AWS deadline-cloud documentation change
Summary
Added documentation for 'License endpoint' concept explaining how it enables secure usage-based licensing within a VPC.
Security assessment
The license endpoint documentation describes a security-related feature (isolating licensing traffic in a VPC), but there is no evidence this addresses a specific security vulnerability. It adds general security documentation about network isolation practices.
Diff
diff --git a/deadline-cloud/latest/developerguide/concepts-terminology.md b/deadline-cloud/latest/developerguide/concepts-terminology.md index 37382a849..5adf9bd24 100644 --- a//deadline-cloud/latest/developerguide/concepts-terminology.md +++ b//deadline-cloud/latest/developerguide/concepts-terminology.md @@ -4,0 +5,2 @@ +Farm resourcesJob execution resourcesOther important concepts and terminology + @@ -9 +11 @@ To help you get started with AWS Deadline Cloud, this topic explains some of its -**Budget manager** +## Farm resources @@ -10,0 +13 @@ To help you get started with AWS Deadline Cloud, this topic explains some of its +This diagram shows how Deadline Cloud farm resources work together. @@ -12,3 +15 @@ To help you get started with AWS Deadline Cloud, this topic explains some of its -Budget manager is part of the Deadline Cloud monitor. Use the budget manager to create and manage budgets. You can also use it to limit activities to stay within budget. - -**Deadline Cloud Client Library** + @@ -15,0 +17 @@ Budget manager is part of the Deadline Cloud monitor. Use the budget manager to +**Farm** @@ -17 +18,0 @@ Budget manager is part of the Deadline Cloud monitor. Use the budget manager to -The Client Library includes a command line interface and library for managing Deadline Cloud. Functionality includes submitting job bundles based on the Open Job Description specification to Deadline Cloud, downloading job attachment outputs, and monitoring your farm using the command line interface. @@ -19 +20 @@ The Client Library includes a command line interface and library for managing De -**Digital content creation application (DCC)** +A farm contains all other resources related to submitting and running jobs. Farms are independent from each other making them useful for separating production environments. @@ -20,0 +22 @@ The Client Library includes a command line interface and library for managing De +**Queue** @@ -22 +23,0 @@ The Client Library includes a command line interface and library for managing De -Digital content creation applications (DCCs) are third-party products where you create digital content. Examples of DCCs are Maya, Nuke, and Houdini. Deadline Cloud provides job submitter integrated plugins for specific DCCs. @@ -24 +25 @@ Digital content creation applications (DCCs) are third-party products where you -**Farm** +A queue holds jobs for scheduling on associated fleets. Users can submit jobs to a queue and manage their priority and status inside the queue. A queue must be associated with a fleet with a queue-fleet association for its jobs to be run, and queues can be associated with multiple fleets. @@ -25,0 +27 @@ Digital content creation applications (DCCs) are third-party products where you +**Fleet** @@ -27 +28,0 @@ Digital content creation applications (DCCs) are third-party products where you -A farm is a where your project resources are located. It consists of queues and fleets. @@ -29 +30 @@ A farm is a where your project resources are located. It consists of queues and -**Fleet** +A fleet contains compute capacity for running jobs. Fleets can be service-managed or customer-managed. Service-managed fleets run in Deadline Cloud and include built-in functionality like autoscaling, licensing, and software access. Customer-managed fleets run on your own compute resources like Amazon EC2 instances or on-premises servers. @@ -30,0 +32 @@ A farm is a where your project resources are located. It consists of queues and +**Budget** @@ -32 +33,0 @@ A farm is a where your project resources are located. It consists of queues and -A fleet is a group of worker nodes that do the rendering. Worker nodes process jobs. A fleet can be associated to multiple queues, and a queue can be associated to multiple fleets. @@ -34 +35 @@ A fleet is a group of worker nodes that do the rendering. Worker nodes process j -**Instance** +A budget sets spending thresholds for your job activity and allows you to take actions when thresholds are reached, such as stopping job scheduling. @@ -35,0 +37 @@ A fleet is a group of worker nodes that do the rendering. Worker nodes process j +**Queue environment** @@ -37 +38,0 @@ A fleet is a group of worker nodes that do the rendering. Worker nodes process j -Fleets use instances for CPU resources. An instance is an Amazon EC2 performance instance. Deadline Cloud