Compute Services Flashcards
Launch instances page that displays popular up-to-date AMIs and includes various releases of Linus and Windows servers.
Amazon Quick Start AMIs
Independent vendors create and maintain AMIs, usually built to meet specific needs.
Community AMIs
Deployed instances are in a single availability zone close to each other, providing low-latency network interconnectivity and useful for HPC applications.
Cluster groups
To minimize the risk of losing data or services due to failures, it is important to physically separate instances across different hardware racks and availability zones. This approach is particularly valuable for hosting environments that cannot afford to experience multiple failures at the same time.
Spread groups
You can group instances and keep them physically separated from the others and never share physical hardware.
Partition groups
A network drive you can attach to your instances to allow for persistent data after an instance has been terminated. They can only be mounted to one instance at a time.
Elastic Block Store (EBS)
Create a policy that can automatically create, keep, and frequently delete snapshots of the storage volumes.
Amazon’s EBS Data Lifecycle Manager
A low latency, cost-effective storage for transactional workloads
General Purpose SSD
For storage for critical applications that need sub-millisecond latency
Provisioned SSD
Storage for large data sets where data access is not critical is primarily used for extensive data warehousing and log processing.
Throughput Optimized HDD
A cost-effective storage option for infrequently accessed data.
Cold HDD
Provisioned deployment allows up to 16 instances to share the same storage, providing higher application availability as needed.
Multi Attached Storage
Ephemeral storage that provides a fast, nonvolatile connectivity
Instance Store Volumes
Examining the denial of incoming traffic and allowance of outgoing traffic at the instance level is based on the specific ports and protocols being used to control traffic statefully
Security Groups
Through this usage, we can ensure a properly secure encrypted session, save the public key to your EC2 server, and save its private half to your local machine
Key Pair
The named document contains the preselected information you proved when manually provisioning an instance, and it cannot be changed once it is deployed.
Launch Configurations
The named document contains the preselected information you proved when manually provisioning an instance, allowing versatility to be changed through versioning once deployed.
Launch Templates
Detects problems related to the host instance, such as memory exhaustion, file corruption, network configuration errors, or startup configuration problems.
Health Checks
Metric that ensures that healthy instances never go below this number. Setting a metric of ‘0’ will remove instances
Scaling Policy Minimum
By utilizing a launch configuration, we have the ability to establish a group of instances that will continuously uphold the specific number of instances for an indefinite period.
Auto Scaling Groups
The metric ensures that the number of healthy instances does not exceed a certain threshold. It involves setting an expected amount to maintain within the budget.
Scaling Policy Maximum
Optional configuration to maintain a desired number of instances. AWS aims to maintain this threshold in case instances fail.
Scaling Policy Desired Capacity
Monitor a specific metric in real-time. If this metric surpasses the predefined threshold, the autoscaling feature will automatically adjust the desired capacity while accounting for the cool-down period.
Simple scaling policy
Even if the cloud watch metric violates the cloud watch policy, we will wait 5 minutes before implementing another simple scaling policy.
Cool-down period
An advanced technique for adding instances based on specific metrics and parameters enables precise and detailed addition and adjustments of instances.
Step Scaling policy (upper/lower bounds)
Track a particular metric in real time. If this metric exceeds the predetermined threshold, the autoscaling feature will automatically modify the desired capacity. It will also scale in and remove instances if the metric decreases to maintain the desired capacity.
Target tracking policy
Proactively ensure that enough instances are provided before the demand hits for predetermined workload patterns.
Scheduled scaling policy
You can automatically or manually perform (Or bulk) actions against your AWS resources, whether in the cloud or on-prem, ranging from installation to upgrades.
AWS Systems Manager
By providing an instance profile role containing permissions, you can run commands on your managed instance, whether on-prem or in the cloud.
Systems Manager Agent
For Linux and Windows instances, you can access bash or PowerShell without needing to enable any security groups or NACLs or move an instance to a public subnet.
Session Manager
Facilitates the automation of patches for Linux and Windows instances, leveraging AWS-RunPatchBaseline as a baseline to define which patches to install.
Patch Manager
Although you can individually patch an instance, you can also patch a collection of instances with the tag
Patch Group
A configuration management tool ensures that your instances have the required software and are configured according to your specifications. It automatically executes commands and policy documents on your instances, either once or on a schedule.
State Manager
Aggregate health compliance and operational details about your AWS resources
Insights
Collects data from your instance, including operating systems and application versions.
AWS Systems Manager Inventory
Provision EC2 instances and maintain the infrastructure within a cluster. Each instance has an agent that monitors the running services.
Amazon ECS
Open source container orchestrator tool that manages the control plane of your cluster, enabling you to manage clusters with high availability in multiple availability zones with additional master nodes.
Amazon EKS
A serverless method to based on task definitions to launch containers, allowing AWS to handle the provisioning of resources. Billing is based on your resource consumption.
Amazon Fargate
The ECS agent uses EC2 instance profiles to make API calls to ECS services, such as sending logs to CloudWatch and referencing sensitive data in Secrets Manager.
IAM roles for ECS
Permission is granted for each task to have a specific role for interacting with the services you are running, as defined in the task definition.
ECS task roles
Fully integrated with Amazon S3, supporting vulnerability scanning, versioning, and image lifecycle to store and manage your docker images on AWS.
Amazon ECR
