Easy ReviewCLF-C02
Cloud Concepts22

AWS global infrastructure

Regions, Availability Zones and edge locations, and how to use them for availability, latency and data residency.

AWS runs its services from Regions around the world. Each Region holds several isolated Availability Zones, and edge locations bring content closer to users.

Regions, Availability Zones, and edge locations

The AWS global infrastructure has three building blocks. A Region is a geographic area. Inside it, Availability Zones (AZs) are isolated groups of data centers. Edge locations, outside the Regions, deliver content close to users. Regions · Availability Zones

How it works

  1. You pick a Region for each resource. Regions are isolated from each other, and data stays in its Region unless you copy it.
  2. Inside the Region, you spread resources over two or more AZs. Each AZ has its own power, networking and connectivity, and AZs share no generators or cooling.
  3. AZs are close enough for synchronous replication, so a database or load-balanced fleet can span them.
  4. Edge locations cache content for Amazon CloudFront near users; on a miss they fetch it from the origin in the Region.
Building block What it is Used for
Region Geographic area with at least three AZs Where resources and data live; latency, data residency
Availability Zone One or more data centers with independent power and networking High availability inside a Region
Edge location Cache site near users Fast content delivery with CloudFront

Exam cues

  • If the question says survive a data center failure, think multiple Availability Zones.
  • If the question says users far away see slow static content, think edge locations with CloudFront.
  • If the question says data must stay in a country, think choose a Region there; data stays in its Region unless copied.

Remember

  • Every public Region has at least three AZs.
  • AZs are up to about 100 km (60 miles) apart, with single-digit millisecond latency between them.
  • AZ codes are the Region code plus a letter, such as us-east-2a.
  • Global: IAM, Organizations, CloudFront, Route 53, Global Accelerator. Most others are Regional.

Trap: One AZ is not highly available. A single AZ can fail as a unit, so high availability needs at least two AZs.

A Region, its Availability Zones, and edge locations
AWS Region12Availability Zone Adata centersAvailability Zone Bdata centersAvailability Zone Cdata centersEdge locationCloudFront cacheUsers
  1. Users reach the nearest edge location, which answers from its cache when it can.

  2. On a cache miss, the edge location fetches the content from the origin in the Region.

From the AWS docs (18)
  • A Region is a geographical location where AWS clusters data centers; each Region is made of multiple independent, physically separate Availability Zones.
  • Each public AWS Region has at least three Availability Zones.
  • Regions are isolated and independent from each other, which limits a service failure to a single Region.
  • Resources and data created in one Region don't exist in any other Region unless you replicate or copy them.
  • Choose a Region that has the services and features you need, that is close to most of your users for lower latency, and that meets your legal requirements.
  • An Availability Zone is one or more discrete data centers with independent and redundant power, networking and connectivity in a Region.
  • Availability Zones in a Region are up to about 100 km (60 miles) apart to prevent correlated failures, but close enough for synchronous replication with single-digit millisecond latency.
  • Common points of failure such as generators and cooling equipment aren't shared across Availability Zones, which are supplied by different power substations.
  • An application that uses multiple Availability Zones is better protected from issues that affect one Availability Zone and, if architected correctly, stays up when one fails.
  • An Availability Zone code is the Region code followed by a letter, such as us-east-2a.
  • DNS routes a CloudFront request to the edge location (point of presence) that can best serve it, typically the nearest in latency, which returns cached content or fetches it from the origin.
  • CloudFront regional edge caches keep content closer to viewers even when it isn't popular enough to stay at an edge location.
  • A few AWS services are global within a partition, such as IAM, AWS Organizations, Amazon CloudFront, AWS Global Accelerator and Amazon Route 53; most services, such as EC2, RDS and Lambda, are Regional.
  • By default, S3 general purpose bucket names must be unique across all AWS accounts in a partition, but each bucket is created in one Region.
  • Most AWS services, such as Amazon EC2, Amazon RDS, AWS Lambda and Amazon Rekognition, have Regional endpoints, so their resources live in the Region you choose.
  • AWS maps physical Availability Zones to AZ names randomly per account, so us-east-1a can differ between accounts; the AZ ID (such as use1-az1) identifies the same physical zone in every account.
  • Zonal services, such as Amazon EC2 and Amazon EBS, create resources in an Availability Zone you choose, and each zone fails independently.
  • Regional services, such as Amazon S3, are built across several Availability Zones of one Region and spread data and requests over them.

