Please enable JavaScript.

Coggle requires JavaScript to display documents.

AWS Cloud Services - Coggle Diagram

- - - - Root user
        
        First user created
        
        Connect using email address and password used to create AWS account
        
        Complete access to all the AWS services and ressources in the account
        
        Best practice => Don't use it for everyday tasks
      - Users are like identities you can create
      - Users represent the personn or application that interacts with AWS services and ressources
      - By default, it has no permissions associated
      - Best practice => 1 user per personn / service
    - - Allows or deny permissions to AWS services and ressources
      - Enable to customize users' level of access to ressources
      - Best practice => follow the rule of least privilege
      - Document (coded)
    - - Hardware security key
      - Hardware device
      - MFA Application
      - Can be enabled on users (best practice => on all users)
    - - Collection of users
      - When a policy is assigned to a group, all users in the group are granted permissions of the policy
    - - Identity that you can assume to gain temporary access to permissions
      - Only one a the time
      - Best practice => Use it for temporary situation instead of long-term and groups
  - - - Centrally control permissions for the accounts
      - Place restrictions on the AWS services, resources, and individual API actions that users and roles in each account can access
    - - Group of accounts
      - Make the things easier to manage
  - - - No cost
      - Protect against from the most common, frenquently occuring types of DDoS
      - Uses a variety of analysis techniques on the traffic going into your app to detect malicious traffic in real time and auto matigates it
    - - Paid service
      - Provides detailled attack diagnostics
      - Provide ability to detect and mitigate sophisticated DDoS attacks
      - Integrate with services such as Route 53, CloudFront, ELB
- - - - Directly to the apps on it
      - Using online interface on AWS website
    - - General Purpose
      - Compute Optimized
      - Memory Optimized
      - Accelerated computing
      - Storage optimized
    - - On-Demande
      - Saving Plans
      - Reserved Instances
        
        Convertible
        
        Scheduled
      - Spot Instances
      - Dedicated Hosts
    - - Dynamic scaling
      - Predictive scaling
  - - - :one: Upload the code to Lambda
      - :two: Set the code to trigger from an event source
      - :three: Lambda runs the code only when triggered
      - :four: Pay only for the comute time used
- - - - Check if a packet has the permissions to leave or enter the subnet
      - Based on
        
        Who sent the packet
        
        How it's trying to communicate
      - Access Control List
      - Doesn't check if the packet is allowed to reach an instance
      - Control inbound and outbound traffic at subnet level
    - - Instance level access
      - Automatically create when an EC2 instance is launched
      - By default
        
        Doesnt allow any traffic into the instance
        
        Allows all traffic out
      - Control inbounds and outbounds at EC2 level
- - - - Can be used with EC2 Instances in the same AZ
      - When EC2 Instance is stoped or terminated, the data in the EBS Volume is still available
      - Configuration
        
        Volume size
        
        Volume type
      - Import to back up the data on it using snapshots
      - Snapshots
        
        Incremental
      - Provide block-level storage volumes
      - Stores data in a single AZ
    - - Object storage system
      - Storage classes
        
        Standard
        
        Frequently accessed data
        
        Stores data in minimum 3 AZ
        
        Higher cost
        
        Standard Infrequent Access
        
        Ideal of infrequently accessed data
        
        Lower storage price and higher retrieval price
        
        Data that requires HA when needed
        
        Same availability level than Standard
        
        One Zone-Infrequent Access
        
        Data stored in a single AZ
        
        Want to save cost on storage
        
        You are easily able to reproduce the data if there is a AZ failure
        
        Intelligent-Tiering
        
        Data with unknown or changing access patterns
        
        Require a small monthly monitoring and automation fee per objetcs
        
        Glacier
        
        Low cost storage
        
        Designed for data archiving
        
        Retrieve objetcs within a few minutes to hours
        
        Glacier Deep Archive
        
        Lowest-cost objects storage
        
        Ideal for archiving
        
        Retrieve objetcs within 12 hours
      - Different classes depending one
        
        How often you want to retrieve your data
        
        How available you need your data to be
    - - Scalable automatically
      - Used with AWS Cloud and on-premise resources
      - Can scale on demand to Petabytes without disrupting applications
      - Managed file system
      - Automatic replications
      - Multiple instances accessing the data at the same time
      - Regional service
      - Stores data in and across multiple AZ
  - - - Managed service
      - Automates tasks such as hardware provisioning, db setup, patching and backups
      - Security options
        
