Availability, Resilience & Elasticity Flashcards
When using Target Tracking Policies, what do you need to define?
A single target value for a single metric, for instance 50% CPU utilization. You’re NOT defining a range such as “between 25% and 75% CPU usage”.
What is the difference between the Cooldown Period and the Healthcheck Grace Period in the context of Auto Scaling?
The Cooldown Period is the interval that Auto Scaling waits, before attempting any further scaling activities. This helps avoiding that too many instances are started or terminated in a short time.
The Healthcheck Grace Period is related to EC2 Auto Scaling and defines the interval that Auto Scaling waits before evaluating if a EC2 target instance is healthy or not. This is helpful if instances require a longer initialization time, for instance because of a length bootstrap script. Without Grace Period, the instance might be terminated due to being detected as unhealthy before the script had a chance finishing.
At which point in time can you set the attributes for a Launch Configuration in the context of ELB?
- At the time of creating the Launch Configuration
- After creating the Launch Configuration
Only at the time of creating the Launch Configuration
Which scaling policies make use of Cool Down Periods with Amazon EC2 Auto Scaling?
Only with simple scaling policies
What are the three dynamic scaling policies available for you to use with Amazon EC2 Auto Scaling?
- Simple scaling
- Step scaling
- Target Tracking scaling
What is the main difference between simple scaling policy, step scaling policy, and target tracking policy in Amazon EC2 Auto Scaling?
Simple scaling policies allow to increase/decrease capacity when a threshold value is hit. After changing capacity, it waits until the cooldown period expires and new health checks take effect, due to which the scaling can be slower than required by the demand.
Step scaling allow for a more granular scale out/in definition, as capacity can be added/removed with different values at different thresholds.
Target Tracking policy doesn’t use upper and lower thresholds, but instead uses a single target value to decide whether to add or remove capacity.
How do you achieve high-availability with Redshift?
Couple of options:
- do frequents snapshots and restore from snapshot when Redshift goes down (backup + restore)
- run Redshift in a multi-node cluster (which supports data replication and node recovery)
- run whole different Redshift cluster, either in same, or different region (active/active)
What are the five pillars of the Well-Architected Framework?
- Security
- Performance Efficiency
- Cost Optimization
- Reliability
- Operational Excellence
Which of the following design principles are accounted to the Reliability pillar of the Well-Architected Framework (5 in total)?
- Prepare for security events
- Keep people away from data
- Lear from [..] failures
- Test recovery procedures
- Anticipate failure
- Implement cloud financial management
- Perform operations as code
- Enable traceability
- Automate security best practices
- Refine operations procedures frequently
- Analyze and attribute expenditure
- Scale horizontally [..]
- Apply security at all layers
- Stop guessing capacity
- Automatically recover from failure
- Manage change in automation
- Experiment more often
- Go global in minutes
- Make frequent, small, reversible changes
- Democratize advanced technologies
- Stop spending money on undifferentiated heavy lifting
- Consider mechanical empathy
- Measure overall efficiency
- Adopt a consumption model
- Protect data in transit and at rest
- Implement a strong identity foundation
- Use serverless architectures
- Automatically recover from failure
- Test recovery procedures
- Scale horizontally [..]
- Stop guessing capacity
- Manage change in automation
Which of the following design principles are accounted to the Operational Excellence pillar of the Well-Architected Framework (5 in total)?
- Prepare for security events
- Keep people away from data
- Lear from [..] failures
- Test recovery procedures
- Anticipate failure
- Implement cloud financial management
- Perform operations as code
- Enable traceability
- Automate security best practices
- Refine operations procedures frequently
- Analyze and attribute expenditure
- Scale horizontally [..]
- Apply security at all layers
- Stop guessing capacity
- Automatically recover from failure
- Manage change in automation
- Experiment more often
- Go global in minutes
- Make frequent, small, reversible changes
- Democratize advanced technologies
- Stop spending money on undifferentiated heavy lifting
- Consider mechanical empathy
- Measure overall efficiency
- Adopt a consumption model
- Protect data in transit and at rest
- Implement a strong identity foundation
- Use serverless architectures
- Perform operations as code
- Make frequent, small, reversible changes
- Refine operations procedures frequently
- Anticipate failure
- Lear from [..] failures
Which of the following design principles are accounted to the Security pillar of the Well-Architected Framework (7 in total)?
- Prepare for security events
- Keep people away from data
- Lear from [..] failures
- Test recovery procedures
- Anticipate failure
- Implement cloud financial management
- Perform operations as code
- Enable traceability
- Automate security best practices
- Refine operations procedures frequently
- Analyze and attribute expenditure
- Scale horizontally [..]
- Apply security at all layers
- Stop guessing capacity
- Automatically recover from failure
- Manage change in automation
- Experiment more often
- Go global in minutes
- Make frequent, small, reversible changes
- Democratize advanced technologies
- Stop spending money on undifferentiated heavy lifting
- Consider mechanical empathy
- Measure overall efficiency
- Adopt a consumption model
- Protect data in transit and at rest
- Implement a strong identity foundation
- Use serverless architectures
- Implement a strong identity foundation
- Enable traceability
- Apply security at all layers
- Automate security best practices
- Protect data in transit and at rest
- Keep people away from data
- Prepare for security events
Which of the following design principles are accounted to the Cost Optimization pillar of the Well-Architected Framework (5 in total)?
- Prepare for security events
- Keep people away from data
- Lear from [..] failures
- Test recovery procedures
- Anticipate failure
- Implement cloud financial management
- Perform operations as code
- Enable traceability
- Automate security best practices
- Refine operations procedures frequently
- Analyze and attribute expenditure
- Scale horizontally [..]
- Apply security at all layers
- Stop guessing capacity
- Automatically recover from failure
- Manage change in automation
- Experiment more often
- Go global in minutes
- Make frequent, small, reversible changes
- Democratize advanced technologies
- Stop spending money on undifferentiated heavy lifting
- Consider mechanical empathy
- Measure overall efficiency
- Adopt a consumption model
- Protect data in transit and at rest
- Implement a strong identity foundation
- Use serverless architectures
- Implement cloud financial management
- Adopt a consumption model
- Measure overall efficiency
- Stop spending money on undifferentiated heavy lifting
- Analyze and attribute expenditure
Which of the following design principles are accounted to the Performance Efficiency pillar of the Well-Architected Framework (5 in total)?
- Prepare for security events
- Keep people away from data
- Lear from [..] failures
- Test recovery procedures
- Anticipate failure
- Implement cloud financial management
- Perform operations as code
- Enable traceability
- Automate security best practices
- Refine operations procedures frequently
- Analyze and attribute expenditure
- Scale horizontally [..]
- Apply security at all layers
- Stop guessing capacity
- Automatically recover from failure
- Manage change in automation
- Experiment more often
- Go global in minutes
- Make frequent, small, reversible changes
- Democratize advanced technologies
- Stop spending money on undifferentiated heavy lifting
- Consider mechanical empathy
- Measure overall efficiency
- Adopt a consumption model
- Protect data in transit and at rest
- Implement a strong identity foundation
- Use serverless architectures
- Democratize advanced technologies
- Go global in minutes
- Use serverless architectures
- Experiment more often
- Consider mechanical empathy
What’s the difference between EC2 “Termination Protection” and “Instance Protection”?
“Termination Protection” protects EC2 instances from being terminated via the console, SDK or API commands. But it does not protect from termination that is triggered by an Auto Scaling Group. For this scenario, “Instance Protection” must be used instead.
