Main Menu

Section Menu

• Introduction to Elastic Block Store
• AWS EBS Volume Types
• Monitoring AWS EBS volumes
• Quiz

Important Links

AWS EBS Metrics Descriptions: https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/monitoring-volume-status.html?icmpid=docs_ec2_console

Introduction to Elastic Block Store

https://www.udemy.com/aws-monitoring-alerting-with-aws-cloudwatch-and-aws-sns/learn/v4/t/lecture/7082812?start=0

Objectives

• Understand AWS EBS
• EBS performance factors
• EBS storage options
• Monitoring EBS

What is EBS

• Elastic Block Storage
• Provides block level storage for EC2 instances
• There are other options
  • Instance Store services
    • Temporary Block level storage
    • Physically attached to the Host
    • Deleted whenever the VM is stopped.
• EBS volumes can be attached, detached and re-attached to EC2 VMs
  • Must be in the same Availability Zone
  • Add data is preserved

Types of EBS volumes

• GP2 – General Purpose SSD
  • Default
  • Balances price vs. performance for a wide variety of transactional workloads
  • Computing IOPS
    • 3 IOPS per GB up to 10,000 IOPS
    • 100 IOPS minimum
    • As volume size increases, speed increases
    • 100GB volume = 300 IOPS
  • 8 GB Minimum size
    • Smaller volumes can burst to 3000 IOPS
  • Maximum throughput of 160MB/Second
  • IO Credits represent the available bandwidth that GP2 volumes can use to burst large amounts of IOs when more baseline performance is needed.
    • The more credits the volume has for IO, the more time it can burst beyond the baseline performance level.
      • With bursting, the better it performs when more performance is needed.
    • Volumes accumulate IOPS credits at the rate of 3 IOPS per GB of the volume size.
    • Each volume receives an initial balance of 5.4 million I/O credits
      • This is enough to sustain the maximum burst performance of 3000 IOPs for 30 minutes.
    • When a volume requires more than the baseline, it draws on its I/O credits to burst to the required performance level up to a maximum of 3000 IOPS.
    • When volumes use fewer I/O credits than it is in a second, unused I/O credits are added to the credit balance.
      • If a volume sits idle, it accumulates credits.
      • These are then consumed as needed
• IO1 – Provisioned IOPS SSD
  • Best suited for high performance scenarios and Mission critical applications
  • Costs more than GP2
• ST1 – Throughput Optimized HDD
• SC1 – Cold HDD
• Magnetic (Previous version?)

• Input/Output Operations per Second IOPS is standard measurement for EBS performance
  • SSD performance measured in blocks of 256KB I/O Operations (Random)
    • Better performance for random operations
  • HDD performance measured in blocks of 1024KB I/O Operations (Sequential)
    • Better performance for sequential operations

AWS EBS Volume Types

https://www.udemy.com/aws-monitoring-alerting-with-aws-cloudwatch-and-aws-sns/learn/v4/t/lecture/7082814?start=0

io1

• Provisioned SSD
• Best suited for throughput intensive workloads
• Baseline peformance is 30 IOPS/GB (10x GP2)
  • Maximum 20,000 IOPs
  • Maximum 320 MB/Second
  • No Credit system

HDD

Both types designed for hard streaming workloads

ST1 – Throughput Optimized

• Frequently accessed data volumes
• Optimized for throughput intensive workloads

SC1 – Cold

• Less frequently accessed workloads
• Not good for high disk throughput
• Least expensive option available for AWS volumes

Monitoring AWS EBS volumes

https://www.udemy.com/aws-monitoring-alerting-with-aws-cloudwatch-and-aws-sns/learn/v4/t/lecture/7082816?start=0

• Two types
  • Basic 5 minute intervals
  • Detailed 1 minute
  • io1 sends at 1 min. intervals by default
    • This is why best for mission-critical apps.
• Status checks
  • Help the end user track and manage potential inconsistencies in the data
  • Automated test that run and return Pass/Fail every 5 minutes
  • Status
    • Okay – All checks pass
    • Impaired – One or more checks fail
      • Disables I/O between the volume and the VM by default
        • This can be over-ridden by enabling “AutoEnableIO” volume attribute
          • EC2 > Volumes > Select volume > Actions > Change Auto-Enable IO Settings

Demo

• EC2 > Volumes > Select Volume > Status Checks
  • Volume Status: Okay
  • IO Status: Enabled
  • IO Performance: N/A (Only available with io1 volumes)
• EC2 > Volumes > Select Volume > Monitoring
  • R/W Bandwidth (aka R/W Throughput)

Quiz

https://www.udemy.com/aws-monitoring-alerting-with-aws-cloudwatch-and-aws-sns/learn/v4/t/lecture/7082816?start=0

AWS EBS stands for

• Elastic Block Store
• Elastic Balancer Store
• Elastic Back Up Store
• Elastic Block Stabilizer

AWS provides 2 types of SSD volumes. They are

• io1 and gp1
• ios and gp2
• io1 and gp2
• io2 and gp2

“The Status Checks help the end user to better track and manage potential inconsistencies in the data on an EBS volume”

• True
• False

CloudWatch has two tiers of metrics available for EBS volumes. They are:

• Basic and Detailed
• Load and Detailed
• Run and Detailed
• Basic and Load

“When EBS determines that a volume’s data is potentially inconsistent, it disables the I/O to the volume from the EC2 instance by default”

• True
  • Trying to talk to the instructor regarding this.  Seems it is NOT enabled by default.
• False

After some research, it appears this setting is Enabled on Root volumes and Disabled on additional volumes.