PuglieseWeb
  • Home
  • Software development
    • Cloud Data Security Principles
      • Separation of Duties (SoD)
      • Security Controls and Data Protection Framework
      • Vaultless Tokenization
    • Multi-cloud strategies
    • DMS
      • How CDC Checkpoints Work
      • Oracle to PostgreSQL Time-Window Data Reload Implementation Guide
      • Join tables separate PostgreSQL databases
      • Multi-Stage Migration Implementation Plan
      • Notes
      • Oracle Golden Gate to PostgreSQL Migration
      • Step-by-Step CDC Recovery Guide: Oracle to PostgreSQL Migration
    • AWS Pro
      • My notes
        • Data Migration Strategy
        • OpsWorks VS CloudFormation
      • Implementation Guides
        • AWS Lambda Scaling and Concurrency Optimization Guide
        • Understanding Cross-Account IAM Roles in AWS
        • HA TCP with Redundant DNS
        • Understanding 429 (Too Many Requests) & Throttling Pattern
        • EC2 Auto Scaling Log Collection Solutions Comparison
        • AWS PrivateLink Implementation Guide for Third-Party SaaS Integration
        • AWS Cross-Account Network Sharing Implementation Guide
        • Cross-Account Route 53 Private Hosted Zone Implementation Guide
          • Route 53
            • Routing Policies
              • Using a Weighted Routing Policy
              • Simple Routing Policy
              • Multivalue Answer Routing
            • Latency Routing Policy
            • Route 53 Traffic Flow
        • Direct Connect Gateway Implementation Guide
        • CICD for Lambda
        • AWS IAM Identity Center Integration with Active Directory
        • AWS Transit Gateway Multi-Account Implementation Guide
          • AWS Multi-Account Network Architecture with Infrastructure Account
      • Links
      • Cloud Adoption Framework
      • Data Stores
        • Data Store Types and Concepts in AWS
        • S3
          • Amazon S3 (Simple Storage Service)
            • Bucket Policies
          • Managing Permissions in Amazon S3
          • Amazon Glacier: AWS Archive Storage Service
          • Lab: Querying Data in Amazon S3 with Amazon Athena
          • LAB: Loading Data into a Redshift Cluster
        • Attached Storage
          • EBS
          • AWS Elastic File System (EFS): From Sun Microsystems to Modern Cloud Storage
          • AWS FSx Service Guide
          • Amazon Storage Gateway Guide
        • Databases
          • Amazon Storage Gateway Guide
          • Amazon RDS (Relational Database Service)
          • Aurora DB
          • Dynamo DB
          • Document DB
          • Amazon Redshift Overview
          • Data Pipeline
            • Data Lake VS Lake Formation
          • AWS Data Preparation Services
          • Amazon Neptune
          • Amazon ElastiCache
          • AWS Specialized Database Services
          • LAB - Deploy an Amazon RDS Multi-AZ and Read Replica in AWS
      • Networking
        • Concept
        • Basics
          • VPG
          • VPC
            • VPC endpoints
              • Interface Endpoint VS Elastic Network Interface (ENI)
            • PrivateLink
              • PrivateLink SAAS Use case
            • Transit Gateway
            • 5G Networks
            • VPN CloudHub
            • VPC security
            • VPC peering
            • VPC Endpoint
            • Route Table (and Routers)
            • Network Access Control List (NACL)
            • Network Security Group
            • NAT Gateway
              • NACL vs NAT
          • Elastic Load Balancing (ELB)
            • Gateway Load Balancer (GWLB)
          • CIDR ranges examples
          • Enhanced Networking
          • Elastic Fabric Adapter (EFA)
          • Elastic Network Interface (ENI)
        • Network to VPC Connectivity
          • Transit VS Direct Connect Gateway
          • Direct Connect
            • VIF (Virtual Interfaces)
            • VIF VS ENI
            • Customer Routers VS Customer Gateways
        • VPC-to-VPC
        • NAT & Internet Gateway
        • Routing
          • IPv4 Address Classes and Subnet Masks
          • VPC's DNS server
          • Transit VPC VS Transit Gateway
          • Example Routing tables configuration
          • Cross-regions failover
          • Loopback
        • Enhanced Networking
        • Hybrid and Cross-Account Networking
        • AWS Global Accelerator
        • Route 53
        • Cross-Account Route 53
        • CloudFront SSL/TLS and SNI Configuration
        • ELB
        • Lab: Creating a Multi-Region Network with VPC Peering Using SGs, IGW, and RTs
        • LAB - Creating a CloudFront Distribution with Regional S3 Origins
        • Lab: Creating and Configuring a Network Load Balancer in AWS
        • Lab: Troubleshooting Amazon EC2 Network Connectivity
        • Lab: Troubleshooting VPC Networking
      • Security
        • Cloud Security
          • IAM
            • SCIM
