining a logical separation between software applications running on other virtual machine instances. This process, which is generally referred to as “native execution,” may be utilized to increase the speed or performance of virtual machine instances. Other techniques that allow direct utilization of underlying hardware, such as hardware pass-through techniques, may be used as well.

While a virtual machine executing an operating system is described herein as one example of an execution environment, other execution environments are also possible. For example, tasks or other processes may be executed within a software “container,” which provides a runtime environment without itself providing virtualization of hardware. Containers may be implemented within virtual machines to provide additional security, or may be run outside of a virtual machine instance.

Embodiments of the disclosure will now be described with reference to the accompanying figures, wherein like numerals refer to like elements throughout. The terminology used in the description presented herein is not intended to be interpreted in any limited or restrictive manner, simply because it is being utilized in conjunction with a detailed description of certain specific embodiments of the disclosure. Furthermore, embodiments of the disclosure may include several novel features, no single one of which is solely responsible for its desirable attributes or which is essential to practicing the disclosure herein described.

FIG. 1 is a block diagram of an illustrative operating environment 100 in which an on-demand code execution system 110 may operate based on communication with user computing devices 102, auxiliary services 106, and network-based data storage services 108. In general, the user computing devices 102 can be any computing device such as a desktop, laptop or tablet computer, personal computer, wearable computer, server, personal digital assistant (PDA), hybrid PDA/mobile phone, mobile phone, electronic book reader, set-top box, voice command device, camera, digital media player, and the like. The on-demand code execution system 110 may provide the user computing devices 102 with one or more user interfaces, command-line interfaces (CLIs), application programing interfaces (APIs), and/or other programmatic interfaces for generating and uploading user-executable code (including metadata identifying dependency code objects for the uploaded code), invoking the user-provided code (e.g., submitting a request to execute the user code on the on-demand code execution system 110), scheduling event-based jobs or timed jobs, tracking the user-provided code, and/or viewing other logging or monitoring information related to their requests and/or user code. Although one or more embodiments may be described herein as using a user interface, it should be appreciated that such embodiments may, additionally or alternatively, use any CLIs, APIs, or other programmatic interfaces.

The illustrative environment 100 further includes one or more network-based data storage services 108, which are configured to enable the on-demand code execution system 110 to store and retrieve data from one or more persistent or substantially persistent data sources. Illustratively, the network-based data storage services 108 may enable the on-demand code execution system 110 to store information corresponding to a task, such as code or metadata, to store additional code objects representing dependencies of tasks, to retrieve data to be processed during execution of a task, and to store information (e.g., results) regarding that execution. The network-based data storage services 108 may represent, for example, a relational or non-relational database. In another example, the network-based data storage services 108 may represent a network-attached storage (NAS), configured to provide access to data arranged as a file system. The network-based data storage services 108 may further enable the on-demand code execution system 110 to query for and retrieve information regarding data stored within the on-demand code execution system 110, such as by querying for a number of relevant files or records, sizes of those files or records, file or record names, file or record creation times, etc. In some instances, the network-based data storage services 108 may provide additional functionality, such as the ability to separate data into logical groups (e.g., groups associated with individual accounts, etc.). While shown as distinct from the auxiliary services 106, the network-based data storage services 108 may in some instances also represent a type of auxiliary service 106.

The user computing devices 102, auxiliary services 106, and network-based data storage services 108 may communicate with the on-demand code execution system 110 via a network 104, which may include any wired network, wireless network, or combination thereof. For example, the network 104 may be a personal area network, local area network, wide area network, over-the-air broadcast network (e.g., for radio or television), cable network, satellite network, cellular telephone network, or combination thereof. As a further example, the network 104 may be a publicly accessible network of linked networks, possibly operated by various distinct parties, such as the Internet. In some embodiments, the network 104 may be a private or semi-private network, such as a corporate or university intranet. The network 104 may include one or more wireless networks, such as a Global System for Mobile Communications (GSM) network, a Code Division Multiple Access (CDMA) network, a Long Term Evolution (LTE) network, or any other type of wireless network. The network 104 can use protocols and components for communicating via the Internet or any of the other aforementioned types of networks. For example, the protocols used by the network 104 may include Hypertext Transfer Protocol (HTTP), HTTP Secure (HTTPS), Message Queue Telemetry Transport (MQTT), Constrained Application Protocol (CoAP), and the like. Protocols and components for communicating via the Internet or any of the other aforementioned types of communication networks are well known to those skilled in the art and, thus, are not described in more detail herein.

The on-demand code execution system 110 is depicted in FIG. 1 as operating in a distributed computing environment including several computer systems that are interconnected using one or more computer networks (not shown in FIG. 1). The on-demand code execution system 110 could also operate within a computing environment having a fewer or greater number of devices than are illustrated in FIG. 1. Thus, the depiction of the on-demand code execution system 110 in FIG. 1 should be taken as illustrative and not limiting to the present disclosure. For example, the on-demand code execution system 110 or various constituents thereof could implement various Web services components, hosted or “cloud” computing environments, and/or peer to peer network configurations to implement at least a portion of the processes described herein.

Further, the on-demand code execution system 110 may be implemented directly in hardware or software executed by hardware devices and may, for instance, include one or more physical or virtual servers implemented on physical computer hardware configured to execute computer executable instructions for performing various features that will be described herein. The one or more servers may be geographically dispersed or geographically co-located, for instance, in one or more data centers. In some instances, the one or more servers may operate as part of a system of rapidly provisioned and released computing resources, often referred to as a “cloud computing environment.”

In the example of FIG. 1, the on-demand code execution system 110 is illustrated as connected to the network 104. In some embodiments, any of the components within the on-demand code execution system 110 can communicate with other components of the on-demand code execution system 110 via the network 104. In other embodiments, not all components of the on-demand code execution system 110 are capable of communicating with other components of the operating environment 100. In one example, only the frontend 120 (which may in some instances represent multiple frontends 120) may be connected to the network 104, and other components of the on-demand code execution system 110 may communicate with other components of the environment 100 via the frontends 120.