When to use multiple Regions

A single Region already spans several Availability Zones, which handles the loss of a data center. A second Region protects against problems that hit a whole Region. It also brings the workload closer to users elsewhere, and keeps data in the countries that require it.

How it works

  1. Decide what you are protecting against. For a data center loss, multiple AZs and backup and restore may be enough. For a Region loss, or when regulations require it, use a multi-Region strategy.
  2. Replicate data to the second Region. Nothing crosses Regions unless you replicate or copy it.
  3. Prepare the infrastructure there too, preferably with infrastructure as code, so it can be deployed quickly.
  4. For users in other parts of the world, serve them from a Region close to them.
Reason Why a second Region helps
Disaster recovery A failure in one Region doesn’t affect the others
Business continuity The business keeps running while a Region is disrupted
Low latency for end users A Region near distant users shortens network paths
Data sovereignty Data created in a Region stays there unless copied

Exam cues

  • If the question says survive the loss of an entire Region, think multiple Regions, not more AZs.
  • If the question says data must remain in a country or area, think choose the Region there.
  • If the question says users on another continent see high latency, think a Region near them, or edge locations for cached content.

Remember

  • Multiple AZs protect against data center failures; multiple Regions protect against Region failures.
  • In active/passive, the passive Region serves no traffic until failover.
  • Regions connect over the private AWS global network.

Trap: More AZs don’t help when a whole Region is down. Only a second Region covers that case.

From the AWS docs (8)
  • Separation between Regions limits a service failure to a single Region, leaving other Regions' operations unaffected.
  • For the loss of one data center, a well-architected, highly available workload may need only backup and restore; for the loss of a Region or regulatory requirements, consider pilot light, warm standby or multi-site active/active.
  • Active/passive disaster recovery hosts the workload in one active Region and uses a passive Region for recovery; the passive Region serves no traffic until failover.
  • Backup and restore can protect against a regional disaster by replicating data to other Regions, but the infrastructure, configuration and code must also be redeployed there.
  • Selecting a Region close to most of your users lowers network latency.
  • An AWS account offers multiple Regions so resources can be created where requirements demand, for example in Europe to be close to European customers or to meet legal requirements.
  • Resources and data stay in the Region where they are created unless they are explicitly replicated or copied.
  • Regions are interconnected through the AWS global network, which provides private, high-bandwidth, low-latency connectivity.

Local Zones, encryption, and regional pricing

AWS Local Zones bring selected resources close to large population and industry centers for low-latency access. When choosing where to run resources, also consider Region pricing and encryption between Availability Zones.

How it works

  1. Use a Local Zone to place compute, storage, database and other selected resources closer to users who need low latency.
  2. Consider Region pricing alongside latency, data residency and service availability. Local costs, including land, electricity and taxes, can make resource prices differ by Region.
  3. All traffic between Availability Zones is encrypted.

Exam cues

  • If the question says a game studio needs very low latency for players in a large city without a Region, think Local Zones, because they place selected resources close to population centers.
  • If the question says the same workload could run in different Regions, think about price as well as latency, data residency and service availability, because resource prices can differ.
  • If the question asks about traffic between Availability Zones, remember that it is encrypted.

Remember

  • Local Zones offer selected compute, storage and database resources near population and industry centers.
  • Region prices can differ with local costs.
  • All traffic between Availability Zones is encrypted.

Trap: Do not assume the same resource costs the same in every Region. Compare prices along with latency, data residency and service availability.

From the AWS docs (3)
  • AWS Local Zones place compute, storage, database and other select resources close to large population and industry centers for low-latency access.
  • All traffic between Availability Zones is encrypted.
  • Resource pricing can differ between Regions because of local costs such as land, electricity and taxes, so Region cost is a selection factor alongside latency, data residency and service availability.

