S3 Access Points

AWS S3 Security

~ Last updated on : ~ jayendrapatil

AWS S3 Security

📌 Last Updated: June 2026. This post reflects the latest S3 security features including default SSE-S3 encryption, SSE-C disabled by default (April 2026), S3 Access Grants, ABAC for S3 buckets, Resource Control Policies (RCPs), and DSSE-KMS dual-layer encryption.

  • AWS S3 Security is a shared responsibility between AWS and the Customer
  • S3 is a fully managed service that is protected by the AWS global network security procedures
  • AWS handles basic security tasks like guest operating system (OS) and database patching, firewall configuration, and disaster recovery.
  • Security and compliance of S3 are assessed by third-party auditors as part of multiple AWS compliance programs including SOC, PCI DSS, HIPAA, etc.
  • S3 provides several other features to handle security, which are the customers’ responsibility.
  • S3 Encryption supports both data at rest and data in transit encryption.
    • Data in transit encryption can be provided by enabling communication via SSL/TLS or using client-side encryption
    • Data at rest encryption can be provided using Server Side or Client Side encryption
    • All new objects are automatically encrypted with SSE-S3 by default (since January 2023)
    • SSE-C is disabled by default on all new general purpose buckets (since April 2026)
  • S3 permissions can be handled using
    • IAM User Policies
    • Resource-based policies which include Bucket policies (ACLs are now disabled by default)
    • S3 Access Points
    • S3 Access Grants – scalable access control mapping corporate identities to S3 data
    • Attribute-Based Access Control (ABAC) – tag-based authorization for S3 buckets (Nov 2025)
  • S3 Object Lock helps to store objects using a WORM model and can help prevent objects from being deleted or overwritten for a fixed amount of time or indefinitely.
  • S3 Access Points simplify data access for any AWS service or customer application that stores data in S3.
  • S3 Versioning with MFA Delete can be enabled on a bucket to ensure that data in the bucket cannot be accidentally overwritten or deleted.
  • S3 Block Public Access provides controls across an entire AWS Account or at the individual S3 bucket level to ensure that objects never have public access, now and in the future.
  • S3 Access Analyzer monitors the access policies, ensuring that the policies provide only the intended access to your S3 resources.
  • S3 Object Ownership – by default set to Bucket owner enforced, disabling ACLs on all new buckets.
  • Resource Control Policies (RCPs) – centrally restrict external access to S3 resources across an organization (Nov 2024).

S3 Encryption

  • S3 allows the protection of data in transit by enabling communication via SSL/TLS or using client-side encryption
  • All new objects uploaded to S3 are automatically encrypted at rest with SSE-S3 by default (since January 5, 2023). This applies to all buckets at no additional cost.
  • SSE-C (Server-Side Encryption with Customer-Provided Keys) is disabled by default on all new general purpose buckets since April 2026. Applications that need SSE-C must explicitly enable it using the PutBucketEncryption API after bucket creation.
  • S3 provides data-at-rest encryption using
    • Server-Side Encryption: S3 handles the encryption
      • SSE-S3
        • S3 handles the encryption and decryption using S3 managed keys
        • This is now the base level of encryption for every bucket in S3 (default)
      • SSE-KMS
        • S3 handles the encryption and decryption using keys managed through AWS KMS.
        • Provides additional audit trail via AWS CloudTrail for key usage
      • DSSE-KMS (Dual-Layer Server-Side Encryption)
        • Applies two layers of encryption using AWS KMS keys
        • Even if one encryption layer were compromised, data remains protected by the second layer
        • Designed for highly regulated workloads requiring CNSSP 15 compliance
        • Higher cost due to increased processing overhead and additional KMS API calls
      • SSE-C
        • S3 handles the encryption and decryption using keys managed and provided by the Customer.
        • Disabled by default on new buckets since April 2026 — must be explicitly enabled via PutBucketEncryption API
        • Most modern use cases prefer SSE-S3 or SSE-KMS for greater flexibility
    • Client Side Encryption: Customer handles the encryption
      • CSE-KMS
        • Customer handles the encryption and decryption using keys managed through AWS KMS.
      • Client-side Master Key
        • Customer handles the encryption and decryption using keys managed by them.

S3 Permissions

Refer blog post @ S3 Permissions

S3 Object Ownership

  • S3 Object Ownership is a bucket-level setting to control ownership of objects uploaded to a bucket and to disable or enable ACLs.
  • By default, Object Ownership is set to Bucket owner enforced, and all ACLs are disabled.
  • When ACLs are disabled, the bucket owner owns all the objects in the bucket and manages access exclusively using access-management policies.
  • A majority of modern use cases in S3 no longer require the use of ACLs.
  • Object Ownership has three settings:
    • Bucket owner enforced (default) – ACLs are disabled. The bucket owner automatically owns and has full control over every object. Policies define access control.
    • Bucket owner preferred – The bucket owner owns new objects written by other accounts if uploaded with the bucket-owner-full-control canned ACL.
    • Object writer – The uploading account owns the object. ACLs are enabled.
  • With ACLs disabled, you can use policies to control access to all objects in your bucket, regardless of who uploaded them.