In FIG. 1, users, by way of user computing devices 102, may interact with the on-demand code execution system 110 to provide executable code, and establish rules or logic defining when and how such code should be executed on the on-demand code execution system 110, thus establishing a “task.” For example, a user may wish to run a piece of code in connection with a web or mobile application that the user has developed. One way of running the code would be to acquire virtual machine instances from service providers who provide infrastructure as a service, configure the virtual machine instances to suit the user's needs, and use the configured virtual machine instances to run the code. In order to avoid the complexity of this process, the user may alternatively provide the code to the on-demand code execution system 110, and request that the on-demand code execution system 110 execute the code. The on-demand code execution system 110 can handle the acquisition and configuration of compute capacity (e.g., containers, instances, etc., which are described in greater detail below) based on the code execution request, and execute the code using the compute capacity. The on-demand code execution system 110 may automatically scale up and down based on the volume, thereby relieving the user from the burden of having to worry about over-utilization (e.g., acquiring too little computing resources and suffering performance issues) or under-utilization (e.g., acquiring more computing resources than necessary to run the codes, and thus overpaying). In accordance with embodiments of the present disclosure, and as described in more detail below, the on-demand code execution system 110 may configure the virtual machine instances with customized operating systems to execute the user's code more efficiency and reduce utilization of computing resources.

To enable interaction with the on-demand code execution system 110, the system 110 includes one or more frontends 120, which enable interaction with the on-demand code execution system 110. In an illustrative embodiment, the frontends 120 serve as a “front door” to the other services provided by the on-demand code execution system 110, enabling users (via user computing devices 102) to provide, request execution of, and view results of computer executable code. The frontends 120 include a variety of components to enable interaction between the on-demand code execution system 110 and other computing devices. For example, each frontend 120 may include a request interface providing user computing devices 102 with the ability to upload or otherwise communication user-specified code to the on-demand code execution system 110 and to thereafter request execution of that code. In one embodiment, the request interface communicates with external computing devices (e.g., user computing devices 102, auxiliary services 106, etc.) via a graphical user interface (GUI), CLI, or API. The frontends 120 process the requests and makes sure that the requests are properly authorized. For example, the frontends 120 may determine whether the user associated with the request is authorized to access the user code specified in the request.

References to user code as used herein may refer to any program code (e.g., a program, routine, subroutine, thread, etc.) written in a specific program language. In the present disclosure, the terms “code,” “user code,” and “program code,” may be used interchangeably. Such user code may be executed to achieve a specific function, for example, in connection with a particular web application or mobile application developed by the user. As noted above, individual collections of user code (e.g., to achieve a specific function) are referred to herein as “tasks,” while specific executions of that code (including, e.g., compiling code, interpreting code, or otherwise making the code executable) are referred to as “task executions” or simply “executions.” Tasks may be written, by way of non-limiting example, in JavaScript (e.g., node.js), Java, Python, and/or Ruby (and/or another programming language). Tasks may be “triggered” for execution on the on-demand code execution system 110 in a variety of manners. In one embodiment, a user or other computing device may transmit a request to execute a task may, which can generally be referred to as “call” to execute of the task. Such calls may include the user code (or the location thereof) to be executed and one or more arguments to be used for executing the user code. For example, a call may provide the user code of a task along with the request to execute the task. In another example, a call may identify a previously uploaded task by its name or an identifier. In yet another example, code corresponding to a task may be included in a call for the task, as well as being uploaded in a separate location (e.g., storage of an auxiliary service 106 or a storage system internal to the on-demand code execution system 110) prior to the request being received by the on-demand code execution system 110. As noted above, the code for a task may reference additional code objects maintained at the on-demand code execution system 110 by use of identifiers of those code objects, such that the code objects are combined with the code of a task in an execution environment prior to execution of the task. The on-demand code execution system 110 may vary its execution strategy for a task based on where the code of the task is available at the time a call for the task is processed. A request interface of the frontend 120 may receive calls to execute tasks as Hypertext Transfer Protocol Secure (HTTPS) requests from a user. Also, any information (e.g., headers and parameters) included in the HTTPS request may also be processed and utilized when executing a task. As discussed above, any other protocols, including, for example, HTTP, MQTT, and CoAP, may be used to transfer the message containing a task call to the request interface.

A call to execute a task may specify one or more third-party libraries (including native libraries) to be used along with the user code corresponding to the task. In one embodiment, the call may provide to the on-demand code execution system 110 a file containing the user code and any libraries (and/or identifications of storage locations thereof) corresponding to the task requested for execution. In some embodiments, the call includes metadata that indicates the program code of the task to be executed, the language in which the program code is written, the user associated with the call, and/or the computing resources (e.g., memory, etc.) to be reserved for executing the program code. For example, the program code of a task may be provided with the call, previously uploaded by the user, provided by the on-demand code execution system 110 (e.g., standard routines), and/or provided by third parties. Illustratively, code not included within a call or previously uploaded by the user may be referenced within metadata of the task by use of a URI associated with the code. In some embodiments, such resource-level constraints (e.g., how much memory is to be allocated for executing a particular user code) are specified for the particular task, and may not vary over each execution of the task. In such cases, the on-demand code execution system 110 may have access to such resource-level constraints before each individual call is received, and the individual call may not specify such resource-level constraints. In some embodiments, the call may specify other constraints such as permission data that indicates what kind of permissions or authorities that the call invokes to execute the task. Such permission data may be used by the on-demand code execution system 110 to access private resources (e.g., on a private network). In some embodiments, individual code objects may also be associated with permissions or authorizations. For example, a third party may submit a code object and designate the object as readable by only a subset of users. The on-demand code execution system 110 may include functionality to enforce these permissions or authorizations with respect to code objects.

In some embodiments, a call may specify the behavior that should be adopted for handling the call. In such embodiments, the call may include an indicator for enabling one or more execution modes in which to execute the task referenced in the call. For example, the call may include a flag or a header for indicating whether the task should be executed in a debug mode in which the debugging and/or logging output that may be generated in connection with the execution of the task is provided back to the user (e.g., via a console user interface). In such an example, the on-demand code execution system 110 may inspect the call and look for the flag or the header, and if it is present, the on-demand code execution system 110 may modify the behavior (e.g., logging facilities) of the container in which the task is executed, and cause the output data to be provided back to the user. In some embodiments, the behavior/mode indicators are added to the call by the user interface provided to the user by the on-demand code execution system 110. Other features such as source code profiling, remote debugging, etc. may also be enabled or disabled based on the indication provided in a call.

