AWS Storage Type
- S3 (Simple Storage Services) [Object Level Based Storage]
- EFS (Elastic File System)
- EBS (Elastic Block Storage)
- Glacier
- Snowball
Amazon S3(Simple Storage Services)
- Amazon S3 is public storage resource in AWS (Amazon Web Services) Cloud which is object based storage.
- AWS S3 Bucket is storage same like a file folders which store objects with data and its Metadata.
- S3 stores all the files (like pdf, images, video, documents etc ) as a object in AWS Cloud. You cannot have the Bucket inside bucket, Bucket can have the folders which groups object inside in it. S3 bucket is having flat architecture (Bucket inside bucket you can't create.).
- If you want to use AWS S3 services, First you have to create the S3 Bucket in any of the AWS Region. Bucket name is unique Globally. Once Bucket is created it get shared with all the AWS Account. The bucket name cannot be used by any other AWS account.
- Amazon S3 services comes under IaaS (Infrastructure as a Services). The Important AWS Services which is offered as Iaas is S3 (Simple Storage Service), EC2 (Elastic computing 2) and RDS (Relational Database Service). These IaaS services is charged per hour basis.
- We can't Install operating System in S3 Storage.
- AWS S3 have distributed Data Architecture where copy of data is stored in multiple location within the region.
- Bucket can store maximum 5TB Data, 100 bucket per aws Account which you can extend further.
- S3 Bucket cannot be transferred from one aws account to another AWS account. It can be shared and can be access with multiple location.
S3 Bucket and Naming Rules
- Bucket name is Globally Unique. In AWS Cloud two bucket can't have the same name. It must be unique.
- Once the Bucket name is created it cannot be changed. If bucket name is deleted then again it will be available for other user to choose that Bucket name.
- Bucket name must contain atleaste 3 Character and max 63 Character.
- Bucket Name is URL part through which we try to access the bucket.
- Bucket name can be created using lower Case letter, number and hyphen. We cant use Upper case letter while naming the bucket. You cant start and end the bucket name with -(hyphen). Bucket name should not be an IP address.
- Whenever we create bucket or Object, By default it is private. Only bucket owner can access that.
S3 Bucket Sub-resources
- Object Life cycle : In this we can define the object life cycle. We can move the object based on the time frame from one storage type to other storage. this can reduce the cost. See the below image where we are moving the object from one storage to other based on the recent data.
- Website : We can use S3 for hosting Static website.
- Versioning : You can enable the versioning on your storage which will maintain your previous updated file and the current file too. You cant disable the versioning. You can enable or suspend the versioning.
- Access Control List : In this you can decide the security of your bucket regarding access.
S3 Bucket Versioning
Bucket Versioning help us to recover the data in the case of accidental delete. Suppose you deleted the Data accidentally, to retrieve the data we enable versioning. Versioning maintain the older data as well as the current data. If the versioning is enabled and the object / Data got deleted, in that case you can recover / restore the data easily.
Versioning Help us in data retention as well as data archive. Once you enable the Versioning, you can't disable it. You can suspend it only.
When Versioning is enabled and we delete the object / data , delete marker is placed on the object which we can recover whenever we want.
You will be charged for all the multiple versions. Example Just imagine you have 10 MB file initially, next versioning your file size became 14 MB, again in next versioning it became 18 MB. It will be charged for incremental only which will be 8 MB not for 42 MB. This is incremental versioning.
You can apply the life-cycle on versioning which means as the object / data become older we can move it to cheaper storage or we can delete the data.
The Object which exist before version enabling will have the version ID Null.
Multi Factor Authentication (MFA)
MFA is extra level of security which we can apply on our S3 Bucket. By Enabling the Multi-factor authentication you restrict the other users to delete the S3 object accidentally. Once the Multi-factor Authentication is enabled you can delete until unless the security code received by the registered physical device is shared with you within 30 second.
S3 Multipart Upload
S3 Multipart Upload allow us to upload our file to server in parts. This object parts get uploaded independently and in parallel. You must use the multipart Upload facility to upload the file in the server if the size is equal or greater than 5 gb using Multipart Upload API.
Type of AWS S3 Storage
1) Amazon S3 Standard [Storage Cost is High but Accessing / Retrieving cost is low. ]
This Storage is useful if you need data access frequently.
2) Amazon S3 Glacier Deep Archive [Storage Cost is cheapest but Retrieving cost is High.]
Amazon S3 Glacier Deep Archive is suitable when you have rare access to the data(Once in a year or rare access.). As S3 Glacier Deep Archive storage is damm Cheaper but accessing the data cost higher. Accessing Glacier Deep Archive Data takes time which is not fix.
3) Amazon Glacier
Amazon S3 Glacier is little costlier then Glacier Deep Archive. This also we use when we don't need the data access frequently. Accessing Glacier Data takes time which is not fix. The Main difference between Amazon glacier and Glacier Deep Archive is, in AWS Glacier you have the option to select the data retrieval time. This means you can select the access time if you want to access the data. Three Option Available for data retrieval, you want retrieval within a minute, within a Hour, or within 12 hour.
4) Amazon S3 Standard Infrequent Access
Amazon S3 Standard Infrequent Access This storage is best for infrequent access. Sometime you need to access the data and you need it immediately. This storage provide immediate access it don't delay like Amazon Glacier or Amazon S3 Glacier Deep Archive. Overall it is infrequent access of data but whenever it is access data must available immediately without delay.
5) Amazon S3 One Zone - IA
By Default S3 Create three copy of your data and keep it as a back up and charge you accordingly. And in any disaster it will recover your data using other copy. S3 One Zone allow you to keep only one copy of the data in one zone. In disaster your data is not recoverable.
6) Amazon S3 Intelligent tearing.
We use this Intelligent Tearing storage in case of when we dont know whether we are going to access the data immediately or not.