S3 Object Lock

  • S3 Object Lock helps to store objects using a write-once-read-many (WORM) model.
  • can help prevent objects from being deleted or overwritten for a fixed amount of time or indefinitely.
  • can help meet regulatory requirements that require WORM storage or add an extra layer of protection against object changes and deletion.
  • can be enabled only for new buckets and works only in versioned buckets.
  • provides two retention modes that apply different levels of protection to the objects
    • Governance mode
      • Users can’t overwrite or delete an object version or alter its lock settings unless they have special permissions.
      • Objects can be protected from being deleted by most users, but some users can be granted permission to alter the retention settings or delete the object if necessary.
      • Can be used to test retention-period settings before creating a compliance-mode retention period.
    • Compliance mode
      • A protected object version can’t be overwritten or deleted by any user, including the root user in the AWS account.
      • Object retention mode can’t be changed, and its retention period can’t be shortened.
      • Object versions can’t be overwritten or deleted for the duration of the retention period.
  • Legal Hold
    • Provides the same protection as a retention period, but has no expiration date.
    • Remains in place until explicitly removed.
    • Can be applied to any object in an Object Lock-enabled bucket, whether or not that object has a retention period.
    • Requires the s3:PutObjectLegalHold permission to add or remove.

S3 Access Points

  • S3 access points simplify data access for any AWS service or customer application that stores data in S3.
  • Access points are named network endpoints that are attached to buckets and can be used to perform S3 object operations, such as GetObject and PutObject.
  • Each access point has distinct permissions and network controls that S3 applies for any request that is made through that access point.
  • Each access point enforces a customized access point policy that works in conjunction with the bucket policy, attached to the underlying bucket.
  • An access point can be configured to accept requests only from a VPC to restrict S3 data access to a private network.
  • Custom block public access settings can be configured for each access point.
  • S3 Access Points now support tags for ABAC (Aug 2025), enabling tag-based permissions for access points.
  • S3 Multi-Region Access Points provide a global endpoint for applications to fulfill requests from S3 buckets in multiple AWS Regions.
    • Uses AWS Global Accelerator for intelligent routing to the nearest bucket
    • Supports failover controls for active-passive or active-active configurations
    • Does not support anonymous requests

S3 Access Grants

  • S3 Access Grants is a scalable access control solution that maps identities to S3 data permissions.
  • Acts as a credential vendor — users request credentials from S3 Access Grants, which evaluates grants and vends temporary, least-privilege access credentials.
  • Supports mapping identities from corporate directories such as Microsoft Entra ID (formerly Azure AD), Okta, and other IdPs via AWS IAM Identity Center.
  • Supports up to 100,000 grants per Region per account.
  • Ideal for complex access control requirements:
    • Bucket policy size limit (20 KB) exceeded
    • Human identities from external IdPs need S3 access for analytics and big data
    • Cross-account access without frequent IAM policy updates
    • Object-level access control for unstructured data
  • Integrates with AWS machine learning and analytics services including Amazon SageMaker, Amazon Redshift, and AWS Glue.
  • Simplifies authentication when using both IAM and Identity Provider permissions (March 2025).

S3 Attribute-Based Access Control (ABAC)

  • Launched November 2025 – Tag-based authorization for S3 general purpose buckets.
  • Enables automatically managing permissions for users and roles by controlling data access through tags on S3 buckets.
  • Instead of managing permissions individually, use tag-based IAM or bucket policies to grant or deny access based on matching tags between principals and buckets.
  • Makes it easy to grant S3 access based on project, team, cost center, data classification, or other bucket attributes instead of bucket names.
  • ABAC must be explicitly enabled on each S3 bucket using the PutBucketAbac API.
  • Tag-based condition keys work in AWS Organizations policies, IAM policies, and S3 bucket policies.
  • Can enforce tagging requirements at bucket creation using SCPs or IAM policies with aws:TagKeys and aws:RequestTag condition keys.
  • Same tags can serve dual purpose: access control (ABAC) and cost allocation.
  • Available through Console, API, SDKs, CLI, and CloudFormation at no additional cost.
  • Also supported on S3 directory buckets, S3 access points, and S3 tables buckets.

S3 VPC Gateway Endpoint

  • A VPC endpoint enables connections between a VPC and supported services, without requiring that you use an internet gateway, NAT device, VPN connection, or AWS Direct Connect connection.
  • VPC is not exposed to the public internet.
  • A Gateway Endpoint is a gateway that is a target for a route in your route table used for traffic destined to either S3.

S3 Block Public Access

  • S3 Block Public Access provides controls across an entire AWS Account or at the individual S3 bucket level to ensure that objects never have public access, now and in the future.
  • S3 Block Public Access provides settings for access points, buckets, and accounts to help manage public access to S3 resources.
  • By default, new buckets, access points, and objects don’t allow public access. However, users can modify bucket policies, access point policies, or object permissions to allow public access.
  • S3 Block Public Access settings override these policies and permissions so that public access to these resources can be limited.
  • S3 Block Public Access allows account administrators and bucket owners to easily set up centralized controls to limit public access to their S3 resources that are enforced regardless of how the resources are created.
  • S3 doesn’t support block public access settings on a per-object basis.
  • S3 Block Public Access settings when applied to an account apply to all AWS Regions globally.

Resource Control Policies (RCPs)

  • Launched November 2024 – A new type of authorization policy in AWS Organizations for centrally restricting access to AWS resources.
  • RCPs work alongside Service Control Policies (SCPs) to establish permissions guardrails across multiple accounts in an organization.
  • Can enforce that S3 buckets in your accounts are only accessible by principals belonging to your organization.
  • Resource-based — they restrict what actions can be taken on resources regardless of the identity-based permissions granted.
  • Do not grant access — they only set the maximum available permissions on resources.
  • Key S3 use case: Restrict access to S3 buckets so no principal outside the organization can access them, regardless of individual S3 bucket policies.
  • Available in AWS GovCloud (US) Regions since May 2025.
  • Supported by AWS Control Tower with configurable managed controls.