To manage requests for code execution, the frontend 120 can include an execution queue (not shown in FIG. 1), which can maintain a record of requested task executions. Illustratively, the number of simultaneous task executions by the on-demand code execution system 110 is limited, and as such, new task executions initiated at the on-demand code execution system 110 (e.g., via an API call, via a call from an executed or executing task, etc.) may be placed on the execution queue and processed, e.g., in a first-in-first-out order. In some embodiments, the on-demand code execution system 110 may include multiple execution queues, such as individual execution queues for each user account. For example, users of the on-demand code execution system 110 may desire to limit the rate of task executions on the on-demand code execution system 110 (e.g., for cost reasons). Thus, the on-demand code execution system 110 may utilize an account-specific execution queue to throttle the rate of simultaneous task executions by a specific user account. In some instances, the on-demand code execution system 110 may prioritize task executions, such that task executions of specific accounts or of specified priorities bypass or are prioritized within the execution queue. In other instances, the on-demand code execution system 110 may execute tasks immediately or substantially immediately after receiving a call for that task, and thus, the execution queue may be omitted.

As noted above, tasks may be triggered for execution at the on-demand code execution system 110 based on explicit calls from user computing devices 102 (e.g., as received at the request interface). Alternatively or additionally, tasks may be triggered for execution at the on-demand code execution system 110 based on data retrieved from one or more auxiliary services 106 or network-based data storage services 108. To facilitate interaction with auxiliary services 106, the frontend 120 can include a polling interface (not shown in FIG. 1), which operates to poll auxiliary services 106 or data storage services 108 for data. Illustratively, the polling interface may periodically transmit a request to one or more user-specified auxiliary services 106 or data storage services 108 to retrieve any newly available data (e.g., social network “posts,” news articles, files, records, etc.), and to determine whether that data corresponds to a user-established criteria triggering execution a task on the on-demand code execution system 110. Illustratively, criteria for execution of a task may include, but is not limited to, whether new data is available at the auxiliary services 106 or data storage services 108, the type or content of the data, or timing information corresponding to the data. In some instances, the auxiliary services 106 or data storage services 108 may function to notify the frontend 120 of the availability of new data, and thus the polling service may be unnecessary with respect to such services.

In addition to tasks executed based on explicit user calls and data from auxiliary services 106, the on-demand code execution system 110 may in some instances operate to trigger execution of tasks independently. For example, the on-demand code execution system 110 may operate (based on instructions from a user) to trigger execution of a task at each of a number of specified time intervals (e.g., every 10 minutes).

The frontend 120 can further include an output interface (not shown in FIG. 1) configured to output information regarding the execution of tasks on the on-demand code execution system 110. Illustratively, the output interface may transmit data regarding task executions (e.g., results of a task, errors related to the task execution, or details of the task execution, such as total time required to complete the execution, total data processed via the execution, etc.) to the user computing devices 102 or to auxiliary services 106, which may include, for example, billing or logging services. The output interface may further enable transmission of data, such as service calls, to auxiliary services 106. For example, the output interface may be utilized during execution of a task to transmit an API request to an external service 106 (e.g., to store data generated during execution of the task).

In some embodiments, the on-demand code execution system 110 may include multiple frontends 120. In such embodiments, a load balancer (not shown in FIG. 1) may be provided to distribute the incoming calls to the multiple frontends 120, for example, in a round-robin fashion. In some embodiments, the manner in which the load balancer distributes incoming calls to the multiple frontends 120 may be based on the location or state of other components of the on-demand code execution system 110. For example, a load balancer may distribute calls to a geographically nearby frontend 120, or to a frontend with capacity to service the call. In instances where each frontend 120 corresponds to an individual instance of another component of the on-demand code execution system, such as the active pool 140A described below, the load balancer may distribute calls according to the capacities or loads on those other components. As will be described in more detail below, calls may in some instances be distributed between frontends 120 deterministically, such that a given call to execute a task will always (or almost always) be routed to the same frontend 120. This may, for example, assist in maintaining an accurate execution record for a task, to ensure that the task executes only a desired number of times. While distribution of calls via a load balancer is illustratively described, other distribution techniques, such as anycast routing, will be apparent to those of skill in the art.

To execute tasks, the on-demand code execution system 110 includes one or more worker managers 140 that manage the instances used for servicing incoming calls to execute tasks. In the example illustrated in FIG. 1, each worker manager 140 manages an active pool of virtual machine instances 154A-C, which are currently assigned to one or more users and are implemented by one or more physical host computing devices 150A-B. The physical host computing devices 150A-B and the virtual machine instances 154A-C may further implement one or more containers 158A-F, which may contain and execute one or more user-submitted codes 160A-G. Containers are logical units created within a virtual machine instance, or on a host computing device, using the resources available on that instance or device. For example, each worker manager 140 may, based on information specified in a call to execute a task, create a new container or locate an existing container 158A-F and assign the container to handle the execution of the task. Each container may correspond to an execution environment for the task, providing at least some isolation from other execution environments. For example, each container may provide a file system isolated from other file systems on the device, and code executing in the container may have limited or no access to other file systems or memory space associated with code executing outside of the container.

The containers 156A-F, virtual machine instances 154A-C, and host computing devices 150A-B may further include language runtimes, code libraries, or other supporting functions (not depicted in FIG. 1) that facilitate execution of user-submitted code 160A-G. The physical computing devices 150A-B and the virtual machine instances 154A-C may further include operating systems 152A-B and 156A-C. In various embodiments, operating systems 152A-B and 156A-C may be the same operating system, variants of the same operating system, different operating systems, or combinations thereof.

Although the virtual machine instances 154A-C are described here as being assigned to a particular user, in some embodiments, an instance 154A-C may be assigned to a group of users, such that the instance is tied to the group of users and any member of the group can utilize resources on the instance. For example, the users in the same group may belong to the same security group (e.g., based on their security credentials) such that executing one member's task in a container on a particular instance after another member's task has been executed in another container on the same instance does not pose security risks. Similarly, the worker managers 140 may assign the instances and the containers according to one or more policies that dictate which requests can be executed in which containers and which instances can be assigned to which users. An example policy may specify that instances are assigned to collections of users who share the same account (e.g., account for accessing the services provided by the on-demand code execution system 110). In some embodiments, the requests associated with the same user group may share the same containers (e.g., if the user codes associated therewith are identical). In some embodiments, a task does not differentiate between the different users of the group and simply indicates the group to which the users associated with the task belong.

Once a triggering event to execute a task has been successfully processed by a frontend 120, the frontend 120 passes a request to a worker manager 140 to execute the task. In one embodiment, each frontend 120 may be associated with a corresponding worker manager 140 (e.g., a worker manager 140 co-located or geographically nearby to the frontend 120) and thus the frontend 120 may pass most or all requests to that worker manager 140. In another embodiment, a frontend 120 may include a location selector configured to determine a worker manager 140 to which to pass the execution request. In one embodiment, the location selector may determine the worker manager 140 to receive a call based on hashing the call, and distributing the call to a worker manager 140 selected based on the hashed value (e.g., via a hash ring). Various other mechanisms for distributing calls between worker managers 140 will be apparent to one of skill in the art.

