EBS Flashcards
of test questions related to EBS
4-5
EBS
Elastic Block Store
- storage volumes you can attach to your EC2 instances
What are ways you can use EBS?
Use them the same way you would use any system disk
- to create a file system
- run a database
- run an operating system
- to store data
- install applications
Can EBS be used for mission critical data?
Yes, it is highly available = automatically replicated within a single AZ to protect against hardware failures
How much downtime is required to increase capacity or change volume type?
No downtime required - it is scalable.
- You can make changes without downtime or performance impact
IOPS
Input/Output per second
GP2
General Purpose Solid State (SSD)
- a balance of price and performance
- 99.9% durability
GP3
General Purpose SSD
- predictable
- can scale up for an additional fee
(don’t need to know the IOPS or MiB/s for the exam)
What is GP2 good for?
- Good for boot volumes or development & test applications that are not latency sensitive
- 99.9% durability
What is GP3 good for?
- applications that require high performance at low cost
- mySQL, Cassandra, virtual desktops, Hadoop analytics
How much faster is the top performance of GP3 over GP2?
GP3 is 4x faster than the max throughput of GP2 volumes
io1
Provisioned IOPS SSD - (PIOPS)
- the most expensive
- high performance options
- 99.9% durability
What is io1 (PIOPS) good for?
- if you need more than 16,000 IOPS
- I/O-intensive applications
- large databases
- latency-sensitive workloads/transactions
io2
Provisioned IOPS SSD
- latest generation
- higher durability - 99.999%
- same price as io1
What is io2 (PIOPS) good for?
- I/O-intensive apps
- large databases
- latency-sensitive workloads
- applications that need high durability
- transactions
st1
ST1 - Throughput Optimized Hard Disk Drive
- low cost HDD volume
- maximum throughput of 500 MB/s per volume
What is st1 good for?
- frequently accessed, throughput-intensive workloads
- big data, data warehouse, ETLS and log processing
- cost effective way to store mountains of data
Which types of EBS cannot be boot volumes?
st1 and sc1
sc1
sc1 - Cold Hard Disk Drive
- lowest cost option
- poor performance
What is sc1 good for?
- for cold data requiring fewer scans per day
- for applications where lowest cost is needed when performance is not a factor
IOPS basics
IOPS:
- measures # of reads & writes per second
- important metric for quick transactions, low-latency apps, and transactional workloads
- the ability to action reads/writes very quickly
Throughput basics
Throughput:
- measures # of bits read or written per second
- important metric for: Large datasets, large I/O sizes, complex queries
- the ability to deal with large datasets
EBS volume
a virtual hard disk
- need a minimum of 1 volume per EC2 instance for the OS “root device volume”
EBS Snapshot
an S3 snapshot is a point-in-time copy/backup of the volume
- snapshots are incremental, only the changed pieces are moved to S3 (like gitlab)
- saves space and time because only the changes get moved over
- first snapshot will take the longest to establish the baseline
Snapshot consistency
stop the instance and detach the volume before doing a snapshot, this will allow any cached data to be saved
if you take a snapshot of an encrypted EBS volume, will it be automatically encrypted?
yes
Sharing Snapshots**
- you can share within the region where it was created
- to share with other regions, you will need to copy them to the new destination region and then create an image once it is in the new region.
Are volumes always in the same AZ as your EC2 instance?**
Yes
What must you do when you resize the EBS volume on the fly?
must extend the filesystem in the OS so the OS can see the resized volume
EBS Encryption
- uses a data key w/ industry standard AES-256 algorithm
- uses KMS & Customer-Managed Keys (CMKs)
What do you get with an encrypted volume?*
End-to-End Encryption
- data at rest is encrypted
- data in flight moving between instance and volume is encrypted
- all snapshots are encrypted
- all volumes created from the snapshot are encrypted
Steps to encrypt an unencrypted volume**
1) create a snapshot of the root device volume
2) create a copy of the snapshot and chose the encrypt option**
3) Create an AMI from the encrypted snapshot
4) Use the AMI to launch new encrypted instances
EC2 Hibernation
when an EC2 instance is terminated, by default the root device volume is terminated
Hibernation saves RAM to Root volume. When you wake it up everything is restored.
What does EC2 hibernation do?*
- saves contents of instance memory (RAM) to your EBS root volume
- persists the root volume and any attached EBS data volumes
What happens when you wake EC2 up from hibernation?
- EBS root volume is restored to its previous state
- RAM contents are reloaded
- processes that were previously running on the instance are resumed
- previous attached data volumes are reattached
- instance retains its Instance ID
Why does an instance boot faster from hibernation?
The OS does not need to reboot b/c the RAM is preserved
What is hibernation useful for?*
- long-running processes
- services that take a long time to initialize
What is a restriction for hibernation
RAM must be less than 150 GB.
What is max hibernation period?
60 days
Which instance families can hibernate?
C3, C4, C5
M3, M4, M5
R3, R4, R5