S3 Access Analyzer

  • S3 Access Analyzer (part of IAM Access Analyzer) monitors the access policies, ensuring that the policies provide only the intended access to your S3 resources.
  • Evaluates bucket access policies and enables you to discover and swiftly remediate buckets with potentially unintended access.
  • External Access Findings — identifies resources accessible by principals outside your account or organization.
  • Internal Access Findings (June 2025) — identifies who within your AWS organization has access to critical resources including S3 buckets.
    • Monitors selected resources daily and surfaces findings in a unified dashboard
    • Combines internal and external access findings for a 360-degree view
  • Unused Access Analysis — identifies unused roles, access keys, and permissions to help achieve least privilege.
  • Custom Policy Checks — validate policies before deployment, including:
    • Check No Public Access (Jul 2024)
    • Check Access Not Granted
    • Guided revocation for unused access (Jun 2024)
  • Can be integrated into CI/CD pipelines for automated policy reviews.

S3 Security Best Practices

S3 Preventative Security Best Practices

  • Ensure S3 buckets use the correct policies and are not publicly accessible
    • Use S3 block public access
    • Keep ACLs disabled (Bucket owner enforced) unless specifically needed
    • Use AWS Trusted Advisor to inspect the S3 implementation.
    • Use Resource Control Policies (RCPs) to restrict access to organization principals only
  • Implement least privilege access
    • Use S3 Access Grants for complex identity-based access at scale
    • Use ABAC for tag-based authorization to simplify policy management
    • Use IAM Access Analyzer to identify and remediate overly permissive policies
  • Use IAM roles for applications and AWS services that require S3 access
  • Enable Multi-factor authentication (MFA) Delete to help prevent accidental bucket deletions
  • Consider Data at Rest Encryption
    • SSE-S3 is now applied by default to all new objects
    • Use SSE-KMS for additional audit trails and key management
    • Use DSSE-KMS for compliance requiring two independent layers of encryption
  • Enforce Data in Transit Encryption
  • Consider S3 Object Lock to store objects using a “Write Once Read Many” (WORM) model.
  • Enable versioning to easily recover from both unintended user actions and application failures.
  • Consider S3 Cross-Region replication
  • Consider VPC endpoints for S3 access to provide private S3 connectivity and help prevent traffic from potentially traversing the open internet.
  • Secure presigned URLs with short expiration times, IP restrictions, and scoped IAM permissions.

S3 Monitoring and Auditing Best Practices

  • Identify and Audit all S3 buckets to have visibility of all the S3 resources to assess their security posture and take action on potential areas of weakness.
  • Implement monitoring using AWS monitoring tools
  • Enable S3 server access logging, which provides detailed records of the requests that are made to a bucket useful for security and access audits
  • Use AWS CloudTrail, which provides a record of actions taken by a user, a role, or an AWS service in S3.
  • Enable AWS Config, which enables you to assess, audit, and evaluate the configurations of the AWS resources
  • Use Amazon Macie with S3 to automatically discover, classify, and protect sensitive data in AWS.
    • Automated sensitive data discovery continuously samples and analyzes objects
    • Provides sensitivity scores for each bucket
    • Builds interactive data maps showing sensitive data distribution across accounts
  • Use IAM Access Analyzer to monitor both external and internal access to S3 resources
  • Monitor AWS security advisories to regularly check security advisories posted in Trusted Advisor for the AWS account.

AWS Certification Exam Practice Questions

  • Questions are collected from Internet and the answers are marked as per my knowledge and understanding (which might differ with yours).
  • AWS services are updated everyday and both the answers and questions might be outdated soon, so research accordingly.
  • AWS exam questions are not updated to keep up the pace with AWS updates, so even if the underlying feature has changed the question might not be updated
  • Open to further feedback, discussion and correction.
  1. A company needs to ensure all objects stored in their S3 buckets are encrypted with a dual-layer encryption mechanism for CNSSP 15 compliance. Which encryption option should they use?
    1. SSE-S3
    2. SSE-KMS
    3. DSSE-KMS
    4. SSE-C

    Answer: c. DSSE-KMS — Dual-Layer Server-Side Encryption with AWS KMS keys provides two independent layers of encryption for compliance requirements like CNSSP 15.

  2. An organization wants to grant access to S3 buckets based on team membership from their Microsoft Entra ID directory, supporting up to 50,000 data scientists. Which S3 feature is most appropriate?
    1. S3 Bucket Policies
    2. S3 Access Points
    3. S3 Access Grants
    4. S3 ACLs

    Answer: c. S3 Access Grants — Maps corporate directory identities to S3 data permissions at scale (up to 100,000 grants per Region), ideal for large organizations with external IdPs.

  3. A security engineer needs to ensure that no S3 buckets in any account of the organization can be accessed by principals outside the organization, regardless of individual bucket policies. Which feature should they implement?
    1. S3 Block Public Access
    2. Service Control Policies (SCPs)
    3. Resource Control Policies (RCPs)
    4. S3 Bucket Policies with Deny statements

    Answer: c. Resource Control Policies (RCPs) — RCPs centrally restrict external access to AWS resources across an organization, enforcing that S3 buckets are accessible only by principals belonging to the organization.

  4. An administrator wants to simplify S3 bucket access management by automatically granting access based on environment tags (dev, staging, prod) rather than maintaining individual bucket ARNs in policies. What feature should they use?
    1. S3 Access Grants
    2. S3 ABAC (Attribute-Based Access Control)
    3. S3 Access Points
    4. AWS Organizations SCPs

    Answer: b. S3 ABAC — Enables tag-based authorization for S3 buckets, allowing access decisions based on matching tags between principals and buckets rather than individual bucket names.

  5. A company creates a new S3 bucket and attempts to upload objects using SSE-C encryption but receives an AccessDenied error. What is the most likely cause? (Choose the best answer)
    1. The IAM user lacks encryption permissions
    2. SSE-C is disabled by default on new buckets since April 2026
    3. The bucket policy blocks encryption operations
    4. KMS key permissions are missing

    Answer: b. — Since April 2026, SSE-C is disabled by default on all new general purpose buckets. Applications must explicitly enable SSE-C via the PutBucketEncryption API.