The on-demand code execution system 110 further includes a transaction execution system 170, which implements aspects of the present disclosure including, for example, determining whether a task successfully completed a transaction. In some embodiments, the transaction execution system 170 includes a transaction analyzer 172, which may be invoked when the user submits code via the frontend 120 that includes a transaction. In some embodiments, as described in more detail below, the transaction analyzer 172 may analyze the user-submitted code to identify a transaction and determine criteria for evaluating whether the transaction was executed successfully. In other embodiments, the user may provide metadata or other information that identifies a transaction in their submitted code and provides one or more success criteria. The transaction analyzer 172 may, in some embodiments, obtain an output or other result associated with executing the task, and may apply the success criteria to the output to determine whether the transaction succeeded.

The transaction execution system 170 may further include a snapshot manager 174, which may capture state information regarding the particular host computing device 150A-B, virtual machine instance 154A-C, container 158A-F, and/or other computing resources that are used to execute the task. As described in more detail below, the snapshot manager 174 may capture the state information, store it (e.g., in the snapshot data store 176), and then use this information to restore a container 158A-F, virtual machine instance 154A-C, host computing device 150A-B, and/or other computing resource to a previous state, such as the state that existed just prior to attempting the transaction. The snapshot data store 176 may generally be any non-transient computer-readable storage medium, including but not limited to hard drives, tape drives, optical media, magnetic media, solid state devices, RAM, ROM, and the like.

In various embodiments, the transaction execution system 170 may be implemented as a component or components of a host computing device 150A or 150B, or the functionality of these devices and systems may be combined. For example, each of the host computing devices 150A-B may implement its own transaction execution system 170 as a local process that provides the recited functionality with regard to the virtual machine instances 154A-C and containers 158A-F executing on the respective host computing device 150A or 150B. Such implementations may reduce the number of inter-device interactions required to implement the described embodiments, and may thus reduce latency and overall use of computing resources. For ease of illustration, however, the transaction execution system 170 is depicted in FIG. 1 as a separate entity from the host computing devices 150A-B.

As shown in FIG. 1, various combinations and configurations of host computing devices 150A-B, virtual machine instances 154A-C, and containers 158A-F may be used to facilitate execution of user submitted code 160A-G. In the illustrated example, the host computing device 150A implements two virtual machine instances 154A and 154B. Virtual machine instance 154A, in turn, implements two containers 158A and 158B, which contain user-submitted code 160A and 160B respectively. Virtual machine instance 154B implements a single container 158C, which contains user-submitted code 160C. The host computing device 150B further implements a virtual machine instance 154C and directly implements containers 158E and 158F, which contain user-submitted code 160F and 160G. The virtual machine instance 154C, in turn, implements container 158D, which contains user-submitted codes 160D and 160E. It will be understood that these embodiments are illustrated for purposes of example, and that many other embodiments are within the scope of the present disclosure.

While some functionalities are generally described herein with reference to an individual component of the on-demand code execution system 110, other components or a combination of components may additionally or alternatively implement such functionalities. For example, a worker manager 140 may operate to provide functionality associated with execution of user-submitted code as described herein with reference to the transaction execution system 170.

FIG. 2 depicts a general architecture of a computing system (referenced as transaction execution system 170) that operates to determine whether and when all or part of a task should be re-executed within the on-demand code execution system 110. The general architecture of the transaction execution system 170 depicted in FIG. 2 includes an arrangement of computer hardware and software modules that may be used to implement aspects of the present disclosure. The hardware modules may be implemented with physical electronic devices, as discussed in greater detail below. The transaction execution system 170 may include many more (or fewer) elements than those shown in FIG. 2. It is not necessary, however, that all of these generally conventional elements be shown in order to provide an enabling disclosure. Additionally, the general architecture illustrated in FIG. 2 may be used to implement one or more of the other components illustrated in FIG. 1. As illustrated, the transaction execution system 170 includes a processor 202, input/output device interfaces 204, a network interface 206, and a data store 208, all of which may communicate with one another by way of a communication bus. The network interface 206 may provide connectivity to one or more networks or computing systems. The processor 202 may thus receive information and instructions from other computing systems or services via the network 104. The processor 202 may also communicate to and from a memory 220 and further provide output information for an optional display (not shown) via the input/output device interfaces 204. The input/output device interfaces 204 may also accept input from an optional input device (not shown).

The memory 220 may contain computer program instructions (grouped as modules in some embodiments) that the processor 202 executes in order to implement one or more aspects of the present disclosure. The memory 220 generally includes random access memory (RAM), read only memory (ROM) and/or other persistent, auxiliary or non-transitory computer readable media. The memory 220 may store an operating system 222 that provides computer program instructions for use by the processor 202 in the general administration and operation of the transaction execution system 170. The memory 220 may further include computer program instructions and other information for implementing aspects of the present disclosure. For example, in one embodiment, the memory 220 includes an interface module 224 that generates interfaces (and/or instructions therefor) for interacting with the transaction execution system 170, e.g., via an API, CLI, and/or Web interface. In addition, the memory 220 may include and/or communicate with one or more data repositories (not shown), for example, to access user program codes and/or libraries.

In addition to and/or in combination with the interface module 224, the memory 220 may include a transaction analyzer 172 and an snapshot manager 174 that may be executed by the processor 202. In one embodiment, the transaction analyzer 172 and an snapshot manager 174 individually or collectively implement various aspects of the present disclosure, e.g., analyzing code or code execution to determine transaction results, restore previous execution states, and retry code execution, as described further below.

While the transaction analyzer 172 and an snapshot manager 174 are shown in FIG. 2 as part of the transaction execution system 170, in other embodiments, all or a portion of the transaction analyzer 172 and an snapshot manager 174 may be implemented by other components of the on-demand code execution system 110 and/or another computing device. For example, in certain embodiments of the present disclosure, another computing device in communication with the on-demand code execution system 110 may include several modules or components that operate similarly to the modules and components illustrated as part of the transaction execution system 170.

The memory 220 may further include success criteria 226, which may be loaded into memory in conjunction with a user-submitted request to execute a task on the on-demand code execution system 110. The transaction analyzer 172 may illustratively apply the success criteria 226 to an output or result of executing the code to determine whether to restore a snapshot and retry code execution, as described in more detail below. The memory 220 may further include snapshots 228, which may be generated by the snapshot manager 174 and may be used to restore a previous state of a container, virtual machine instance, or other computing resource.