        Encryption at rest (protecting data while it is stored)
        
        Encryption in transit (protecting data while it is being sent and received)
      - Database engines
        
        :star: Amazon Aurora
        
        Enterprise-class relational database
        
        Compatible with MySQL and PostgreSQL relational db
        
        Up to 5x faster than standard MySQL & 3x than PostgreSQL
        
        Reduce cost by reducing unnecessary I/O operations
        
        Ensure your db resources remain reliable and available
        
        Consider if workloads require HA
        
        Replicate 6 copies of your data across 3 AZ
        
        Continuously backs up data to S3
        
        PostgreSQL
        
        MySQL
        
        MariaDB
        
        Oracle Database
        
        Microsoft SQL Server
      - Enables you to run relational databases in AWS Cloud
    - - Serverless database
      - Manage underlying storage for you
      - Stores data redundatnly across AZ
      - Mirrors the data across mutliples drives
      - Very performant (1 ms response time)
      - Non-relational db
      - Not every item in the table has to have the same attributes
      - Add and remove attributes at any time
      - Cannot run complex query on it
      - Write queries based on a small subset of attributes designated as keys
      - Doesn't fit to every workloads
      - Key-Value system
    - - Helps customers migrate existing db into AWS
      - Secure and easy way
      - Source db remains fully operational during migration
      - Source and target don't have to be of the same type
      - Source and target db can be on-premise or RDS db
      - Homogenous => just click the button and done
      - Heterogeneous => 2 steps
        
        Use AWS Schema Conversion Tool to convert source schemas and code it to match target db
        
        Use DMS to migrate data from source to target db
      - Other use cases
        
        Db consolidation
        
        Continuous db replication
        
        Developement and test of db migrations
    - - Schema Conversion Tool => Convert source db schema to make it match target schema in heterogeneous migrations
      - DocumentDB => Document db service that supports MongoDB workloads
      - Neptune => Graph db service used to build and run applications that with highly connected datasets
      - QLDB : Quantum Ledger DataBase => Ledger db service used to review a complete history of all changes made to app data
      - Managed Blockchain => Create and manage blockchain networks with open-source frameworks
      - ElastiCache => Add caching layer on top of the db to improve read times on common requests, supports only redis and memcached data stores types
      - DAX : DynamoDB Accelerator => in-memory cache for DynamoDB that improve response time from ms to µs
- - - - Optional
      - Automatically detect unusual API activities
  - - - Cost optimization
      - Permormances
      - Security
      - Fault tolerance
      - Service limits
    - - Orange triangles => items with recommended investigations
      - Red circles => items with recommended actions
      - Green checks => items with no problems
- - - - Business
        
        Ensure that IT aligns with business needs
        
        Ensure that IT investments link to bey business results
        
        Create a strong business case for cloud adoption and prioritize cloud adoption initiatives
      - People
        
        Evaluate organizational structures and roles
        
        Process requirements and identify gaps
        
        Help to prioritize training, staffing and organizational changes
      - Governance
        
        Focuses on the skills and processes to align IT strategy with business strategy
        
        Ensure that you maximize the business value and minimize tasks
        
        Understand how to update the staff skills
      - Platform
        
        Principles and patterns to implement new solutions on the cloud
        
        Migrate on-premise to the cloud
      - Security
        
        Ensures that the organization meets security objectives for visibility, auditability, control, and agility
        
        Structure the selection and implementation of security controls that meet the organization’s needs
      - Operations
        
        Enable, run, use, operate, and recover IT workloads to the level agreed upon with your business stakeholders
  - - - 2 CPUs
      - 4GB RAM
      - 8To usable storage
    - - Edge Storage Optimized
        
        Large scall data migrations
        
        80 To of HDD, S3 compatible
        
        1To of SSD
        
        40 vCPUs
        
        80 GiB of memory
      - Edge Compute Optimized
        
        Powerfull computing ressources
        
        42 Tb usable HDD, S3 et EBS compatible
        
        7,68TB usable of NVMe, EBS compatible
        
        52 vCPUs
        
        208GiB of memory
        
        Optional NVIDIA Tesla V100 GPU
    - - Exabyte-scale data transfer
      - Move large amount of data to AWS
      - Up to 100 PB