  6. Which IAM Access Analyzer feature, launched in June 2025, helps security teams identify who within their AWS organization has access to critical resources like S3 buckets?
    1. External Access Findings
    2. Internal Access Findings
    3. Unused Access Analysis
    4. Custom Policy Checks

    Answer: b. Internal Access Findings — Monitors resources daily and provides a unified dashboard combining internal and external access findings for a 360-degree view.

References

AWS Simple Storage Service – S3

~ Last updated on : ~ jayendrapatil ~ 50 Comments

AWS Simple Storage Service – S3

  • Amazon Simple Storage Service – S3 is a simple key, value object store designed for the Internet
  • provides unlimited storage space and works on the pay-as-you-use model. Service rates get cheaper as the usage volume increases
  • offers an extremely durable, highly available, and infinitely scalable data storage infrastructure at very low costs.
  • is Object-level storage (not Block level storage like EBS volumes) and cannot be used to host OS or dynamic websites (however, S3 can host static websites).
  • S3 resources e.g. buckets and objects are private by default.
  • As of March 2026, S3 stores more than 500 trillion objects, serves more than 200 million requests per second globally across hundreds of exabytes of data.
  • S3 provides strong read-after-write consistency for all operations (PUT, GET, LIST, DELETE, HEAD) automatically, at no additional cost, in all AWS Regions.
  • Starting January 5, 2023, all new objects are automatically encrypted with SSE-S3 (server-side encryption with Amazon S3 managed keys) by default at no additional cost.
  • Starting April 2023, all new S3 buckets have S3 Block Public Access enabled and ACLs disabled by default.
  • Starting April 2026, SSE-C (server-side encryption with customer-provided keys) is disabled by default on all new S3 general purpose buckets.

S3 Bucket Types

  • Amazon S3 offers multiple bucket types designed for different use cases:
    • General Purpose Buckets – Standard buckets for most workloads, storing objects across multiple Availability Zones for high durability
    • Directory Buckets – Used with S3 Express One Zone storage class, stored in a single Availability Zone for lowest latency access
    • Table Buckets – Store Apache Iceberg tables for analytics workloads with built-in table maintenance and optimization
    • Vector Buckets – Purpose-built for storing and querying vector embeddings for AI/ML applications

S3 Buckets & Objects

S3 Buckets

  • A bucket is a container for objects stored in S3
  • Buckets help organize the S3 namespace.
  • A bucket is owned by the AWS account that creates it and helps identify the account responsible for storage and data transfer charges.
  • Bucket names are globally unique, regardless of the AWS region in which it was created and the namespace is shared by all AWS accounts
  • Even though S3 is a global service, buckets are created within a region specified during the creation of the bucket.
  • Every object is contained in a bucket
  • There is no limit to the number of objects that can be stored in a bucket and no difference in performance whether a single bucket or multiple buckets are used to store all the objects
  • The S3 data model is a flat structure i.e. there are no hierarchies or folders within the buckets. However, logical hierarchy can be inferred using the key name prefix e.g. Folder1/Object1
  • Restrictions
    • 10,000 general purpose buckets (default quota) per AWS account, with the ability to request up to 1 million buckets. (Updated Nov 2024: increased from the previous limit of 100)
    • Bucket names should be globally unique and DNS compliant
    • Bucket ownership is not transferable
    • Buckets cannot be nested and cannot have a bucket within another bucket
    • Bucket name and region cannot be changed, once created
  • Empty or a non-empty buckets can be deleted
  • S3 allows retrieval of 1000 objects and provides pagination support

Objects

  • Objects are the fundamental entities stored in a bucket
  • An object is uniquely identified within a bucket by a key name and version ID (if S3 versioning is enabled on the bucket)
  • Objects consist of object data, metadata, and others
    • Key is the object name and a unique identifier for an object
    • Value is actual content stored
    • Metadata is the data about the data and is a set of name-value pairs that describe the object e.g. content-type, size, last modified. Custom metadata can also be specified at the time the object is stored.
    • Version ID is the version id for the object and in combination with the key helps to uniquely identify an object within a bucket
    • Subresources help provide additional information for an object
    • Access Control Information helps control access to the objects
  • S3 objects allow two kinds of metadata
    • System metadata
      • Metadata such as the Last-Modified date is controlled by the system. Only S3 can modify the value.
      • System metadata that the user can control, e.g., the storage class, and encryption configured for the object.
    • User-defined metadata
      • User-defined metadata can be assigned during uploading the object or after the object has been uploaded.
      • User-defined metadata is stored with the object and is returned when an object is downloaded
      • S3 does not process user-defined metadata.
      • User-defined metadata must begin with the prefix “x-amz-meta“, otherwise S3 will not set the key-value pair as you define it
  • Object metadata cannot be modified after the object is uploaded and it can be only modified by performing copy operation and setting the metadata
  • Objects belonging to a bucket that reside in a specific AWS region never leave that region, unless explicitly copied using Cross Region Replication
  • Each object can be up to 5 TB in size
  • An object can be retrieved as a whole or a partially
  • With Versioning enabled, current as well as previous versions of an object can be retrieved