In some embodiments, the transaction execution system 170 may further include components other than those illustrated in FIG. 2. For example, the memory 220 may further include user-submitted code, results of executing user-submitted code, or metadata or other information that was submitted with the request, such as an indication of which portion of the code is executed to perform a transaction. FIG. 2 is thus understood to be illustrative but not limiting.

FIGS. 3A-3C depict illustrative interactions for virtualizing transaction retries in accordance with aspects of the present disclosure. With reference now to FIG. 3A, at (1), a user device 102 sends a request via a frontend 120 that an on-demand code execution system execute a particular task. The task may illustratively be associated with code that a user previously submitted to the on-demand code execution system, or in some embodiments code may be included as part of the request to execute a task. While the request is shown in FIG. 3A as stemming from the user device 102, in some embodiments, other devices may additionally or alternatively request task execution. For example, a different device external to the on-demand code execution system 110 (e.g., an auxiliary service 106) may request execution of a task. In some instances, the on-demand code execution system 110 itself may request execution of a task (e.g., by detecting fulfillment of a triggering event for the task).

At (2) the frontend 120 requests that the transaction analyzer 172 analyze the code to determine whether it includes a transaction. Generally described, a “transaction” may refer to code that attempts to access or change data and that has a possibility of success or failure depending on factors that are external to the code. At (3), the transaction analyzer 172 may illustratively identify a transaction based on metadata provided with the request, historical executions of the code, information included in the code, or other data. For example, the code may include particular statements or API calls (e.g., an instruction to access an external data store or shared resource) that the transaction analyzer 172 identifies as a transaction. As a further example, the transaction analyzer 172 may analyze previous executions of the code to determine a typical output pattern, such as a “success” status code or an expected response to a particular API call. In some embodiments, the on-demand code execution system may include code libraries or APIs in which particular calls are designated as transactional, and the transaction analyzer 172 may identify a transaction by determining that the code contains a function or API call that is known to invoke a transaction. In other embodiments, the user may designate sections of their code as being transactional in nature by inserting markers or statements (e.g., a beginTransaction or endTransaction statement) that allow the transaction analyzer 172 to identify these sections, or may provide metadata (e.g., information indicating that a transaction begins on line 72 of the code and ends on line 96) that facilitates identifying a transaction.

The transaction analyzer 172 may further determine success criteria associated with the identified transaction. In some embodiments, the success criteria may be defined or provided by the user (e.g., as metadata submitted in conjunction with the task). In other embodiments, the success criteria may be associated with obtaining a particular result that is pre-defined at the on-demand code execution system as being associated with a successful transaction or a failed transaction. For example, an API call that updates a record in a database may return a status code that indicates whether the update was successfully applied. As a further example, a command may attempt to read content from a shared data store, and may return the content if successful or silently fail if not. The transaction analyzer 172 may thus determine criteria for assessing whether the transaction succeeded based on known responses to identified transactions.

At (4), the transaction analyzer 172 reports to the frontend 120 that the code does contain a transaction, and provides any information that the frontend 120 may need to facilitate capturing a pre-transaction snapshot of the computing resources that execute the code and capturing a post-transaction result that can be analyzed to determine whether the transaction succeeded or failed. For example, the transaction analyzer 172 may identify a particular section of the code as corresponding to the transaction, and may indicate to the frontend 120 that a snapshot should be taken just prior to executing this section of the code. As a further example, the transaction analyzer 172 may identify the scope of computing resources to be included in the snapshot (e.g., that specific memory pages, register values, or other state information should be preserved).

In some embodiments, the interactions at (2), (3), and (4) may be carried out prior to receiving a request to execute the code. For example, the user may have previously submitted the code to on-demand code execution system, and the interactions at (2), (3), and (4) may be carried out at the time the code is submitted rather than waiting for a request to execute the code. In other embodiments, the interactions at (2), (3), and (4) may be carried out in conjunction with the user's initial request to execute the code, and subsequent requests to execute the code may obtain the results of these interactions rather than carrying them out again. One skilled in the art will appreciate that carrying out the interactions at (2), (3), and (4) prior to receiving a request to execute user-submitted code may reduce the time required to fulfill the request when it is received.

At (5), the frontend 120 sends a request to execute the code to the worker manager 140, and may provide all or part of the information supplied by the transaction analyzer 172. In some embodiments, the interaction at (5) may precede the interaction at (2), and the worker manager 140 may ask the transaction analyzer 172 to analyze code after receiving a request from the frontend to execute the code. In other embodiments, code may be submitted to the on-demand code execution system prior to making a request to execute the code, and the interactions at (2), (3) and (4) may be carried out at the time the code is submitted.

At (6), the worker manager 140 determines the resources that will be allocated to execute the requested code. Illustratively, the worker manager 140 may identify a host computing device (e.g., host computing device 150A), a virtual machine instance that is executing on the host computing device 150A (e.g., the virtual machine instance 154A depicted in FIG. 1), a container within a virtual machine instance (e.g., container 158A), other computing resources, or various combinations thereof.

At (7), the frontend 120 instructs the host computing device 150A to allocate the resources that were determined at (6) to execute the code containing the transaction. In various embodiments, the frontend 120 may instruct the host computing device 150A to execute the code in a new or existing virtual machine instance, a new or existing container, or to use other computing resources to execute the code.

At (8), the host computing device 150A interacts with the snapshot manager 174 to request a pre-transaction snapshot of the resources that will be allocated to the code execution request. In some embodiments, the interaction at (8) may take place while the code is being executed (e.g., during the interaction at (12)). For example, if the transaction analyzer 172 determines that a particular section of the code is associated with a transaction, then the interaction at (8) may occur when the host computing device 150A begins to execute that section. In other embodiments, the entire user-submitted code may be associated with the transaction, and the request to take a pre-transaction snapshot of resources may precede any code execution. In some embodiments, the request to take a pre-transaction snapshot may be carried out by the frontend 120 when determining or assigning the resources that will execute the code.

