Certified Cloud Practitioner Flashcards
S3 access logs
see all requests made to bucket
Two types of S3 object replications
Cross-region and Same-region
S3 - Cross-region replication use case (3)
compliance, lower latency access, replication across accounts
S3 - Same-region replication use case (2)
log aggregation, live replication between production and test accounts
S3 - Name 6 storage classes
- Standard General Purpose
- Standard Infrequent Access (IA)
- One Zone Infrequent Access
- Intelligent Tiering
- Glacier
- Glacier Deep Archive
S3 - Standard General Purpose class (availability, use case, failures)
99.99% availability
Frequently accessed data
Sustain 2 concurrent facility failures
S3 - Standard Infrequent Access class (availability, use case, fees)
99.9% availability
Less frequently accessed, but rapid when needed
Cheaper at rest, but retrieval fee
S3 - One Zone Infrequent Access class (availability, replication, cost, use case)
99.5% availability
Not replicated across AZ
Lower cost than IA
Storing backups or easily recreated data
S3 - Intelligent Tiering class (availability, latency, what does it do)
99.9% availability
Same low latency as standard access
Auto moves objects between frequent/infrequent access
S3 - Glacier class (cost, retrieval times and fees)
Low cost
Different retrieval times and fees
S3 - Glacier Deep Archive class (retrieval for one file, retrieval for bulk files, relative cost)
12 hours retrieval for one file
48 hours for bulk files
Cheapest option
S3 - Durability and Availability
Durability = 99.999999999% (11 9's) Availability = 99.99% for standard
S3 - Object Lock & Vault Lock (what it does, use case)
File written once and never deleted
Useful for compliance and data retention
What is Snow used for? (2)
Data migration and Edge computing
Snow - Which services are for Data Migration? (3)
Snowcone
Snowball Edge
SnowMobile
Snow - Which services are for Edge Computing? (2)
Snowcone
Snowball Edge
Snow - Snowball Edge (use case, two classes w/ capacity)
Physical transport of data, large data cloud migration
Storage Optimized: 80TB
Compute Optimized: 42TB
Snow - Snowcone (capacity, size, what service to send data to AWS?)
8TB
Rugged, small
AWS DataSync
Snow - SnowMobile (capacity for one, capacity for multiple, best for this capacity and more)
100PB of data per
Exabytes of data for multiple
Better for more than 10PB
Snow - Edge Computing use cases (2)
Limited/no internet or computing power
Useful to preprocess data, machine learning, transcoding media streams
Snow - Edge Computing can run which two services?
EC2 instances
Lambda functions
Snow - Software used to connect to Snow devices
OpsHub
Storage Gateway - use case, services (3)
Hybrid cloud
Bridge on-prem with cloud, EBS, S3, Glacier
RDS - capabilities (3)
Automatic provisioning and OS patching
Backups and restore
Monitoring
RDS - additional features (3)
Read replicas
Multi AZ - failover
Multi region - for read replicas
RDS - read replicas - what it is, how many can you have, how data moves
Copy of RDS
Up to 5
Data only written to one, copied to replicas
RDS - one drawback
No SSH access
Aurora - two DB types
PostreSQL
MySQL
Aurora - two advantages over RDS
Cloud optimized - faster
Storage automatically grows up to 64TB in 10TB increments
Aurora - two disadvantages over RDS
Costs more
Not in free tier
Elasticache - what is it, two techs, why
In memory database
Managed Redis or Memcached
Help reduce load of databases for READ intensive workloads
DynamoDB - what is it
NoSQL database, key/value
DynamoDB - three advantages
Replication across 3 AZ
Low latency
Serverless
DynamoDB - what can be enabled for performance improvement, how does it work?
DynamoDB Accelerator (DAX) Fully managed in-memory cache for performance improvement
Redshift - use cases (3)
OLAP (online analytical processing)
Analytics
Data warehousing
Redshift - difference between other databases
Load data once every hour instead of every second or on demand
Redshift - cost model
Pay as you go based on instances provisioned
Redshift - how to query data
SQL
Redshift - what does it integrate with? (2)
BI tools like AWS Quicksight or Tableau
Elastic MapReduce (EMR) - what is it used for?
Hadoop clusters (big data)
Elastic MapReduce (EMR) - how can it be provisioned? (2)
Auto-scaling and spot instances
Athena - what is it
Full serverless database with SQL capabilities to query data in S3
Athena - cost model
Pay per query
Athena - where do query results go?
Back to S3
Quicksight - what is it
Serverless machine learning-powerd BI tool to create interactive dashboards
Quicksight - cost model
Per session pricing
Quicksight - integrations (5)
RDS Aurora Athena Redshift S3
DocumentDB - what is it
NoSQL database
Aurora version of MongoDB
Neptune - what is it
Graph database
Neptune - replication (2)
Across 3 AZ and 15 read replicas
Neptune - use cases (3)
Fraud detection
Recommendation engines
Social networking
Quantum Ledger Database (QLDB) - what is it (2)
Ledger of recorded changes to data
Review history of all changes made to application data
Quantum Ledger Database (QLDB) - advantages (2)
Immutable system, cryptographic hash
More performant than common ledger blockchains
Quantum Ledger Database (QLDB) - how is data queried?
SQL
Quantum Ledger Database (QLDB) - is it decentralized?
No, there is no decentralization
Managed Blockchain - what is it used for
Join public blockchain network or create your own private network
Managed Blockchain - compatible with which two blockchains?
Hyperledger
Ethereum
Managed Blockchain - is it decentralized?
Yes
Database Migration Service (DMS) - what is it
Quickly and securly migrate database to AWS
Database Migration Service (DMS) - two types of migrations
Homogenous (postgres –> postgres)
Heterogenous (mysql –> postgres)
Glue - what is it for (2)
Managed ETL
To prepare and transform data for analytics, possibly from multiple sources
Glue - how is it provisioned?
Serverless
Glue - example usage (extract, transform, load)
Extract data from S3 and RDS, transform, load into Redshift
EBS - definition
Network drive attached to EC2 instance
EBS - two constraints
Only attached to one instance at a time
Bound to a specific AZ
EBS - three features
Can be detached/attached quickly
Can increase capacity over time
Can attach multiple to one instance
EBS - delete on termination is on by default when? (2)
By default for root volume
Off for non-root volumes
EBS snapshots - can copy where (2)
Across AZ or Region
AMI - is specific to what?
A region
AMI - three types
Public AMI from AWS
Your own
AWS Marketplace AMI
AMI - creating one also does what?
Creates an EBS snapshot
EC2 Image Builder - does what
Automates the creation and testing of AMIs
EC2 Image Builder - two features
Can distribute resulting AMI across regions
Can be run on a schedule
EC2 Image Builder - cost model
Free service, only pay for underlying resources
Instance Store - what is it
Storage physically connected to server
Instance Store - advantage
Better performance
Instance Store - disadvantage
Storage is ephemeral and is lost when instance is stopped
Instance Store - what is it good for
Buffer, cache, or temporary content
EFS - advantage
Can be mounted to 100s of EC2 instances
EFS - constraint
Only works with Linux EC2
EFS - availability
Multi-AZ
EFS - Infrequent Access - cost and what it does
Lower cost
Move files to IA based on a lifecycle policy such as last time they were accessed
FSx - two types
Windows File Server
Lustre
FSx - one advantage
Can be accessed from AWS or on-prem infrastructure
FSx - Windows can integrate with what
Microsoft AD
FSx - Lustre is good for what
High performance computing