uses On-Demand and Spot instances. @@ -39 +40 @@ Fleets use instances for CPU resources. An instance is an Amazon EC2 performance -**On-Demand instance** +A queue environment defines scripts that run on each worker to set up or tear down the workload environment. They are useful for setting environment variables, installing software, and configuring asset storage. @@ -40,0 +42 @@ Fleets use instances for CPU resources. An instance is an Amazon EC2 performance +**Storage profile** @@ -42 +43,0 @@ Fleets use instances for CPU resources. An instance is an Amazon EC2 performance -On-Demand instances are priced by the second, have no long-term commitment, and will not be interrupted. @@ -44 +45 @@ On-Demand instances are priced by the second, have no long-term commitment, and -**Spot instance** +A storage profile is a configuration for a group of hosts and workstations, that tells where data is located on the file system. Deadline Cloud uses storage profiles to map paths when running jobs on differently configured hosts, such as a job submitted from Windows and running on Linux. @@ -45,0 +47 @@ On-Demand instances are priced by the second, have no long-term commitment, and +**Limit** @@ -47 +48,0 @@ On-Demand instances are priced by the second, have no long-term commitment, and -Spot instances are unreserved capacity that you can use at a discounted price, but may be interrupted by On-Demand requests. @@ -49 +50 @@ Spot instances are unreserved capacity that you can use at a discounted price, b -**Wait and Save** +A limit allows you to track usage of shared resources such as floating licenses and control how they are allocated between jobs. Limits are associated with queues with queue-limit associations. @@ -50,0 +52 @@ Spot instances are unreserved capacity that you can use at a discounted price, b +**Monitor** @@ -52 +53,0 @@ Spot instances are unreserved capacity that you can use at a discounted price, b -The Wait and Save feature provides delayed job scheduling for lower cost and can be interrupted by On-Demand and Spot requests. Wait and Save is only available within Deadline Cloud service-managed fleets. @@ -54 +55 @@ The Wait and Save feature provides delayed job scheduling for lower cost and can -Wait and Save is for managing the execution of visual computing workloads in AWS Deadline Cloud. See [AWS service terms](https://aws.amazon.com/service-terms/) for details. +The monitor configures the URL for the Deadline Cloud monitor web application, allowing end users to monitor and manage jobs. It can be accessed in a browser or through the Deadline Cloud monitor desktop application. @@ -56 +57 @@ Wait and Save is for managing the execution of visual computing workloads in AWS -**Job** +## Job execution resources @@ -57,0 +59 @@ Wait and Save is for managing the execution of visual computing workloads in AWS +This diagram shows how Deadline Cloud job resources work together. @@ -59 +61 @@ Wait and Save is for managing the execution of visual computing workloads in AWS -A job is a rendering request. Users submit jobs. Jobs contain specific job properties that are outlined as steps and tasks. + @@ -61 +63 @@ A job is a rendering request. Users submit jobs. Jobs contain specific job prope -**Job attachments** +**Job** @@ -64 +66 @@ A job is a rendering request. Users submit jobs. Jobs contain specific job prope -A job attachment is a Deadline Cloud feature that you can use to manage inputs and outputs for jobs. Job files are uploaded as job attachments during the rendering process. These files can be textures, 3D models, lighting rigs, and other similar items. +A job is a set of work that a user submits to Deadline Cloud to be scheduled and run on available workers. A job may render a 3D scene or run a simulation. Jobs are created from reusable job templates, which define the runtime environment and processes, and job-specific parameters. Jobs contain steps and tasks that define the work to be performed, and they can be configured with priorities, maximum worker counts, and retry settings. @@ -76 +78 @@ Job properties are settings that you define when submitting a render job. Some e -**Job template** +**Step** @@ -79 +81 @@ Job properties are settings that you define when submitting a render job. Some e -A job template defines the runtime environment and all processes that run as part of a Deadline Cloud job. +A step is part of a job that provides a template for running many tasks that are identical except for the task parameter values. Steps can have dependencies on other steps, allowing you to create complex workflows with sequential or parallel execution paths. In rendering jobs, a step often defines the command for rendering a frame and uses the frame number as the task parameter. @@ -81 +83 @@ A job template defines the runtime