At (9), the snapshot manager 174 interacts with the host computing device 150A that will execute the user-submitted code to request state information. The state information may illustratively include information regarding the state of the host computing device 150A, the state of a virtual machine instance hosted on the host computing device 150A, the state of a container that is executing within a virtual machine instance (or directly on the host computing device 150A), or the state of another computing resource that is a component of or available to the host computing device 150A. At (10), the host computing device 150A provides the requested state information, which at (11) the snapshot manager 174 then stores in the snapshot data store 176. In some embodiments, the snapshot manager 174 may be a component of the host computing device 150A, and the interactions at (8), (9), (10), and (11) may be combined into fewer interactions (e.g., collecting and then storing state information).

In various embodiments, the snapshot manager 174 may preserve different types of state information relating to the computing environment in which the code will be executed. For example, the snapshot manager 174 may preserve the contents of virtual memory, the registers of a virtual processor, a local data store (e.g., a “scratch” partition or temporary folder), or other state information. In some embodiments, the snapshot manager 174 may begin logging changes to state information rather than preserving the state information. For example, on a virtual machine instance with a relatively large amount of memory, the snapshot manager 174 may implement a copy-on-write scheme and preserve only the portions of memory that are overwritten during execution of the transaction. In further embodiments, the snapshot manager 174 may determine that a virtual machine instance is in a known baseline state prior to execution of the user-submitted code, and may store information that identifies this state and facilitates returning to the baseline rather than storing particular memory contents or register values.

At (12), the host computing device 150A executes the code (or, in some embodiments, executes at least the portion of the code that corresponds to the transaction). In some embodiments, the transaction analyzer 172 may identify and warn the user regarding any code that causes “side effects” or would otherwise not be idempotent (e.g., code that increments a value or would otherwise cause a different result if run more than once). In other embodiments, the user who submits code to the on-demand code execution system must ensure that any code marked as being part of a transaction does not cause undesired side effects if executed repeatedly.

Turning now to FIG. 3B, at (13), the host computing device 150A reports the results of executing the transaction to the transaction analyzer 172. In some embodiments, the host computing device 150A may only communicate with the transaction analyzer 172 and/or the snapshot manager 174 via the frontend 120. In other embodiments, the host computing device 150A may interact directly with other components. In further embodiments, as described above, the transaction analyzer 172 and/or the snapshot manager 174 may be implemented as a component of the host computing device 150A, and the results may thus be obtained and analyzed locally rather than being reported. Further, in some embodiments, the host computing device 150A may detect that the transaction has completed without receiving or generating a report to that effect. For example, the host computing device 150A may determine that the code has reached or executed an endTransaction call, or has completed execution of a section of the code that is associated with the transaction.

At (14), in some embodiments, the transaction analyzer 172 may determine, based on the results output by the host computing device 150A and the criteria determined or obtained at (3), that the transaction did not succeed. For example, the transaction analyzer 172 may compare the result to the criteria to determine that the result contains an error code defined by the criteria (or that the result does not contain a success code defined by the criteria) and thus represents a failed transaction. The transaction analyzer 172 may thus, at (15), request that the snapshot manager 174 use the information captured during the interactions at (8), (9), and (10) to restore the host computing device 150A to its pre-transaction state. At (16), the snapshot manager 174 requests this information from the snapshot data store 176, and at (17) the snapshot data store 176 provides the information that was stored during the interaction at (10).

At (18), the snapshot manager 174 provides the snapshot information and instructs the host computing device 150A to restore the pre-transaction state of the resources that were used to execute the code. Thereafter, at (19), the host computing device 150A reverts to the pre-transaction state and re-executes the code. Illustratively, the host computing device 150A may use the information stored in the snapshot to restore memory contents, page tables, register values, and other computing resources to the state they had just prior to executing the transaction. The host computing device 150A may then repeat the interaction at (13) to resume execution of the code from the point at which the transaction began, and may again report the results of executing the code. The interactions at (14)-(19) may then be further repeated if the transaction has failed again. In some embodiments, the interactions at (13)-(19) may be repeated until a particular condition has been met, such as a maximum number of retries, a maximum time elapsed, a threshold amount of consumption of a computing resource, a “hard failure” (e.g., an indication that the transaction will never succeed), or other condition that prevents the interactions at (13)-(19) from being repeated indefinitely. For example, the host computing device 150A may keep track of the elapsed time since its first attempt to execute the transaction, and may stop further attempts to execute the transaction after a specified time period has elapsed. As a further example, the transaction analyzer 172 may apply criteria to determine that a particular error code found in the results of a failed transaction attempt (e.g., “403 Forbidden”) indicates that the failure is not intermittent and cannot be overcome by repeating the same attempt.

Turning now to FIG. 3C, an alternate embodiment is described in which, at (14′), the transaction analyzer 172 determines that the transaction was successful based on the output obtained at (13) and the criteria determined or obtained at (3). Illustratively, the interaction at (14′) may be reached after any number of repetitions of the interactions at (13)-(19), including no repetitions (i.e., the transaction may succeed on the first try). In some embodiments, the transaction analyzer 172 may, at (15′), make these results available to the user device 102. In other embodiments, the host computing device 150A may perform other actions after the transaction successfully completes. For example, the transaction analyzer 172 may report to the host computing device 150A that the transaction was successful, thereby enabling the host computing device 150A to continue executing the post-transaction portion of the code. The host computing device 150A may then, in some embodiments, make the results of executing the post-transaction code (or of executing the entire code) available to the user device 102. One skilled in the art will thus appreciate that the user-submitted code need not include any routines for handling an unsuccessful transaction, since the post-transaction portion of the code will only be reached if and when the transaction completes successfully. The user-submitted code may thus be simplified and made easier to maintain, and this shorter code will execute more quickly in scenarios where the transaction succeeds.

It will be understood that FIGS. 3A-3C are provided for purposes of example, and that many variations on the depicted interactions are within the scope of the present disclosure. For example, as described above, the transaction analyzer 172 may determine and provide information that allows the host computing device 150A to assess whether a transaction was successful, and the interactions at (13), (14), and (15) may thus be internal to the host computing device 150A rather than being carried out via the transaction analyzer 172. As a further example, the frontend 120 may act as an intermediary between the host computing device 150A and other computing devices (e.g., the auxiliary services 106 or data storage services 108 depicted in FIG. 1), and may determine whether transactions are successful by analyzing the responses received from the external computing devices. FIGS. 3A-3C are thus understood to be illustrative and not limiting.

FIG. 4 is a flow diagram of an illustrative routine 400 for executing user-submitted code that contains a transaction. The routine may be carried out, for example, by the transaction analyzer 172, the snapshot manager 174, or various combinations thereof. The routine 400 begins at block 402, where code to be executed and criteria for measuring successful execution may be obtained. In various embodiments, as described above, the criteria for measuring successful transaction execution may be obtained from the user, determined by analyzing the code, determined by analyzing previous executions of the code, defined in terms of, e.g., API calls returning particular result codes, and so forth.

