Overview of AWS Flashcards
What are the six advantages of Cloud Computing?
- Variable expense rather than capitol expense (Pay for what you need, rather than an upfront expense for everything you might need).
- Economies of Scale (AWS has hundreds of thousands of customers on their cloud infrastructure, meaning they can achieve better economies of scale, which translates to lower aggregate expenses of customers than if they operated data centers themselves).
- Stop having to guess capacity needs (AWS is easy to scale, meaning that if your needs exceed your current capacity, you can easily scale up resources in a very short amount of time).
- Increase speed an agility (Make resources available to developers in just minutes).
- You do not have to spend money and time maintaining data centers (focus on what your business does best, rather than infrastructure).
- Go global in minutes (using AWS’s global network, you can deliver content and resources globally with the flip of a switch).
What Kinds of EC2 Instances are there?
General Purpose: A good balance of compute, memory, and networking resources. They are good for application servers, gaming servers, enterprise applications, and small to medium databases.
Compute optimized: Focus on compute power. Good for gaming, scientific modeling, high performance computing.
Memory Optimized: Good for high performance Databases.
Accelerated Computing: Use Hardware Accelorators. good for floating point number calculations, Graphics processing, Data Pattern Matching.
Storage Optimise: good for applications that require high performance localized storage.
Types of EC2 Pricing
On Demand: Pay for the duration that you instance runs for.
Amazon EC2 Savings Plus: Low prices in exchange for a committment to use a consistent amount of compute resources over a 1-3 year period.
Reserved Instances: Best for steady state usages. You commit to a 1 or 3 year term.
Spot instances: Request spare capacity for up to 90% off the on demand price. AWS can reclaim the instance anytime, if they need it.
Dedicated Hosts: for meeting certain compliance requirments. You will not share resources with other instances.
Types of EC2 Auto Scaling and parameters.
Amazon EC2 Auto Scaling enables you to automatically add or remove Amazon EC2 instances in response to changing application demand.
Dynamic Scaling responds to demand.
Predictive Scaling automatically schedules the right number of amazon EC2 instances based on predicted demand.
Minimum capacity, is the number of instances that launch immediately when creating an auto scaling group.
Desired capacity, will be the default number of instances if different from minimum capacity.
Maximum capacity sets a hard limit on how many instances you can have in the group.
Load Balancer
A tool that distributes traffic to different EC2 instances. It can scale with EC2 auto scaling groups.
It can serve as a single point of contact between front end and back end instances, as they scale up and down.
Amazon Simple Queue Service
Amazon Simple Notification Service
SQS: send, store, receive messages between software components at any volume. This is without losing messages or requiring other services to be available.
SNS: Shares messages through a publish / subscribe model. Create a topic, and then subscribers to that topic will receive messages that are published to that topic. Subscribers can be endpoints like SQS queues, HTTP Web hooks, or even end uses, using SMS, Email and others.
AWS Lamda
Serverless compute service. Upload code to a “Lamda Function” and set a trigger. When that that trigger happens the code is run
Amazon Elastic Container Service
Amazon Elastic Kubernetes Service.
Amazon Fargate
Container orchestration tools.
Containers carry code or applications with the dependencies and configuration settings to allow them to run.
ECS: Highly Scalable Container Manager. Comparable with Docker.
EKS: Fully managed service that allows you to run Kubernetes on AWS. Kubernetes is maintained by a volunteer community.
Fargate: A serverless compute engine for containers. Works with both ECS and EKS.
AWS Edge Locations
You can cache resources in edge locations using Amazon Cloudfront, to be able to deliver them to customers in regions distant from the region your servers are deployed in.
AWS Elastic Beanstalk
You can set code and desired resources in a template which can be automatically executed to deploy your environment.
AWS Cloudformation
Allows you to treat your infrastructure as code. Build a environment by writing lines of code which can be executed in any Amazon region to have a consistent way of automatically deploying resources.
Factors to consider when selecting a region.
Data governance and regulations.
Proximity to customers.
Resource availability.
Cost.
VPC
Virtual Private Cloud. Used to group resources in a virtual network with defined traffic and security rules.
You can provision your resources into subnets.
Internet Gateway
Virtual Private Gateway
AWS Direct Connect
An internet gateway allows public traffic into and out of your VPC.
A VPC component that allows private traffic into the network from defined sources over the public internet.
Direct Connect allows you to establish a private fiberoptic connection between your data center and your VPC.
Network Acess Control List
A list of rules for what can enter and leave a subnet. This is a stateless security check, which checks all traffic passing through.
Security Group
A set of rules about what which IP addresses can access which ports on a individual instance. A security group can be applied to multiple instances. This is a stateful security check, which will remember allowed traffic.
Amazon Elastic Block Storage
Create Virtual Hard Drive called EBS volumes. They are seperate from instances, and are persistent. You can take backups called snapshots from EBS volumes.
Instance Store Volume
Block storage attached to the host that an instance is running on. This will be deleted if you stop and start your instance, because it will be started on a different host with different built in storage. Only use in situation that you can lose the data.
Amazon S3
Simple Storage Server. You can store and retrieve unlimited amounts of data in buckets. You can version data, set access permissions, and place data into multiple tiers of accessibility. You can create lifecycle policies, which will autimatically move data between tiers.
Amazon S3 Storage Classes
S3 Standard: for frequently accessed data. Stores data in a minimum of three availability zones.
S3 Infrequent Access: Similar to S3 Standard but with a lower storage price and higher retrieval price.
S3 One Zone Infrequent Access: Similar to regular infrequent access but with a lower price and only stores data in a single availability zone.
S3 Intelligent tiering: Best for data with unknown of changing access patterns, but requires a small monthly monitoring fee.
S3 Glacier: Low cost storage for archiving. Slow retrieval with times from minutes to hrs.
S3 Glacier Deep Archive: Lowest cost storage. Can retrieve data in 12 hrs.