You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As the term is wrapping up, one thing I would like us to do is deploy whatever we end up with.
Things that need to be deployed:
The database
The web server
The extension (We can get this onto the chrome store once it's ready, shouldn't be too much to worry about)
The database
Mongo Atlas:
Pros:
Well it's run by mongo...
All we need is modify the URI and we are done
We get $200 as students
Deploy a docker container containing the mongo database
Pros:
We can use any service we want (i.e Aws, GCP, Azure)
Most similar to our development environment
Cons:
Not sure about the cost... but we probably get free credits from those services. That said deploying docker containers can sometimes get expensive (well usually not, unless we wanna get fancy about it)
Use a cloud service that is similar to Mongo, i.e GCP's firestore
Pros:
We can choose one that works best with where we deploy the web server, for example, firestore credentials are managed automatically if we deploy on App Engine
Cons:
We will need to change code to work with that new service.
The web server:
"serverless", like AWS Lambdas and GCP's App Engine
Pros:
Pretty much the best option for pricing, it scales with usage, and we get a decent buffer where it's free
No need to worry about the internals of deployment, usually it's one .yml file and one CLI command and we are done
Cons:
Depending on where we deploy, we might need to slightly modify code to be the format the service expects it
AWS EC2, or virtual instances:
Pros:
We get full control over the environment if we need to do something fancy
Cons:
Might need a decent amount of setup and config to make sure service is up, DNS is working etc
Usually a flat fee, and we can choose to scale it manually (or automatically)
CD:
Whichever approach we take, we'll want to make sure it's deployed continuously. We can setup a Github Action for that, we can talk details once we come up with an approach.
The text was updated successfully, but these errors were encountered:
As the term is wrapping up, one thing I would like us to do is deploy whatever we end up with.
Things that need to be deployed:
The database
Mongo Atlas:
Deploy a docker container containing the mongo database
Use a cloud service that is similar to Mongo, i.e GCP's firestore
The web server:
"serverless", like AWS Lambdas and GCP's App Engine
.yml
file and one CLI command and we are doneAWS EC2, or virtual instances:
CD:
Whichever approach we take, we'll want to make sure it's deployed continuously. We can setup a Github Action for that, we can talk details once we come up with an approach.
The text was updated successfully, but these errors were encountered: