AWS, section 4, S3 and CORS

yes

unlimited, 0-5TB

• safe space to store data
• Data is stored across multiple locations
• object based storage

• Read after write after PUTs
• Its called eventual consistency. It might take time to delete stuff or edit the file though.

• They are bucket specific config
• eg bucket policies, access control lists

Lets you transfer lots of files in s3

data about the data you're storing, eg team name using it

• Key - name of object
• Value - the data
• Version ID - supports version control
• Metadata (data about the data you're sharing)
• Subresources - bucket policies, access control lists
• cross origin sharing. - allows buckets to access to other buckets

Standard S3. 99.99% availability, 99.9999999% durability. stored redundanly over multiple storage facilites capapble of 2 data centre losses.

S3 (IA) - infrequenctly accesssed. only pay when you retrieve

s3- One zone IA. Only stored in on availabiliity zone. Slighltly less aavailability. Cost is 20% less than regular IA.

reduced redundancy storage. no longer recommended. Use for data that can be recreated if lost

Glacier. Used for archiving and stuff you really dont use

It moves you into two different storage tiers depending on how often you acess stuff. Its more cost effective.

Private

An HTTP 200 status code

By bucket policies-applied at bucket level. Cant apply to individual object

and access control lists - applied at object level

Json

In transit - encrypting between the buckets

• At rest (server side encryption)
• SSE S3 - s3 managed keys
• SSE KMS - kms managed keys
• SSE C - Server side encryption with customer provided keys

• Amazon manage the keys for you.
• You get the strong multifactor authentication and they rotate the keys for you.

AMazon manage keys but additionally the keys are encrypted by an envelope which is encrypted. You get an audit trail too.

AWS manage encryption and decryption but you manage the keys and the lifecycle

You choose your own encryption methodology

• We make a put request and then we put a request header on it. Two types of request headers are possible. 
• SSE--S3 (s3 managed keys)
• SSE-KMS (KMS managed keys)
• If the header is not included, it wont get uploaded

CDN is for content delivery eg downloads and stuff and trasnfer acceleration is about faster uploads to s3

Its a CDN. System of distributed servers that deliver webpages and other web content based on where they are and where the page came from and where the content delivery server is.

It caches fequently requested stuff.

origin of files that CDN will distribute.

• Distribution is a name for CDN (cloud distribution network)
• The two types are:
• Web distribution - for websites
• RTMP distribution - for media streaming adobe real time messaging

• Places you cache frequently accessed shit to
• Its read and wirte. 
• Used by S3 transfer acceleration to reduce latency for s3 uploads.
• Object cached
• Remember its not the same place as availability zone

To provide users access to things only if they have paid for it.

Use cloudfront

its web application firewall to protect stuff on the application layer from sql injection attacks.

5GB

• Use the key name. 
• If you use sequential names, you might end up with multiple ojects aiming for the same slot

So S3 actually add random keys in there. like hex hash. Spreads the load a bit.

UPDATE: This is not longer necessary

200

no

It uses the bucket key name