            • Use case 1
          • Core Concepts of AWS Cloud Security
            • OAuth VS OpenID Connect
          • Understanding User Access Security in AWS Organizations
          • Exploring Organizations
          • Controlling Access in AWS Organizations
            • SCP (Service Control Policy) implementation types
        • Network Controls and Security Groups
          • Firewalls
            • Network Controls and Security Groups Overview
          • AWS Directory Services
          • AWS Identity and Access Management (IAM) and Security Services
            • ASW Identity Sources
          • AWS Resource Access Manager (RAM): Cross-Account Resource Sharing
            • AWS App Mesh
        • Encryption
          • History and Modern Implementation of Encryption in AWS
          • Secret Manager
          • DDoS Attacks and AWS Protection Strategies: Technical Overview
          • AWS Managed Security Services Overview
          • IDS and IPS
          • AWS Service Catalog
      • Migrations
        • Migration Concepts
          • Hybrid Cloud Architectures
          • Migration Strategies
        • Migration Application
          • Services and Strategies
          • AWS Data Migration Services
          • Network Migrations and Cutovers
            • Network and Broadcast Addresses
            • VPC DNS
          • AWS Snow Family
      • Architecting to scale
        • Scaling Concepts and Services
          • Auto-Scaling
          • Compute Optimizer
          • Kinesis
          • DynamoDB Scaling
          • CloudFront Part Duex
            • CloudFront's Behavior
            • Lambda@Edge and CloudFront Functions
        • Event-Driven Architecture
          • SNS and Fan-out Architecture
            • SNS & outbox pattern
          • AWS Messaging Services: SQS and Amazon MQ
          • Lab: Scaling EC2 Using SQS
          • Lambda
          • Scaling Containers in AWS
          • Step Function and Batch
          • Elastic MapReduce
          • AWS Data Monitoring and Visualization Services
      • Business Continuity
        • AWS High Availability and Disaster Recovery
        • AWS Disaster Recovery Architectures
        • EBS Volumes
        • AWS Compute Options for High Availability
        • AWS Database High Availability Options
        • AWS Network High Availability Options
        • Lab: Connect Multiple VPCs with Transit Gateway
        • Deployment and Operations Management
          • Software Deployment Strategies
            • AWS CI/CD
            • Elastic Beanstalk
              • Elastic Beanstalk and App Runner
            • CloudFormation
            • Cross-Account Infrastructure Deployment
              • Example Code Pipeline
            • AWS Container Services
            • AWS API Gateway
            • LAB: Understanding CloudFormation Template Anatomy
          • Management Tool
            • Config and OpsWorks
            • System Manager
            • Enterprise Apps
            • AWS Machine Learning Landscape
            • AWS IoT Services
      • Cost Management and Optimization
        • Concepts
        • AWS Cost Optimization Strategies
        • AWS Tagging and Resource Groups
        • Managing Costs Across AWS Accounts
        • AWS Instance Purchasing Options
        • AWS Cost Management Tools
      • Others
        • SCPs vs AWS Config
        • Questions notes
        • Comparison of Deployment Strategies in AWS
        • Bedrock vs EMR
        • Software Deployment Strategies
    • AWS
      • Others
        • AWS Example architectures
          • Gaming application
          • Digital Payment System
            • Marketplace Application
            • Analytics & Reporting System MVP
            • Reporting System 2
            • Data Pipeline
            • Monitoring and visualization solution for your event-driven architecture (EDA) in AWS.
              • Visualize how services are linked together for each business flow
              • Visualize flow and metrics
            • Reporting
            • Data
        • AWS Key Learning
        • AWS NFRs
          • AWS Integration Pattern Comparison Matrix
          • AWS 99.999% Architecture
        • AWS Best Practices
          • use S3 for data migration
          • Principle of centralized control
          • For CPU Spikes in DB use RDS Proxy
          • API Security
          • Lambda VS ECS
          • Use CloudFront for Dynamic content
        • ECS Sizing
        • AWS Q&A
          • AWS Prep
          • prepexam
          • Big Data/ AI Q&A
          • DB Q&A
          • AWS Application Servers Q&A
          • General Q&A
          • VPC Q&A
      • DRs
      • AI, Analytics, Big Data, ML
        • EMR
          • Flink
          • Spark
          • Hadoop
            • Hive
        • Extra
          • Glue and EMR
          • Redshift Use Cases
        • AI