At block 404, the start of a transaction may be detected. As described above, the start of a transaction may correspond to reaching a particular section of the code during execution, such as an API call that is known to invoke a transaction or a “beginTransaction” statement in the code. In some embodiments, the start of a transaction may correspond to the start of code execution, and a snapshot may thus be taken before any code is executed. In other embodiments, the start of a transaction may be determined based on activities performed by the code rather than an analysis of the code itself. For example, a transaction may be detected when the code attempts to obtain exclusive access to a shared resource, or when the code attempts to write a record to a database.

At block 406, a snapshot of the execution environment may be taken. In some embodiments, as described above, a change log may be created and maintained such that the pre-transaction execution environment can be recreated based on the change log. For example, the “snapshot” may be implemented as a copy-on-write scheme that retains information being overwritten from that point forward. In other embodiments, contents of memory pages, values of registers, network configurations, or other state information may be captured and stored. For example, where a task is executing in a virtual machine instance execution environment, a snapshot may be created by utilizing a “snapshot” functionality of a host operating system (e.g., a hypervisor), which functionality is known in the art. The execution environment may illustratively include any configuration of physical and/or virtual computing devices, including virtual machine instances, containers, host computing devices, data stores, and the like.

At block 408, the code obtained at block 402 may be executed in the execution environment. In some embodiments, a portion of the code may be executed prior to taking a snapshot at block 406 or may be executed prior to carrying out the routine 400. In further embodiments, only a portion of the code may be executed at block 408. For example, a transaction may be associated with a particular subroutine, statement, API call, block, function, or other portion of the code. At block 410, a result of executing the code may be obtained. In various embodiments, the result may include a status code (e.g., “200,” “503,” etc.) an object, a message, a register value, a pointer, or in some embodiments may be null (which may indicate success or failure). In some embodiments, a result may be obtained by inspecting the execution environment. For example, the contents of a virtual machine instance's memory, CPU registers, etc. may be analyzed to determine the outcome of executing the code. In further embodiments, the result may be obtained by monitoring communications between the execution environment and an external resource (e.g., a database, a storage service, etc.), or by monitoring the external resource directly.

At block 412, the execution results may be evaluated against the criteria obtained or determined at block 402. Illustratively, the results may be analyzed to determine whether they have a particular format, size, status code, error message, content, or otherwise compared to the success criteria. At decision block 414, a determination may be made as to whether the results satisfy the success criteria. If so, then at block 416 the code execution may be continued (or, in embodiments where the entire code corresponds to a transaction, completed) and the routine 400 ends. If not, then at decision block 418 a determination may be made as to whether the transaction should be retried. Illustratively, the determination may be as to whether a threshold number of retries has been exceeded, a threshold amount of time has elapsed, that the result indicates a “permanent” failure (e.g., a “403 Forbidden” error code), or that some other condition has been met. In some embodiments, a user-specified condition may be obtained and the determination at decision block 414 may be as to whether this condition has been met.

If the determination at decision block 418 is that the transaction should not be retried, then at block 420 the transaction's failure to execute may be reported. Illustratively, a notification may be provided to a user requesting that the code be executed, or to the computing device from which a request was obtained. In some embodiments, other code may be obtained and executed in response to a transaction failure. If the determination at decision block 418 is that the transaction should be retried, then the routine 400 branches to block 422, where in some embodiments a retry count may be incremented, and then to block 424, where the pre-transaction execution environment may be restored. The routine 400 then returns to block 408, and re-executes the code iteratively until either an attempt meets the success criteria or a determination is made that no further retries should be attempted.

In various embodiments, the transaction execution routine 400 may include more, fewer, different, or different combinations of blocks than those depicted in FIG. 4. For example, the routine 400 may, in some embodiments, output the results of executing the transaction rather than continuing execution at block 414. As a further example, in embodiments where the entire code corresponds to a transaction, block 414 may be omitted and the routine 400 may end after the transaction completes successfully. As a further example, blocks 408, 410, and 412 may be combined, and the routine 400 may evaluate execution results as it obtains them (e.g., by determining whether a “success” status code can be obtained). The routine 400 depicted in FIG. 4 is thus understood to be illustrative and not limiting.

It is to be understood that not necessarily all objects or advantages may be achieved in accordance with any particular embodiment described herein. Thus, for example, those skilled in the art will recognize that certain embodiments may be configured to operate in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other objects or advantages as may be taught or suggested herein.

All of the processes described herein may be embodied in, and fully automated via, software code modules, including one or more specific computer-executable instructions, that are executed by a computing system. The computing system may include one or more computers or processors. The code modules may be stored in any type of non-transitory computer-readable medium or other computer storage device. Some or all the methods may be embodied in specialized computer hardware.

Many other variations than those described herein will be apparent from this disclosure. For example, depending on the embodiment, certain acts, events, or functions of any of the algorithms described herein can be performed in a different sequence, can be added, merged, or left out altogether (e.g., not all described acts or events are necessary for the practice of the algorithms). Moreover, in certain embodiments, acts or events can be performed concurrently, e.g., through multi-threaded processing, interrupt processing, or multiple processors or processor cores or on other parallel architectures, rather than sequentially. In addition, different tasks or processes can be performed by different machines and/or computing systems that can function together.

The various illustrative logical blocks and modules described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a processing unit or processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor can be a microprocessor, but in the alternative, the processor can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor can include electrical circuitry configured to process computer-executable instructions. In another embodiment, a processor includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor may also include primarily analog components. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computational engine within an appliance, to name a few.

Conditional language such as, among others, “can,” “could,” “might,” or “may,” unless specifically stated otherwise, are otherwise understood within the context as used in general to convey that certain embodiments include, while other embodiments do not include, certain features, elements and/or steps. Thus, such conditional language is not generally intended to imply that features, elements and/or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without user input or prompting, whether these features, elements and/or steps are included or are to be performed in any particular embodiment.

Disjunctive language such as the phrase “at least one of X, Y, or Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and/or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each be present.

Any process descriptions, elements or blocks in the flow diagrams described herein and/or depicted in the attached figures should be understood as potentially representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or elements in the process. Alternate implementations are included within the scope of the embodiments described herein in which elements or functions may be deleted, executed out of order from that shown, or discussed, including substantially concurrently or in reverse order, depending on the functionality involved as would be understood by those skilled in the art.