environment and all processes that run as par -**Queue** +**Task** @@ -84 +86 @@ A job template defines the runtime environment and all processes that run as par -A queue is where submitted jobs are located and scheduled to be rendered. A queue must be associated with a fleet to create a successful render. A queue can be associated with multiple fleets. +A task is the smallest unit of work in Deadline Cloud. Tasks are part of steps and are executed by workers, representing individual operations that need to be performed as part of a job. Tasks can be configured with specific parameters and are assigned to workers based on their capabilities and availability. In rendering jobs, a task often renders a single frame. @@ -86 +88 @@ A queue is where submitted jobs are located and scheduled to be rendered. A queu -**Queue-fleet association** +**Worker** @@ -89 +91 @@ A queue is where submitted jobs are located and scheduled to be rendered. A queu -When a queue is associated with a fleet, there is a queue-fleet association. Use an association to schedule workers from a fleet to jobs in that queue. You can start and stop associations to control scheduling of work. +Workers are part of a fleet and execute tasks from jobs. Workers can be configured with specific capabilities such as GPU accelerators, CPU architecture, and operating system. In service-managed fleets, workers are created automatically as the fleet scales out and in. @@ -91 +93,11 @@ When a queue is associated with a fleet, there is a queue-fleet association. Use -**Session** +**Instance** + + +Fleets use instances for CPU resources. An instance is an Amazon EC2 performance instance. Deadline Cloud uses On-Demand and Spot instances. + +**On-Demand instance** + + +On-Demand instances are priced by the second, have no long-term commitment, and will not be interrupted. + +**Spot instance** @@ -94 +106,10 @@ When a queue is associated with a fleet, there is a queue-fleet association. Use -A session is an ephemeral runtime environment on a worker host created to run a set of tasks from the same job. The session ends when the worker host finishes running tasks for that job. +Spot instances are unreserved capacity that you can use at a discounted price, but may be interrupted by On-Demand requests. + +**Wait and Save** + + +The Wait and Save feature provides delayed job scheduling for lower cost and can be interrupted by On-Demand and Spot requests. Wait and Save is only available within Deadline Cloud service-managed fleets. + +Wait and Save is for managing the execution of visual computing workloads in AWS Deadline Cloud. See [AWS service terms](https://aws.amazon.com/service-terms/) for details. + +**Session** @@ -96 +117,2 @@ A session is an ephemeral runtime environment on a worker host created to run a -The session provides a way to configure the environment with resources shared across multiple task runs, such as defining environment variables or starting a background process or container. + +A session represents a worker's sequence of work on a job. During a single session, a worker may be assigned multiple tasks which it runs one after another. Sessions often have setup actions which configure environments and load assets before running the task actions. @@ -101 +123 @@ The session provides a way to configure the environment with resources shared ac -A session action is a discrete unit of work executed by a worker within a session. It can encompass the core task run operations of a task, or it might include preparatory steps such as environment setup and post-execution processes like tear-down and cleanup. +A session action represents specific operations performed during a session such as setting up the environment, running a task, and syncing assets. @@ -103 +125 @@ A session action is a discrete unit of work executed by a worker within a sessio -**Step** +## Other important concepts and terminology @@ -104,0 +127 @@ A session action is a discrete unit of work executed by a worker within a sessio +**Usage explorer** @@ -106 +128,0 @@ A session action is a discrete unit of work executed by a worker within a sessio -A step is one particular process to run in the job. @@ -108 +130 @@ A step is one particular process to run in the job. -**Deadline Cloud submitter** +Usage explorer is a feature of Deadline Cloud monitor. It provides an approximate estimate of your costs and usage. @@ -109,0 +132 @@ A step is one particular process to run in the job. +**Budget manager** @@ -111 +133,0 @@ A step is one particular process to run in the job. -A Deadline Cloud submitter is a digital content creation (DCC) plugin. Artists use it to submit jobs from a third-party DCC interface that they are familiar with. @@ -113 +135 @@ A Deadline Cloud submitter is a digital content creation (DCC) plugin. Artists