          • Media Services (Elastic Transcoder, Kinesis)
          • Textract
          • Rekognition (part of the exam)
          • Comprehend
          • Kendra
          • Fraud Detector
          • Transcribe, Polly, Lex
          • Translate
          • Time-series and Forecast
        • Big Data
          • Processing & Analytics
            • Amazon Athena VS Amazon Redshift
            • Athena & AWS Glue: Serverless Data Solutions
          • BigData Storage Solutions
          • EMR
        • Business intelligence
        • Sagemaker
          • SageMaker Neo
          • Elastic Inference (EI)
          • Integration patterns with Amazon SageMaker
          • Common Amazon SageMaker Endpoint usage patterns
          • Real-time interfaces
          • ML Example
        • Machine Learning
          • Data Engineering
            • Understanding Data Preparation
            • Feature Engineering: Transforming Raw Data into Powerful Model Inputs
            • Feature Transformation and Scaling in Machine Learning
            • Data Binning: Transforming Continuous Data into Meaningful Categories
          • Exploratory Data Analysis
            • Labs
              • Perform Feature Engineering Using Amazon SageMaker
            • Categorical Data Encoding: Converting Categories to Numbers
            • Text Feature Extraction for Machine Learning
            • Feature Extraction from Images and Speech: Understanding the Fundamentals
            • Dimensionality Reduction and Feature Selection in Machine Learning
          • Modelling
            • Prerequisites for Machine Learning Implementation
            • Classification Algorithms in Machine Learning
            • Understanding Regression Algorithms in Machine Learning
            • Time Series Analysis: Fundamentals and Applications
            • Clustering Algorithms in Machine Learning
      • Databases
        • Capturing data modification events
        • Time-Series Data (Amazon Timestream)
        • Graph DBs
          • Amazon Neptune
        • NoSQL
          • Apache Cassandra (Amazon Keyspaces)
          • Redshift
            • Redshift's ACID compliance
          • MongoDB (Amazon DocumentDB)
          • DynamoDB
            • Additional DynamoDB Features and Concepts
            • DynamoDB Consistency Models and ACID Properties
            • DynamoDB Partition Keys
          • Amazon Quantum Ledger DB (QLDB)
        • RDS
          • DR for RDS
          • RDS Multi-AZ VS RDS Proxy
          • Scaling Relational Databases
          • Aurora Blue/Green deployments
          • Aurora (Provisioned)
          • Amazon Aurora Serverless
        • Sharing RDS DB instance with an external auditor
      • Caching
        • DAX Accelerator
        • ElastiChache
        • CloudFront (External Cache)
        • Global Accelerator (GA)
      • Storages
        • S3
          • MFA Delete VS Object Lock
          • S3 Standard VS S3 Intelligent-Tiering
        • Instance Storage
        • EBS Volumes
          • Burst Capacity & Baseline IOPS
          • Provisioned IOPS vs GP3
          • EBS Multi-Attach
        • Snapshots
        • AWS Backup
        • File Sharing
          • FSx (File system for Windows or for Lustre)
          • EFS (Elastic File System)
      • Migration
        • Migration Hub
        • Application Discovery Service
        • Snow Family
        • DMS
        • SMS (Server Migration Service)
        • MGN (Application Migration Service)
        • Transfer family
        • DataSync
        • Storage Gateway
          • Volume gateway
          • Tape Gateway
          • File Gateway
          • Storage Gateway Volume Gateway VS Storage Gateway File Gateway
        • DataSync VS Storage Gateway File Gateway
      • AWS Regional Practices and Data Consistency Regional Isolation and Related Practices
      • Front End Web application
        • Pinpoint
        • Amplify
        • Device Farm
      • Glossary
      • Governance
        • Well-Architected Tool
        • Service Catalog and Proton
          • AWS Service Catalog
          • AWS Proton
        • AWS Health
        • AWS Licence Manager
        • AWS Control Tower
        • AWS Trusted Advisor
        • Saving Plans
        • AWS Compute Optimizer
        • AWS CUR
        • Cost Explorer and Budgets
        • Directory Service
        • AWS Config
        • Cross-Account Role Access
        • Resource Access Manager (RAM)
        • Organizations, Accouts, OU, SCP
      • Automation
        • System Manager (mainly for inside EC2 instances)
        • Elastic Beanstalk (for simple solutions)
        • IaC
          • SAM
          • CloudFormation
            • !Ref VS !GetAtt
            • CloudFormation examples
      • Security
        • Identity Management Services
          • IAM
            • Identity, Permission, Trust and Resource Policies
              • IAM Policy Examples
