Change EC2 Instance Type
—
### **1. Stop the Instance**
1. Go to the **EC2 Dashboard** in the AWS Management Console.
2. Select the instance you want to modify.
3. Click on **Actions > Instance State > Stop Instance**.
4. Confirm the action to stop the instance.
> **Note:** Stopping the instance will temporarily interrupt its operation, so ensure you plan this action during a maintenance window if needed.
—
### **2. Change the Instance Type**
1. After the instance is stopped, select it again.
2. Click on **Actions > Instance Settings > Change Instance Type**.
3. In the dialog that appears:
– Select the desired **Instance Type** from the dropdown list.
– Ensure the new instance type is compatible with your instance's AMI (Amazon Machine Image) and the underlying architecture.
4. Click **Apply**.
—
### **3. Start the Instance**
1. Select the modified instance.
2. Click on **Actions > Instance State > Start Instance**.
—
### **4. Verify the Changes**
1. Once the instance is running, check the **Instance Type** column in the EC2 Dashboard to confirm the changes.
2. Test the instance to ensure that your application and services are functioning correctly.
—
### **Additional Considerations**
– **EBS Volume Compatibility:** Ensure that the instance type supports the current storage configuration.
– **Networking:** Some instance types require an enhanced networking setup (e.g., Elastic Network Adapter). Make necessary adjustments.
– **Billing:** Switching to a larger instance type may increase costs. Be sure to review the pricing.
– **Elastic IPs:** If you're using an Elastic IP, it will remain attached to your instance even after the type is changed.
Would you like detailed instructions on verifying compatibility or adjusting networking settings?
—
### **1. Pricing Model**
Amazon EC2 instances can be billed under different pricing models:
– **On-Demand Pricing:** Pay for the compute capacity by the hour or second (minimum of 1 minute) with no long-term commitments.
– **Reserved Instances:** Commit to a 1- or 3-year term to receive significant discounts compared to On-Demand pricing.
– **Savings Plans:** Flexible pricing model offering lower prices compared to On-Demand in exchange for a commitment to a certain amount of usage (measured in $/hour).
– **Spot Instances:** Purchase unused capacity at a significant discount but with the risk of termination if AWS needs the capacity back.
—
### **2. Components of Cost**
The cost includes:
#### **a. Instance Running Time**
– **Billing Unit:** AWS bills based on seconds, with a minimum of 60 seconds.
– **Hourly Rate:** Depends on the region and pricing model. For example, On-Demand pricing for T3.small in **US East (N. Virginia)** might be around **$0.0208 per hour** (as of Dec 2024).
– **Formula:**
[
text{Instance Cost} = text{Hourly Rate} times text{Number of Hours Used}
]
#### **b. vCPU Credit Charges (for T3 instances)**
– T3 instances are **burstable performance instances** and earn **CPU credits** when usage is low.
– If the instance exceeds the baseline performance, additional vCPU credits are used.
– **vCPU Credit Cost:** $0.05 per vCPU-hour (varies by region) if exceeding earned credits.
– **Formula:**
[
text{vCPU Credit Cost} = (text{vCPU Usage} – text{Baseline vCPU Allocation}) times 0.05
]
#### **c. Additional Costs**
Other associated costs may include:
– **EBS Storage:** Charged separately based on the size and type of EBS volume attached to the instance.
– **Data Transfer:**
– Inbound data: Usually free.
– Outbound data: Charged based on the amount of data transferred out of AWS.
—
### **3. Example Calculation**
#### Scenario:
– **T3.small instance** in **US East (N. Virginia)**:
– On-Demand Rate: $0.0208/hour
– Baseline Performance: 20% CPU (1 vCPU × 0.2 = 0.2 vCPU-hours/hour).
– Instance runs for **10 hours** and uses an additional **0.1 vCPU-hours** beyond baseline performance.
#### Calculation:
1. **Instance Cost:**
[
10 , text{hours} times 0.0208 = 0.208 , text{USD}
]
2. **vCPU Credit Cost:**
[
0.1 , text{vCPU-hours} times 0
—
### **Assumptions:**
1. **Instance Type:** T3.small (2 vCPUs, 2 GB RAM).
2. **Pricing Model:** On-Demand.
3. **Region:** Example, **US East (N. Virginia)**.
– **On-Demand Rate:** $0.0208/hour.
– **vCPU Credit Cost:** $0.05 per vCPU-hour (only if usage exceeds baseline performance).
4. **Baseline Performance:** 20% CPU (0.2 vCPU-hours/hour).
5. **Month Duration:** 30 days (720 hours).
—
### **Cost Breakdown:**
#### **1. On-Demand Cost (Hourly Rate)**
[
text{On-Demand Cost} = 720 , text{hours} times 0.0208 , text{USD/hour}
]
[
text{On-Demand Cost} = 14.98 , text{USD}
]
#### **2. vCPU Credit Cost**
– Baseline CPU usage (covered by credits):
[
720 , text{hours} times 0.2 , text{vCPU-hours/hour} = 144 , text{vCPU-hours}
]
– If the instance uses more than this, additional credits are charged at $0.05 per vCPU-hour. Let’s assume **10% extra usage** beyond baseline:
[
text{Extra vCPU Usage} = 720 , text{hours} times 0.1 , text{vCPU-hours/hour} = 72 , text{vCPU-hours}
]
[
text{vCPU Credit Cost} = 72 , text{vCPU-hours} times 0.05 , text{USD/vCPU-hour} = 3.60 , text{USD}
]
#### **3. EBS Storage**
– Assume **30 GB of General Purpose (GP2) SSD** at $0.10 per GB-month:
[
text{Storage Cost} = 30 , text{GB} times 0.10 , text{USD/GB-month} = 3.00 , text{USD}
]
#### **4. Data Transfer (Outbound)**
– First 1 GB is free. Additional outbound data is typically **$0.09 per GB**.
– Let’s assume **50 GB outbound traffic** in a month:
[
text{Data Transfer Cost} = (50 – 1) , text{GB} times 0.09 , text{USD/GB} = 4.41 , text{USD}
]
—
### **Total Monthly Cost**
[
text{Total Cost} = text{On-Demand Cost} + text{vCPU Credit Cost} + text{Storage Cost} + text{Data Transfer Cost}
]
[
text{Total Cost} = 14.98 + 3.60 + 3.00 + 4.