Billy mills
Lawn
last online: 02/18, 4:01
Verified User (6 years, 6 months)
Long Term User
Shoutout0

How difficult is it to build a back-end rest API with Amazon Web Services?

I have experience with front-end but not much back-end.

I have a client that wants me to put together an estimate for a project and I'm not sure how long it will take. It needs to be able to capture logging data from about 60 different locations into a central database in the cloud. Not much traffic compared to a regular website, but lots of data.

Then I need to build an admin website where a small number of users can log in securely and view/filter/sort the data or view graphs. Also need ability to create new users and reset username/password. I can handle the front-end part of it, but I'm not sure about the backend API and the database. Also not sure about ongoing costs.

I found a really great tutorial here that I hope I can use to figure out a lot of it:
https://serverless-stack.com

Not sure if the DynamoDB database is right for the job though. Any input is helpful. Thanks.

This open post was written |
Views: 19, Subscribers: 4 |
Leave a reply | Report Post

⇩ Zoom to bottom
Reciprocity (0)
Reciprocity
Since writing this post Lawn may have helped people, but has not within the last four (4) days.
Post Tags (5)
data, users, front-, api, database
Replies (5)
1581744157174 1581744149313 miss bot
last online: 03/19, 3:49
Verified User (6 years, 6 months)
Long Term User
Shoutout0
#
(2 days after post)
Quote this reply Report this reply to moderators

You lost me at "How"...

Dr. ralph club zps9ornptsl
(2 days after post)
Quote this reply Report this reply to moderators

Rockster could probably help you, he made this site.

Roccoflip
(2 days after post)
Quote this reply Report this reply to moderators

Fairly simple to set up. AWS is great for building scaleable solutions, though might not be the cheapest option if you only need something small and/or of set size/traffic.

Storing data is dead easy. You can use S3 to store everything and dump all of your data in- the hard part is getting it back out for use/aggregation- and that depends on how you need to use that data. I wouldn't recommend going serverless, simply because serverless makes it difficult to "serve" the data. You'd have to do all of the data filtering and aggregation on the front end, which means you first have to collect all of the raw data, store it in client-side memory, and then run functions on it there. It works, but it's slow.

I'm a full-stack dev, but always use a server. Serverless doesn't really make sense to me for a number of reasons, especially along the lines of user auth. However, setting up user auth is generally extremely easy as there are frameworks that provide the whole process for you. It's also possible to run your users through a dedicated server and store/serve your data simply through S3 without it touching the server's database at all.

DynamoDB is a no-sql service, which to me, falls under the "serverless" category. You'll generally experience significantly longer load times than you would with a database where you can query exactly what you need. That said- if you don't have backend experience and you're just trying to get something up and running, I think that's a great start.

1581744157174 1581744149313 miss bot
last online: 03/19, 3:49
Verified User (6 years, 6 months)
Long Term User
Shoutout0
#
(3 days after post)
Quote this reply Report this reply to moderators

What he said ^^^^^^^^^!

Billy mills
last online: 02/18, 4:01
Verified User (6 years, 6 months)
Long Term User
Shoutout0
#
(3 days after post)
Quote this reply Report this reply to moderators

Thanks for the great info. Ok, I think DynamoDB will be too slow since there is going to be more and more data in the database over time. Would definitely want to query by date range and other criteria so we are not having to load massive chunk of data at once.

A
⇧ Zoom to top

Help-QA supports basic Markdown, emoji ๐Ÿ˜, and tagging friends with @username!