              • Trust policy
            • IAM roles cannot be attached to IAM Groups
            • AWS IAM Policies Study Guide
            • Cross-Account Access in AWS: Resource-Based Policies vs IAM Roles
            • EC2 instance profile VS Trust policy
          • Cognito
        • STS
        • AI based security
          • GuardDuty
          • Macie (S3)
        • AWS Network Firewall
        • Security Hub
        • Detective (Root Cause Analysis)
        • Inspector (EC2 and VPCs)
        • System Manager Parameter Store
        • Secret Manager
          • Secret Manger VS System Manager's Parameter Store
          • Secret Manager VS AWS KMS
        • Shield
          • DDoS
        • KMS vs CloudHSM
        • Firewall Manager
        • AWS WAF
      • Compute
        • Containers
          • ECS
            • ECS Anywhere
          • EKS
            • EKS Anywhere
          • Fargate
            • ECS Fargate VS EKS Fargate
          • ECR (Elastic Container Registry)
        • EC2
          • EC2 Purchase Options
            • Spot instances VS Spot Fleet
          • EC2 Instance Types
            • T Instance Credit System
          • Auto Scaling Groups (ASG)
          • Launch Template vs. Launch Configuration
          • AMI
          • EC2 Hibernation
        • Lambda
          • Publish VS deploy
      • Data Pipeline
      • ETL
      • AppFlow
      • AppSync
      • Step Functions
      • Batch
        • Spring Boot Batch VS AWS Batch
      • Decoupling Workflow
      • Elastic Load Balancers
      • Monitoring
        • OpenSearch
        • CloudWatch Logs Insights VS AWS X-Ray
        • QuickSight
        • Amazon Managed Service for Prometheus
        • Amazon Managed Grafana
        • CloudWatch Logs Insights
          • CloudWatch Logs Insights VS Kibana VS Grafana
        • CloudWatch Logs
        • CloudTrail
        • CloudWatch
        • X-Ray
      • On-Premises
        • ECS/EKS Anyware
        • SSM Agent
      • Serverless Application Repository
      • Troubleshooting
      • Messaging, Events and Streaming
        • Kinesis (Event Stream)
        • EventBridge (Event Router)
          • EventBridge Rule Example
          • EventBridge vs Apache Kafka
          • EventBridge VS Kinesis(Event Stream)
          • Event Bridge VS SNS
        • SNS (Event broadcaster)
        • SQS (Message Queue)
        • MSK
        • Amazon MQ
        • DLQ
    • Software Design
      • CloudEvents
        • CloudEvents Kafka
      • Transaction VS Operation DBs
      • Event-based Microservices
        • Relations database to event messages
      • Hexagonal Architecture with Java Spring
      • Distributed Systems using DDD
        • Scaling a distributed system
        • Zookeeper
        • Aggregates
        • Bounded Context
      • API Gateway
      • Cloud
        • The Twelve Factors
        • Open Service Broker API
      • Microservices
    • Design technique
    • Technologies
      • Kafka
      • Docker
        • Docker Commands
        • Artifactory
        • Dockerfile
      • ReactJs
        • Progressive Web App (PWA)
        • Guide to File Extensions in React Projects
    • Guides
      • OCP
      • AWS
        • Creating and Assuming an Administrator AWS IAM Role
        • Standing Up an Amazon Aurora Database with an Automatically Rotated Password Using AWS Secrets Manag
        • Standing Up an Apache Web Server EC2 Instance and Sending Logs to Amazon CloudWatch
        • Creating a Custom AMI and Deploying an Auto Scaling Group behind an Application Load Balancer
        • Assigning Static IPs to NLBs with ALB Target Groups
        • Hosting a Wordpress Application on ECS Fargate with RDS, Parameter Store, and Secrets Manager
        • Amazon Athena, Amazon S3, and VPC Flow Logs
      • Creating a CloudTrail Trail and EventBridge Alert for Console Sign-Ins
      • Load Balancer VS Reverse Proxy
      • Health check
      • Load Balancer
      • HTTP Protocol
      • TCP/IP Network Model
      • Event-base Microservices Implementation Guideline
      • How to write a service
      • Observability
      • Kafka Stream
      • Security
        • Securing Properties
          • HashiCorp Vault
      • Kubernates
      • Unix
        • Networking
        • Firewall
        • File system
        • alternatives
      • Setup CentOS 8 and Docker
    • Dev Tools
      • Docker Commands
      • Intellij
      • CheatSheets
        • Unix Commands
        • Vim Command
      • Templates
  • Working for an enterprise
    • Next step
    • Job roles
      • SME role
    • Common issues
Powered by GitBook
On this page
  • EBS Volume tyeps
  • AWS SSD Volume Types
  • GP2 (General Purpose SSD)
  • GP3 (Latest Generation General Purpose SSD)
  • io1 (Provisioned IOPS SSD):
  • io2 (Provisioned IOPS SSD):
  • AWS HDD Volume Types
  • st1 (Throughput Optimized HDD):
  • sc1 (Cold HDD):