Unless otherwise explicitly stated, articles such as “a” or “an” should generally be interpreted to include one or more described items. Accordingly, phrases such as “a device configured to” are intended to include one or more recited devices. Such one or more recited devices can also be collectively configured to carry out the stated recitations. For example, “a processor configured to carry out recitations A, B, and C” can include a first processor configured to carry out recitation A working in conjunction with a second processor configured to carry out recitations B and C.

Claims

1. A system comprising: a non-transitory data store configured to store computer-executable instructions; anda computing device in communication with the non-transitory data store, wherein the computer-executable instructions, when executed by the computing device, configure the computing device to: obtain user-submitted code executable on an on-demand code execution system, wherein the user-submitted code includes a subsection that, when executed by the on-demand code execution system, configures the on-demand code execution system to attempt a transaction; andin response to a request to execute the user-submitted code: cause the on-demand code execution system to configure a virtual machine instance to execute the user-submitted code;generate a snapshot of the virtual machine instance when execution of the user-submitted code reaches a first location in the user-submitted code, the first location corresponding to the start of the subsection;cause the on-demand code execution system to perform a first execution of the subsection within the virtual machine instance;obtain a first processing result, wherein the first processing result is generated based at least in part on the execution of the user-submitted code reaching a second location in the user-submitted code, the second location corresponding to the end of the subsection;determine, based at least in part on the first processing result, that the transaction did not succeed; andcause the on-demand code execution system to restore the snapshot of the virtual machine instance, wherein restoring the snapshot of the virtual machine instance changes an execution state of the user-submitted code from a state corresponding to the second location to a state corresponding to the first location, and wherein restoring the snapshot of the virtual machine instance causes the on-demand code execution system to resume execution of the user-submitted code within the virtual machine instance from the first location in order to perform a second execution of the subsection.

2. The system of claim 1, wherein the computing device is further configured to obtain a second processing result corresponding to the second execution of the user-submitted code.

3. The system of claim 2, wherein the computing device is further configured to: determine, based at least in part on the second processing result, that the transaction succeeded; andcause the on-demand code execution system to continue execution of the user-submitted code from the second location.

4. The system of claim 2, wherein the computing device is further configured to: determine, based at least in part on the second processing result, that the transaction did not succeed; andcause the on-demand code execution system to restore the snapshot of the virtual machine instance a second time; andcause the on-demand code execution system to perform a third execution of the user-submitted code.

5. The system of claim 1, wherein the transaction comprises one or more of reading data from a database, writing data to a database, reading data from a storage device, writing data to a storage device, obtaining a resource, or communicating via a network.

6. A computer-implemented method comprising: obtaining user-submitted code executable on an on-demand code execution system;in response to a request to execute the user-submitted code: causing the on-demand code execution system to configure a virtual computing environment to execute the user-submitted code;generating a snapshot of the virtual computing environment when execution of the user-submitted code reaches a first location in the user-submitted code, the first location being prior to a portion of the user-submitted code;causing the on-demand code execution system to perform a first execution of the portion of the user-submitted code within the virtual computing environment;determining, based at least in part on a result obtained from the first execution, that the first execution was not successful, wherein the result is generated based at least in part on the execution of the user-submitted code reaching a second location in the user-submitted code, the second location being associated with an end of the portion of the user-submitted code;causing the on-demand code execution system to restore the snapshot of the virtual computing environment, wherein restoring the snapshot of the virtual computing environment changes an execution state of the user-submitted code from a state corresponding to the second location to a state corresponding to the first location; andcausing the on-demand code execution system to perform a second execution of the portion of the user-submitted code within the virtual computing environment.

7. The computer-implemented method of claim 6, wherein the snapshot includes one or more of memory contents, register values, or content of a data store.

8. The computer-implemented method of claim 6, wherein the snapshot includes a change log.

9. The computer-implemented method of claim 6 further comprising: determining that a condition associated with repeated executions of the portion of the user-submitted code has been satisfied; andnotifying a user associated with the user-submitted code that a success criterion has not been satisfied.

10. The computer-implemented method of claim 9, wherein the condition comprises a threshold number of executions, a threshold resource consumption, an elapsed time, or a determination that a success criterion will not be satisfied by further executions.

11. The computer-implemented method of claim 6 further comprising causing the on-demand code execution system to begin executing the user-submitted code within the virtual computing environment prior to generating the snapshot.

12. The computer-implemented method of claim 6 further comprising analyzing the user-submitted code to identify the at least a portion of the user-submitted code.

13. The computer-implemented method of claim 6, wherein the at least a portion of the user-submitted code comprises a function call.

14. The computer-implemented method of claim 6, further comprising determining a success criterion based at least in part on one or more of an analysis of the user-submitted code or a previous execution of the user-submitted code, wherein determining that the first execution was not successful is based on the success criterion.

15. The computer-implemented method of claim 6, wherein the at least a portion of the user-submitted code is associated with an attempt to invoke functionality outside the virtual computing environment.

16. The computer-implemented method of claim 6, wherein the result indicates that a resource is temporarily unavailable.

17. The computer-implemented method of claim 6, wherein the result indicates one or more of a read error, a write error, a network error, network congestion, or rate limiting.

18. Non-transitory computer-readable media including computer-executable instructions that, when executed by a processor, cause the processor to perform operations including: configuring a virtual computing environment to execute user-submitted code, wherein at least a portion of the user-submitted code is associated with a success criterion;obtaining a state of the virtual computing environment when execution of the user-submitted code reaches a first location in the user-submitted code, the first location being prior to a first execution of the at least a portion of the user-submitted code within the virtual computing environment;determining, based at least in part on a result obtained when the execution of the user-submitted code reaches a second location in the user-submitted code, the second location being after the at least a portion of the user-submitted code, that the success criterion has not been satisfied;restoring the state of the virtual computing environment to that obtained when the execution of the user-submitted code reached the first location in the user-submitted code; andperforming a second execution of the at least a portion of the user-submitted code within the virtual computing environment.

19. The non-transitory computer-readable media of claim 18, wherein the operations further include performing the first execution of the at least a portion of the user-submitted code within the virtual computing environment.

20. The non-transitory computer-readable media of claim 18, wherein the operations further include determining the success criterion.

21. The non-transitory computer-readable media of claim 18, wherein the operations further include analyzing the user-submitted code to identify a transaction that the user-submitted code attempts when executed.

22. The non-transitory computer-readable media of claim 18, wherein the operations are performed at a host computing device that hosts the virtual computing environment.