Flashcards

tap to flip
Answer

·

All 22 cards as a list

How do AWS Regions and Availability Zones relate to each other?

A Region is a geographic area that contains multiple isolated Availability Zones. Each public Region has at least three.

What is an AWS Availability Zone?

One or more discrete data centers with independent, redundant power, networking and connectivity inside a Region.

A company runs its web servers in one Availability Zone and wants the application to survive the loss of that zone. What should it do?

Run the servers in at least two Availability Zones of the Region. AZs share no single points of failure, so one can fail while the others keep serving.

Why can a database replicate synchronously between Availability Zones in the same AWS Region?

AZs are close enough (up to about 100 km apart) for single-digit millisecond latency, while far enough apart to avoid correlated failures.

A company stores data in an Amazon S3 bucket in one AWS Region. Does AWS copy that data to other Regions automatically?

No. Data stays in its Region unless you replicate or copy it.

Which three factors should a company weigh when choosing an AWS Region?

  • The services and features it needs are available there
  • Closeness to most users, for lower latency
  • Legal and data residency requirements

What role do edge locations play in the AWS global infrastructure?

They serve Amazon CloudFront content from the location nearest the user, returning cached content or fetching it from the origin.

Does an outage in one AWS Region normally affect workloads running in other Regions?

No. Regions are isolated and independent, which limits a service failure to one Region.

Are Availability Zones the same thing as edge locations?

No. AZs are data centers inside a Region where you run resources. Edge locations are separate sites that cache and deliver content, such as CloudFront content, close to users.

Which of these AWS services are global rather than Regional: IAM, Amazon CloudFront, Amazon EC2, Amazon RDS, Amazon Route 53?

IAM, CloudFront and Route 53. EC2 and RDS are Regional.

Is Amazon S3 a global service because bucket names are unique worldwide?

No. Bucket names share a global namespace by default, but each bucket and its data live in one Region.

How are AWS Regions and Availability Zones related?

Each AWS Region is a geographical location made up of multiple independent, physically separate Availability Zones.

Why can using multiple Availability Zones improve an application’s availability?

An application designed across multiple Availability Zones is better protected from a failure affecting one zone.

Sort these by scope: IAM, CloudFront, Route 53, S3, EFS, AWS Batch, EC2, EBS.

Global: IAM, CloudFront, Route 53. Regional: S3, EFS, AWS Batch. Zonal: EC2 instances and EBS volumes, which live in one Availability Zone.

Name four reasons to run a workload in more than one AWS Region.

  • Disaster recovery
  • Business continuity if a whole Region is disrupted
  • Lower latency for users in other parts of the world
  • Data residency or legal rules that require data in a specific country

A highly available workload already runs across three Availability Zones. It must also keep running if an entire AWS Region becomes unavailable. What does it need?

A second Region, with a disaster recovery strategy such as pilot light, warm standby or multi-site active/active. Multiple AZs protect only against data center failures.

A law requires that customer data of German users stays in Europe. How does AWS Region choice help?

Create the resources in a European Region. Data stays in its Region unless you replicate or copy it elsewhere.

In an active/passive multi-Region setup, does the passive Region serve user traffic?

No. It stays ready for recovery and serves traffic only after a failover.

Is copying backups to another AWS Region enough to recover a whole workload there?

No. The infrastructure, configuration and application code must also be redeployed in the recovery Region, ideally with infrastructure as code.

A game studio needs single-digit millisecond latency for players in a large city that has no AWS Region. Which AWS infrastructure option fits?

AWS Local Zones. They place compute, storage and database resources close to large population centers.

Is traffic between Availability Zones in a Region encrypted?

Yes. All traffic between AZs is encrypted.

Besides latency, legal requirements and service availability, what else can differ between two AWS Regions that host the same workload?

Price. Each Region has its own resource prices, so with no residency or latency constraint the cheaper Region can lower cost.

You finished every topic of CLF-C02!

Your badge turned gold. Next step: book the exam on AWS. Once you pass, add yourself to Who passed.

Share

Share that you finished this certification

I finished every topic of AWS Certified Cloud Practitioner (CLF-C02) on Easy Review. Come study too!