S3 Bucket & Object Operations

  • Listing
    • S3 allows the listing of all the keys within a bucket
    • A single listing request would return a max of 1000 object keys with pagination support using an indicator in the response to indicate if the response was truncated
    • Keys within a bucket can be listed using Prefix and Delimiter.
    • Prefix limits result in only those keys (kind of filtering) that begin with the specified prefix, and the delimiter causes the list to roll up all keys that share a common prefix into a single summary list result.
  • Retrieval
    • An object can be retrieved as a whole
    • An object can be retrieved in parts or partially (a specific range of bytes) by using the Range HTTP header.
    • Range HTTP header is helpful
      • if only a partial object is needed for e.g. multiple files were uploaded as a single archive
      • for fault-tolerant downloads where the network connectivity is poor
    • Objects can also be downloaded by sharing Pre-Signed URLs
    • Metadata of the object is returned in the response headers
  • Object Uploads
    • Single Operation – Objects of size 5GB can be uploaded in a single PUT operation
    • Multipart upload – can be used for objects of size > 5GB and supports the max size of 5TB. It is recommended for objects above size 100MB.
    • Pre-Signed URLs can also be used and shared for uploading objects
    • Objects if uploaded successfully can be verified if the request received a successful response. Additionally, returned ETag can be compared to the calculated MD5 value of the upload object
  • Conditional Writes
    • S3 supports conditional writes using HTTP conditional headers to prevent unintended overwrites (Launched August 2024)
    • If-None-Match – prevents overwrites of existing objects by checking that no object with the same key exists; useful for write-once patterns
    • If-Match – ensures an object has not been modified since last read by comparing ETags; useful for read-modify-write patterns (Added November 2024)
    • Conditional writes can be enforced via bucket policies using s3:if-none-match and s3:if-match condition keys
    • Supported on PutObject, CompleteMultipartUpload, and CopyObject operations
    • Helps coordinate simultaneous writes from multiple writers without external locking mechanisms
  • Copying Objects
    • Copying of objects up to 5GB can be performed using a single operation and multipart upload can be used for uploads up to 5TB
    • When an object is copied
      • user-controlled system metadata e.g. storage class and user-defined metadata are also copied.
      • system controlled metadata e.g. the creation date etc is reset
    • Copying Objects can be needed to
      • Create multiple object copies
      • Copy objects across locations or regions
      • Renaming of the objects
      • Change object metadata for e.g. storage class, encryption, etc
      • Updating any metadata for an object requires all the metadata fields to be specified again
  • Deleting Objects
    • S3 allows deletion of a single object or multiple objects (max 1000) in a single call
    • For Non Versioned buckets,
      • the object key needs to be provided and the object is permanently deleted
    • For Versioned buckets,
      • if an object key is provided, S3 inserts a delete marker and the previous current object becomes the non-current object
      • if an object key with a version ID is provided, the object is permanently deleted
      • if the version ID is of the delete marker, the delete marker is removed and the previous non-current version becomes the current version object
    • Deletion can be MFA enabled for adding extra security
  • Restoring Objects from Glacier
    • Objects must be restored before accessing an archived object stored in S3 Glacier Flexible Retrieval or S3 Glacier Deep Archive
    • S3 Glacier Instant Retrieval provides millisecond access without requiring a restore operation
    • Retrieval options for Glacier Flexible Retrieval include:
      • Expedited – 1-5 minutes
      • Standard – 3-5 hours
      • Bulk – 5-12 hours
    • Restoration request also needs to specify the number of days for which the object copy needs to be maintained.
    • During this period, storage cost applies for both the archive and the copy.

Pre-Signed URLs

  • All buckets and objects are by default private.
  • Pre-signed URLs allows user to be able to download or upload a specific object without requiring AWS security credentials or permissions.
  • Pre-signed URL allows anyone to access the object identified in the URL, provided the creator of the URL has permission to access that object.
  • Pre-signed URLs creation requires the creator to provide security credentials, a bucket name, an object key, an HTTP method (GET for download object & PUT of uploading objects), and expiration date and time
  • Pre-signed URLs are valid only till the expiration date & time.
  • Pre-signed URLs can have a maximum expiration of 7 days when generated using SigV4.

Multipart Upload

  • Multipart upload allows the user to upload a single large object as a set of parts. Each part is a contiguous portion of the object’s data.
  • Multipart uploads support 1 to 10000 parts and each part can be from 5MB to 5GB with the last part size allowed to be less than 5MB
  • Multipart uploads allow a max upload size of 5TB
  • Object parts can be uploaded independently and in any order. If transmission of any part fails, it can be retransmitted without affecting other parts.
  • After all parts of the object are uploaded and completed initiated, S3 assembles these parts and creates the object.
  • Using multipart upload provides the following advantages:
    • Improved throughput – parallel upload of parts to improve throughput
    • Quick recovery from any network issues – Smaller part size minimizes the impact of restarting a failed upload due to a network error.
    • Pause and resume object uploads – Object parts can be uploaded over time. Once a multipart upload is initiated there is no expiry; you must explicitly complete or abort the multipart upload.
    • Begin an upload before the final object size is known – an object can be uploaded as is it being created
  • Three Step process
    • Multipart Upload Initiation
      • Initiation of a Multipart upload request to S3 returns a unique ID for each multipart upload.
      • This ID needs to be provided for each part upload, completion or abort request and listing of parts call.
      • All the Object metadata required needs to be provided during the Initiation call
    • Parts Upload
      • Parts upload of objects can be performed using the unique upload ID
      • A part number (between 1 – 10000) needs to be specified with each request which identifies each part and its position in the object
      • If a part with the same part number is uploaded, the previous part would be overwritten
      • After the part upload is successful, S3 returns an ETag header in the response which must be recorded along with the part number to be provided during the multipart completion request
    • Multipart Upload Completion or Abort
      • On Multipart Upload Completion request, S3 creates an object by concatenating the parts in ascending order based on the part number and associates the metadata with the object
      • Multipart Upload Completion request should include the unique upload ID with all the parts and the ETag information
      • The response includes an ETag that uniquely identifies the combined object data
      • On Multipart upload Abort request, the upload is aborted and all parts are removed. Any new part upload would fail. However, any in-progress part upload is completed, and hence an abort request must be sent after all the parts uploads have been completed.
      • S3 should receive a multipart upload completion or abort request else it will not delete the parts and storage would be charged.

