Monitoring Flashcards
gp2
General Purpose SSD Boot Volumes - minimum of 100 IOPS - 3 IOPS/GB -> max of 16,000 IOPS Virtual Desktops Low-latency interactive apps Dev and Test environments
io1
Provisioned IOPS Boot Volumes More than 10,000 IOPS or 160 MiB/s of throughput per volume 50 IOPS/GB -> max of 64,000 IOPS. Large database workloads: MongoDB, Cassandra, Microsoft SQL Server
Which EBS Volumes can be used as Boot volumes?
gp2 and io1 (SSD volumes)
EBS Volume (?)
Streaming workloads requiring consistent, fast throughput at a low price.
- Big Data
- Data Warehouse
- Log processes
st1
Throughput-oriented storage for large volumes of data that is infrequently accessed.
Scenarios where the lowest storage cost is important
sc1
VolumeReadOPS
VolumeWriteOPS
What do they calculate?
How do you calculate it?
Used to calculate IOPS performance.
Total number of IO Ops in a specified period of time.
1000 in 1 minute => 1000/60 = 16,66 IOPS
Number of read and write ops waiting to be completed in a specified period of time. (EBS)
Volume Queue Length
Warning Volume Status
Degraded
Severly Degraded
Impaired Volume Status
Stalled
Not Available
4 Different Ways to Monitor ELBs
CloudWatch metrics
Access Logs
Request Tracing
CloudTrail logs
Monitor ELB - Access Logs
IT SHOWS WHO IS ACCESSING YOUR LOAD BALANCER.
Captures detailed info about REQUESTS sent to your LB.
You can use it to analyze traffic patterns and troubleshoot issues.
Info like:
- The time the request was received
- Client’s IP address
- Latencies
- Request Paths
- Server Responses
Desabled by default - intervals of 5 or 60 minutes.
Logs are stored in S3, so you can recover logs even after ec2 was terminated.
Track HTTP requests from client to targets or other services.
When the LB receives a request from a client, it adds or updates the X-Amzn-Trace-Id header before sending the request to the target.
APPLICATION LOAD BALANCER only.
Monitor ELB - Request Tracing
Capture detailed info about the calls made to the ELB API and store them as log files in S3.
Used to determine which:
- calls were made
- the source IP address where the call came from
- who made the call
- when the call was made, etc.
Monitor ELB - CloudTrail logs
CloudWatch Dashboards are multi-region, but to create a widget you need to be in the correct region for the service. True or False?
True
How to get automatic notification if your account goes over $1,000?
Go into CloudWatch and create a Billing Alarm, billing alarm uses a SNS Topic, which it will email you letting you know.
Allows you to manage multiple AWS accounts at once.
- centrally manage policies across multiple AWS accounts
- control access to your AWS services
- automate AWS account creation and management
- consolidate your billing across multiple accounts
AWS Organizations
SCPs
service control policies
- allow or deny accounts to use individual AWS services
Example:
- you might deny the use of Kinesis or DynamoDB to your HR group within your AWS Organization
Resource Groups
Lets you manage and automate tasks on large numbers of resources at one time.
Examples:
- Applying updates or security patches.
- Upgrading applications.
- Opening or closing ports to network traffic
- Collecting specific log and monitoring data from your fleet of instances.
Use Resource Groups with AWS Systems manager for automation.
AWS Systems Manager
Allows you to centralize operational data from multiple AWS services and automate tasks across your AWS resources.
Tool that enables you to view and analyze your costs and usage.
- You can view data for up to the last 13 months - Forecast how much you're likely to spend in the next 3 months - Get recommendations for Reserved Instances to purchase.
Cost Explorer
How to use Cost Explorer? 3 Steps
- Use tags to tag resources
- Configure tags for cost centers (dpt, employee id, etc)
- Activate cost allocation tags to track your costs by tags (Billing & Cost Management)
4 Different Ways to Monitor ElastiCache
ElastiCache
- CPU Utilization
- Swap Usage
- Eviction
- Concurrent Connections
Memcached CPU Utilization
Multi-threade
- Can handle loads of up to 90%
- Add more nodes to the cluster
Redis CPU Utilization
Not Multi-threaded
Scale = 90 / num of cores
Example: cache.m1.xlarge node = 4 cores
90/4 = 22.5% -> when to scale