Was this helpful?

  1. Software development
  2. AWS
  3. Storages

EBS Volumes

Elastic Block Store (EBS) is highly available and scalable storage volumes you can attach to EC2 instances.

EBS volumes provide durable, block-level storage that can be attached to Amazon EC2 instances. Key points include:

  1. Usage:

    • Primarily designed for use with EC2 instances.

    • Can be used as storage for some AWS services like Amazon RDS and Amazon WorkSpaces.

  2. Compatibility:

    • Not directly usable with Amazon Redshift or VMware Cloud on AWS.

    • Redshift uses its own cluster storage.

    • VMware Cloud on AWS uses vSAN for storage.

  3. Flexibility:

    • Volumes can be resized on the fly without detaching from instances.

    • Volume types can be changed dynamically (e.g., from GP2 to GP3).

  4. Data Persistence:

    • EBS volumes persist independently from the life of an EC2 instance.

  5. Snapshots:

    • Point-in-time snapshots can be created for backup or volume recreation.

  6. Encryption:

    • Supports encryption at rest using AWS Key Management Service (KMS).

  7. Multi-Attach:

    • Some volume types support attaching to multiple EC2 instances simultaneously.

EBS Volume tyeps

Feature
sc1
st1
GP2
GP3
io1
io2

Type

HDD

HDD

SSD

SSD

SSD

SSD

Use Case

Infrequently accessed data

Big data, warehouses, ETL

Boot volumes, general-purpose

Broad range of workloads

I/O-intensive, databases

I/O-intensive, ultra-low latency

Throughput

Up to 250 MB/s

Up to 500 MB/s

Up to 250 MiB/s

Up to 1,000 MiB/s

Up to 1,000 MiB/s

Up to 1,000 MiB/s (4,000 MiB/s for Block Express)

Max IOPS

250

500

16,000

16,000

64,000

64,000 (256,000 for Block Express)

Baseline IOPS

12 per TB

40 per TB

3 IOPS/GiB (min 100)

3,000 (Additional IOPS can be added in increments of 500 IOPS)

Provisioned (up to 50 IOPS/GiB)

Provisioned (up to 500 IOPS/GiB)

Burst Capability

Yes (up to 80 MB/s per TB)

Yes (up to 250 MB/s per TB)

Yes (up to 3,000 IOPS)

No (consistent performance)

No

No

Volume Size

125 GiB - 16 TiB

125 GiB - 16 TiB

1 GiB - 16 TiB

1 GiB - 16 TiB

4 GiB - 16 TiB

4 GiB - 16 TiB

Durability

99.8-99.9%

99.8-99.9%

99.8-99.9%

99.8-99.9%

99.8-99.9%

Up to 99.999%

Boot Volume

No

No

Yes

Yes

Yes

Yes

Cost

Lowest

Low

Moderate

Lower than GP2

High

Same as io1

AWS SSD Volume Types

GP2 (General Purpose SSD)

GP2 volumes provide millisecond latency and are well-suited for small, random I/O operations.