S3 Transfer Acceleration

  • S3 Transfer Acceleration enables fast, easy, and secure transfers of files over long distances between the client and a bucket.
  • Transfer Acceleration takes advantage of CloudFront‘s globally distributed edge locations. As the data arrives at an edge location, data is routed to S3 over an optimized network path.
  • Transfer Acceleration will have additional charges while uploading data to S3 is free through the public Internet.

S3 Batch Operations

  • S3 Batch Operations help perform large-scale batch operations on S3 objects and can perform a single operation on lists of specified S3 objects.
  • A single job can perform a specified operation on billions of objects containing exabytes of data.
  • S3 tracks progress, sends notifications, and stores a detailed completion report of all actions, providing a fully managed, auditable, and serverless experience.
  • Batch Operations can be used with S3 Inventory to get the object list and use S3 Select to filter the objects.
  • Batch Operations can be used for copying objects, modify object metadata, applying ACLs, encrypting objects, transforming objects, invoke a custom lambda function, etc.

S3 Express One Zone

  • S3 Express One Zone is a high-performance, single-Availability Zone storage class designed for latency-sensitive applications (Launched November 2023)
  • Delivers data access speeds up to 10x faster and request costs up to 50-80% lower than S3 Standard
  • First S3 storage class where you can select a specific Availability Zone to co-locate storage with compute resources
  • Uses directory buckets instead of general purpose buckets, with a hierarchical namespace using forward slash (/) as delimiter
  • Designed for 99.95% availability within a single Availability Zone (vs. 99.99% for S3 Standard across multiple AZs)
  • Supports up to 200,000 reads and 100,000 writes per second per directory bucket
  • Ideal use cases:
    • Machine learning training and inference
    • Interactive analytics
    • Media content creation
    • High-performance computing (HPC)
    • Financial modeling
  • Uses session-based authentication (CreateSession API) for optimized request handling

S3 Tables (Apache Iceberg)

  • S3 Tables provide the first cloud object store with built-in Apache Iceberg support (Launched December 2024)
  • Optimized for analytics workloads with up to 3x faster query throughput and up to 10x higher transactions per second compared to self-managed tables
  • Stores tabular data in table buckets with tables as subresources
  • Provides automatic table maintenance including compaction, snapshot management, and unreferenced file removal
  • Supports Intelligent-Tiering access tiers for automatic cost optimization (Added 2025)
  • Integrates with analytics engines like Apache Spark, Trino, and Amazon Athena
  • Use cases: data lakes, business analytics, real-time analytics, and ML feature stores

S3 Vectors

  • S3 Vectors is the first cloud object storage with native support for storing and querying vector data (GA December 2025)
  • Reduces the total cost of storing and querying vectors by up to 90% compared to specialized vector database solutions
  • Uses a new bucket type — vector bucket — optimized for durable, low-cost vector storage
  • Supports up to 2 billion vectors per index and 10,000 vector indexes per vector bucket
  • Delivers sub-second latency for infrequent queries and ~100ms for frequent queries
  • Supports up to 50 metadata keys alongside each vector for fine-grained filtering
  • Ideal use cases:
    • AI agent persistent memory
    • Retrieval Augmented Generation (RAG)
    • Semantic search
    • Recommendation systems

S3 Files

  • S3 Files makes S3 buckets accessible as high-performance file systems on AWS compute resources (Launched April 2026)
  • First and only cloud object store that provides fully-featured, high-performance file system access via NFS v4.2
  • Provides full file system semantics with sub-millisecond latency on small files
  • Changes to data on the file system are automatically reflected in the S3 bucket
  • Can be attached to multiple compute resources enabling data sharing across clusters without duplication
  • Supported on EC2, Lambda, EKS, and ECS
  • Eliminates the tradeoff between object storage benefits and interactive file capabilities
  • Use cases: AI/ML training, legacy application migration, shared data access across compute

S3 Metadata

  • S3 Metadata automatically captures metadata for objects in general purpose buckets and stores it in read-only, fully managed Apache Iceberg tables (Preview Dec 2024, enhanced 2025)
  • Provides two types of metadata tables:
    • Journal table – records changes as objects are added or modified
    • Live inventory table – provides a complete current snapshot of all objects and their metadata
  • Accelerates data discovery for analytics, AI/ML model training, and content retrieval
  • Supports metadata for existing objects via backfill (Added July 2025)
  • Queryable using standard SQL via Amazon Athena, Spark, and other analytics engines

S3 Access Grants

  • S3 Access Grants provide a simplified model for defining access permissions to S3 data by prefix, bucket, or object (Launched November 2023)
  • Maps corporate identities from directories (Microsoft Entra ID, Okta) directly to S3 datasets without requiring IAM principal mapping
  • Integrates with AWS IAM Identity Center for trusted identity propagation
  • Logs end-user identity and application used to access S3 data in AWS CloudTrail
  • Integrates with AWS Glue, Amazon Redshift, and Lake Formation for analytics workloads
  • Provides fine-grained access control at the prefix or object level

Mountpoint for Amazon S3

  • Mountpoint for Amazon S3 is an open-source file client that mounts an S3 bucket as a local file system on Linux instances (GA August 2023)
  • Translates local file system API calls to S3 REST API calls automatically
  • Optimized for high-throughput read-heavy workloads (sequential and random reads, sequential writes)
  • Available as a CSI driver for Kubernetes/EKS containerized workloads
  • Backed by AWS support for customers with Business and Enterprise Support plans
  • Use cases: data lakes, machine learning training, HPC, media processing
  • Note: For full file system semantics including NFS access, see S3 Files (launched April 2026)

