Generative AI: Can We Resolve Some AWS Issues With Amazon Q?
The era of generative AI is here! As cloud engineers, we can’t handle everything, and AI presents an opportunity to enhance our decision-making in this exciting time. Working on infrastructures requires knowledge and patience to tackle different issues daily. Even world-class products sometimes experience breakdowns, so having Amazon Q as an AI companion is a great opportunity to work closely with this tool.
Every day is an opportunity to learn something new and issues may come anytime without our consent no matter the design and code written!
Discover my experience with Amazon Q.
First Impression
Amazon Q is welcoming and knows a lot about AWS services! once a question is asked it provides links to a particular documentation, a great way to learn and understand AWS services step by step no matter the update.
In this review, we have 5 issues to resolve with Q :
- Understanding the services and documentation easily
- Knowing the number of VPCs with cloud shell
- Detecting issues between a VPC and an Internet gateway
- Knowing if the cost and billing summary is correct
- How to launch an EC2 instance easily
Issue 1: Navigating Through Endless Documentation Online
With more than 100 services, it’s difficult to master all!
and when it comes to searching for a particular service online or through the AWS docs we fall in an ocean of blogs, white papers …
Amazon Q summarizes every service clearly and links to particular documentation.
In the first issue, I asked Q to explain how many services we have with AWS, it summarized the services and gave me the link to sources where I can learn easily.
with such experience, it’s now easier to learn services better than before with detailed and summarized information, with a simple chat we can understand many things before browsing the web or complex documentation.
Issue 2: Knowing the Number of VPCs With Cloud Shell
AWS remains the biggest cloud provider in the world when it comes to the documentation and services to learn; it’s like an ocean of information! Sometimes, we wonder how many things we should master, and the commands to execute every time we encounter a problem. Most of the time, as cloud engineers, we work with the CLI, and we can’t memorize all the commands! Thanks to Q, we can now resolve some problems with less googling!
I asked Q a question, and thanks to its clear explanation, I went on Cloud Shell.
On the cloud shell, the command is run and I can see two VPCs! Do you have to navigate through the graphic user interface? Not mandatory. The shell offers more opportunities to execute complex tasks; with Q, I can learn many things daily!
The AI companion has many data and is a huge opportunity to boost our daily skills.
Issue 3: Detecting Issues Between a VPC and an Internet Gateway
On the console, we have 2 VPCs in the east region; let’s detach the internet gateway to one VPC to see how Q will understand the problem!
Let’s ask Q a question :
The question is asked and Q provides some details and suggests previewing the troubleshooting experience.
Q knows the name I gave to the internet gateway “Route to Dallas” and the ID.
The VPC is attached now to the internet gateway, unfortunately, due to compliance, Q cannot display some information easily. As a tool, it may provide inaccurate information; sometimes, it does not understand some questions no matter the context, even if we regenerate!
Issue 4: Knowing if the Cost and Billing Summary Is Correct
The AWS platform remains complex with so many things to learn, such as the features, the documentation, and updates that come all the time for improvement. I wanted to know if Q would understand why all my cost summary is at zero $.
Q was able to understand the question and provided some details correctly! Q can make mistakes all the time. It’s not 100 % accurate, the information provided is true because this account is new, and not many tasks were performed.
Issue 5: How to Launch an EC2 Instance Easily
The instance was launched after following the detailed steps, and I could see the state “Running.”
I am curious, and I would like to know the IP addresses of my running instances.
Q was able to provide some information, not the IP addresses but different IDs for running instances.
To go deeper, I chose a particular ID and asked Q to provide some details.
With Q, we can read some details in JSON for better readability.
Conclusion
Q seems like a great tool, but it does not understand better everything we ask, and due to compliance, some information cannot be shown!
It’s strong by providing some answers when it comes to AWS services, providing some details, and can make some mistakes all the time.
On a note of 5, I will give it a 3. It’s not perfect, and I will get better as soon as it is trained daily.
The generative AI age is now, and we don’t have to rely on the tools all the time!