GP2 volumes provide 3 IOPS per GB, capped at 16,000 IOPS.

  • Suitable for boot volumes and general-purpose workloads.

  • Baseline performance of 3 IOPS/GiB, with a minimum of 100 IOPS.

  • Can burst up to 3,000 IOPS for volumes 334 GiB and smaller.

  • Maximum of 16,000 IOPS per volume (for volumes larger than 5,334 GiB).

  • Throughput up to 250 MiB/s.

  • 99.8-99.9% durability.

GP2 IOPS calculation

General Purpose SSD (gp2) IOPS are dependent on volume size at a rate of 3 IOPS/GB.

For GP2 (General Purpose SSD) volumes, the IOPS calculation follows these rules:

  1. Baseline: 3 IOPS per GB

  2. Minimum: 100 IOPS

  3. Maximum: 16,000 IOPS

For a 300 GB volume: 300 GB × 3 IOPS/GB = 900 IOPS

To get 1000 IOPS with GP2, you would need approximately 334 GB of storage (1000 ÷ 3 = 333.33).

GP3 (Latest Generation General Purpose SSD)

General Purpose SSD (gp3) includes 3,000 IOPS at no additional cost independent of volume size.

  • Suitable for a broad range of workloads.

  • Predictable 3,000 IOPS baseline performance and 125 MiB/s regardless of volume size.

  • Can provision up to 16,000 IOPS and 1,000 MiB/s per volume.

  • More cost-effective than GP2 for most workloads.

  • 99.8-99.9% durability.

io1 (Provisioned IOPS SSD):

Provisioned IOPS SSD (io1) suitable Sub-millisecond latency and vest for I/O-intensive workloads.

PIOPS would let you provision exactly the IOPS needed without overprovisioning storage.

  • Suitable for I/O-intensive workloads, particularly database workloads.

  • Provision up to 50 IOPS per GiB.

  • Up to 64,000 IOPS per volume.

  • Up to 1,000 MiB/s of throughput per volume.

  • High performance and higher cost.

  • 99.8-99.9% durability.

io2 (Provisioned IOPS SSD):

  • Latest generation of Provisioned IOPS volumes.

  • Designed for I/O-intensive workloads requiring consistent ultra-low latency.

  • Provision up to 500 IOPS per GiB.

  • Up to 64,000 IOPS per volume (256,000 IOPS for io2 Block Express).

  • Up to 1,000 MiB/s of throughput per volume (4,000 MiB/s for io2 Block Express).

  • Higher durability: up to 99.999%.

  • Same price as io1 with better performance and durability.

AWS HDD Volume Types

st1 (Throughput Optimized HDD):

  • Designed for frequently accessed, throughput-intensive workloads.

  • Ideal for big data, data warehouses, log processing, and ETL workloads.

  • Baseline throughput of 40 MB/s per TB.

  • Ability to burst up to 250 MB/s per TB.

  • Maximum throughput of 500 MB/s per volume.

  • Baseline IOPS of 40 per TB, with ability to burst up to 500 IOPS per TB.

  • Maximum IOPS of 500 per volume.

  • Cannot be used as a boot volume.

  • Minimum volume size of 125 GiB.

  • 99.8-99.9% durability.

sc1 (Cold HDD):

  • Designed for less frequently accessed workloads.

  • Suitable for scenarios where the lowest storage cost is important.

  • Baseline throughput of 12 MB/s per TB.

  • Ability to burst up to 80 MB/s per TB.

  • Maximum throughput of 250 MB/s per volume.

  • Baseline IOPS of 12 per TB, with ability to burst up to 80 IOPS per TB.

  • Maximum IOPS of 250 per volume.

  • Cannot be used as a boot volume.

  • Minimum volume size of 125 GiB.

  • 99.8-99.9% durability.

  • Lower cost compared to st1, making it ideal for infrequently accessed data.

PreviousInstance StorageNextBurst Capacity & Baseline IOPS

Last updated 5 months ago

Was this helpful?