Virtual Hosted Style vs Path-Style Request

S3 allows the buckets and objects to be referred to in Path-style or Virtual hosted-style URLs

Path-style

  • Bucket name is not part of the domain (unless region specific endpoint used)
  • Endpoint used must match the region in which the bucket resides for e.g, if you have a bucket called mybucket that resides in the EU (Ireland) region with object named puppy.jpg, the correct path-style syntax URI is http://s3-eu-west-1.amazonaws.com/mybucket/puppy.jpg.
  • A “PermanentRedirect” error is received with an HTTP response code 301, and a message indicating what the correct URI is for the resource if a bucket is accessed outside the US East (N. Virginia) region with path-style syntax that uses either of the following:
    • http://s3.amazonaws.com
    • An endpoint for a region different from the one where the bucket resides for e.g., if you use http://s3-eu-west-1.amazonaws.com for a bucket that was created in the US West (N. California) region
  • Path-style URLs were planned for deprecation after September 30, 2020, but AWS has indefinitely delayed this plan. Virtual hosted-style is still recommended for all new implementations.

Virtual hosted-style

  • S3 supports virtual hosted-style and path-style access in all regions.
  • In a virtual-hosted-style URL, the bucket name is part of the domain name in the URL for e.g. http://bucketname.s3.amazonaws.com/objectname
  • S3 virtual hosting can be used to address a bucket in a REST API call by using the HTTP Host header
  • Benefits
    • attractiveness of customized URLs,
    • provides an ability to publish to the “root directory” of the bucket’s virtual server. This ability can be important because many existing applications search for files in this standard location.
  • S3 updates DNS to reroute the request to the correct location when a bucket is created in any region, which might take time.
  • S3 routes any virtual hosted-style requests to the US East (N.Virginia) region, by default, if the US East (N. Virginia) endpoint s3.amazonaws.com is used, instead of the region-specific endpoint (for e.g., s3-eu-west-1.amazonaws.com) and S3 redirects it with HTTP 307 redirect to the correct region.
  • When using virtual hosted-style buckets with SSL, the SSL wild card certificate only matches buckets that do not contain periods.To work around this, use HTTP or write your own certificate verification logic.
  • If you make a request to the http://bucket.s3.amazonaws.com endpoint, the DNS has sufficient information to route the request directly to the region where your bucket resides.

S3 Pricing

  • S3 costs vary by Region
  • S3 pricing has dropped approximately 85% since launch, with current rates as low as ~$0.021/GB/month for S3 Standard in US regions
  • Charges are incurred for
    • Storage – cost is per GB/month
    • Requests – per request cost varies depending on the request type GET, PUT
    • Data Transfer
      • data transfer-in is free
      • data transfer out is charged per GB/month (except in the same region or to Amazon CloudFront)

S3 Select (Maintenance Mode)

  • S3 Select is closed to new customers as of July 25, 2024. Existing customers can continue to use the service.
  • S3 Select enabled applications to retrieve only a subset of data from an object using simple SQL expressions
  • Recommended alternatives: S3 Object Lambda, Amazon Athena, or S3 Metadata with Apache Iceberg for querying object data