u -**Tags** +Budget manager is part of the Deadline Cloud monitor. Use the budget manager to create and manage budgets. You can also use it to limit activities to stay within budget. @@ -114,0 +137 @@ A Deadline Cloud submitter is a digital content creation (DCC) plugin. Artists u +**Deadline Cloud client library** @@ -116 +138,0 @@ A Deadline Cloud submitter is a digital content creation (DCC) plugin. Artists u -A tag is a label that you can assign to an AWS resource. Each tag consists of a key and an optional value that you define. @@ -118 +140 @@ A tag is a label that you can assign to an AWS resource. Each tag consists of a -With tags, you can categorize your AWS resources in different ways. For example, you could define a set of tags for your account’s Amazon EC2 instances that help you track each instance’s owner and stack level. +The open-source client library includes a command line interface and library for managing Deadline Cloud. Functionality includes submitting job bundles based on the Open Job Description specification to Deadline Cloud, downloading job attachment outputs, and monitoring your farm using the command line interface (CLI). @@ -120 +142 @@ With tags, you can categorize your AWS resources in different ways. For example, -You can also categorize your AWS resources by purpose, owner, or environment. This approach is useful when you have many resources of the same type. You can quickly identify a specific resources based on the tags that you've assigned to it. +**Digital content creation application (DCC)** @@ -122 +143,0 @@ You can also categorize your AWS resources by purpose, owner, or environment. Th -**Task** @@ -123,0 +145 @@ You can also categorize your AWS resources by purpose, owner, or environment. Th +Digital content creation applications (DCCs) are third-party products where you create digital content. Deadline Cloud has built-in integrations with many DCCs such as Autodesk Maya, Blender, and Maxon Cinema 4D allowing you to submit jobs from within the DCC and render on service-managed fleets with pre-configured software and licensing. @@ -125 +147 @@ You can also categorize your AWS resources by purpose, owner, or environment. Th -A task is a single component of a render step. +**Job attachments** @@ -127 +148,0 @@ A task is a single component of a render step. -**Usage-based licensing (UBL)** @@ -128,0 +150 @@ A task is a single component of a render step. +Job attachments are a Deadline Cloud feature that you upload and download assets as part of a job such as textures, 3D models, and lighting rigs. Job attachments are stored in Amazon S3 and avoid the need for shared network storage. @@ -130 +152 @@ A task is a single component of a render step. -Usage-based licensing (UBL) is an on-demand licensing model that is available for select third-party products. This model is pay as your go, and you are charged for the number of hours and minutes that you use. +**Job template** @@ -132 +153,0 @@ Usage-based licensing (UBL) is an on-demand licensing model that is available fo -**Usage explorer** @@ -133,0 +155 @@ Usage-based licensing (UBL) is an on-demand licensing model that is available fo +A job template defines the runtime environment and all processes that run as part of a Deadline Cloud job. @@ -135 +157 @@ Usage-based licensing (UBL) is an on-demand licensing model that is available fo -Usage explorer is a feature of Deadline Cloud monitor. It provides an approximate estimate of your costs and usage. +**Deadline Cloud submitter** @@ -137 +158,0 @@ Usage explorer is a feature of Deadline Cloud monitor. It provides an approximat -**Worker** @@ -138,0 +160,3 @@ Usage explorer is a feature of Deadline Cloud monitor. It provides an approximat +A Deadline Cloud submitter is a plugin for a DCC that allows users to easily submit jobs from within the DCC. + +**License endpoint** @@ -140 +164,12 @@ Usage explorer is a feature of Deadline Cloud monitor. It provides an approximat -Workers belong to fleets and run Deadline Cloud assigned tasks to complete steps and jobs. Workers store the logs from task operations in Amazon CloudWatch Logs. Workers can also use the job attachments feature to sync inputs and outputs to an Amazon Simple Storage Service (Amazon S3) bucket. + +A license endpoint makes Deadline Cloud's usage-based licensing for third-party products available inside your VPC. This model is pay as you go, and you are charged for the number of hours and minutes that you use. License endpoints are not connected to farms and can be used independently. + +**Tags** + + +A tag is a label that you can assign to an AWS resource. Each tag consists of a key and an optional value that you define. With tags, you can categorize your AWS resources in different ways, such as by purpose, owner, or environment.