Additional Topics

AWS Certification Exam Practice Questions

  • Questions are collected from Internet and the answers are marked as per my knowledge and understanding (which might differ with yours).
  • AWS services are updated everyday and both the answers and questions might be outdated soon, so research accordingly.
  • AWS exam questions are not updated to keep up the pace with AWS updates, so even if the underlying feature has changed the question might not be updated
  • Open to further feedback, discussion and correction.
  1. What does Amazon S3 stand for?
    1. Simple Storage Solution.
    2. Storage Storage Storage (triple redundancy Storage).
    3. Storage Server Solution.
    4. Simple Storage Service
  2. What are characteristics of Amazon S3? Choose 2 answers
    1. Objects are directly accessible via a URL
    2. S3 should be used to host a relational database
    3. S3 allows you to store objects or virtually unlimited size
    4. S3 allows you to store virtually unlimited amounts of data
    5. S3 offers Provisioned IOPS
  3. You are building an automated transcription service in which Amazon EC2 worker instances process an uploaded audio file and generate a text file. You must store both of these files in the same durable storage until the text file is retrieved. You do not know what the storage capacity requirements are. Which storage option is both cost-efficient and scalable?
    1. Multiple Amazon EBS volume with snapshots
    2. A single Amazon Glacier vault
    3. A single Amazon S3 bucket
    4. Multiple instance stores
  4. A user wants to upload a complete folder to AWS S3 using the S3 Management console. How can the user perform this activity?
    1. Just drag and drop the folder using the flash tool provided by S3
    2. Use the Enable Enhanced Folder option from the S3 console while uploading objects
    3. The user cannot upload the whole folder in one go with the S3 management console
    4. Use the Enable Enhanced Uploader option from the S3 console while uploading objects (NOTE – The S3 console now natively supports folder upload via drag and drop without any special option)
  5. A media company produces new video files on-premises every day with a total size of around 100GB after compression. All files have a size of 1-2 GB and need to be uploaded to Amazon S3 every night in a fixed time window between 3am and 5am. Current upload takes almost 3 hours, although less than half of the available bandwidth is used. What step(s) would ensure that the file uploads are able to complete in the allotted time window?
    1. Increase your network bandwidth to provide faster throughput to S3
    2. Upload the files in parallel to S3 using mulipart upload
    3. Pack all files into a single archive, upload it to S3, then extract the files in AWS
    4. Use AWS Import/Export to transfer the video files
  6. A company is deploying a two-tier, highly available web application to AWS. Which service provides durable storage for static content while utilizing lower Overall CPU resources for the web tier?
    1. Amazon EBS volume
    2. Amazon S3
    3. Amazon EC2 instance store
    4. Amazon RDS instance
  7. You have an application running on an Amazon Elastic Compute Cloud instance, that uploads 5 GB video objects to Amazon Simple Storage Service (S3). Video uploads are taking longer than expected, resulting in poor application performance. Which method will help improve performance of your application?
    1. Enable enhanced networking
    2. Use Amazon S3 multipart upload
    3. Leveraging Amazon CloudFront, use the HTTP POST method to reduce latency.
    4. Use Amazon Elastic Block Store Provisioned IOPs and use an Amazon EBS-optimized instance
  8. When you put objects in Amazon S3, what is the indication that an object was successfully stored?
    1. Each S3 account has a special bucket named_s3_logs. Success codes are written to this bucket with a timestamp and checksum.
    2. A success code is inserted into the S3 object metadata.
    3. A HTTP 200 result code and MD5 checksum, taken together, indicate that the operation was successful.
    4. Amazon S3 is engineered for 99.999999999% durability. Therefore there is no need to confirm that data was inserted.
  9. You have private video content in S3 that you want to serve to subscribed users on the Internet. User IDs, credentials, and subscriptions are stored in an Amazon RDS database. Which configuration will allow you to securely serve private content to your users?
    1. Generate pre-signed URLs for each user as they request access to protected S3 content
    2. Create an IAM user for each subscribed user and assign the GetObject permission to each IAM user
    3. Create an S3 bucket policy that limits access to your private content to only your subscribed users’ credentials
    4. Create a CloudFront Origin Identity user for your subscribed users and assign the GetObject permission to this user
  10. You run an ad-supported photo sharing website using S3 to serve photos to visitors of your site. At some point you find out that other sites have been linking to the photos on your site, causing loss to your business. What is an effective method to mitigate this?
    1. Remove public read access and use signed URLs with expiry dates.
    2. Use CloudFront distributions for static content.
    3. Block the IPs of the offending websites in Security Groups.
    4. Store photos on an EBS volume of the web server.
  11. You are designing a web application that stores static assets in an Amazon Simple Storage Service (S3) bucket. You expect this bucket to immediately receive over 150 PUT requests per second. What should you do to ensure optimal performance?
    1. Use multi-part upload.
    2. Add a random prefix to the key names.
    3. Amazon S3 will automatically manage performance at this scale. (S3 automatically scales to handle at least 3,500 PUT/COPY/POST/DELETE and 5,500 GET/HEAD requests per second per partitioned prefix, with no prefix randomization needed)
    4. Use a predictable naming scheme, such as sequential numbers or date time sequences, in the key names
  12. What is the maximum number of S3 buckets available per AWS Account?
    1. 100 Per region
    2. There is no Limit
    3. 100 Per Account (Previously correct, but updated Nov 2024)
    4. 500 Per Account
    5. 100 Per IAM User
    6. 10,000 Per Account (default), up to 1 million per account by request (Updated Nov 2024)
  13. Your customer needs to create an application to allow contractors to upload videos to Amazon Simple Storage Service (S3) so they can be transcoded into a different format. She creates AWS Identity and Access Management (IAM) users for her application developers, and in just one week, they have the application hosted on a fleet of Amazon Elastic Compute Cloud (EC2) instances. The attached IAM role is assigned to the instances. As expected, a contractor who authenticates to the application is given a pre-signed URL that points to the location for video upload. However, contractors are reporting that they cannot upload their videos. Which of the following are valid reasons for this behavior? Choose 2 answers { “Version”: “2012-10-17”, “Statement”: [ { “Effect”: “Allow”, “Action”: “s3:*”, “Resource”: “*” } ] }
    1. The IAM role does not explicitly grant permission to upload the object. (The role has all permissions for all activities on S3)
    2. The contractorsˈ accounts have not been granted “write” access to the S3 bucket. (using pre-signed urls the contractors account don’t need to have access but only the creator of the pre-signed urls)
    3. The application is not using valid security credentials to generate the pre-signed URL.
    4. The developers do not have access to upload objects to the S3 bucket. (developers are not uploading the objects but its using pre-signed urls)
    5. The S3 bucket still has the associated default permissions. (does not matter as long as the user has permission to upload)
    6. The pre-signed URL has expired.
  14. A company wants to prevent concurrent writers from accidentally overwriting each other’s data in Amazon S3. Which S3 feature should they use?
    1. S3 Object Lock
    2. S3 Versioning with MFA Delete
    3. S3 Conditional Writes with If-None-Match or If-Match headers
    4. S3 Block Public Access
  15. A machine learning team needs the lowest latency access to frequently accessed training data stored in S3, and their compute resources are in a single Availability Zone. Which S3 storage class is MOST appropriate?
    1. S3 Standard
    2. S3 Intelligent-Tiering
    3. S3 Express One Zone
    4. S3 One Zone-Infrequent Access
  16. An organization wants to grant S3 data access to users based on their corporate directory identity without creating individual IAM users. Which S3 feature enables this? [Choose 1]
    1. S3 Bucket Policies with IAM conditions
    2. S3 ACLs with cross-account access
    3. S3 Access Grants with IAM Identity Center
    4. S3 Object Lambda Access Points
  17. Which of the following are S3 bucket types available as of 2025? (Choose 3)
    1. General purpose buckets
    2. Directory buckets
    3. Archive buckets
    4. Table buckets
    5. Compute buckets
  18. A data engineering team needs to automatically track and query metadata about millions of objects in their S3 bucket for data discovery. Which service should they use?
    1. S3 Inventory with Athena
    2. S3 Select with SQL queries
    3. S3 Metadata with managed Apache Iceberg tables
    4. AWS Glue Data Catalog