Using Amazon S3 to Store Data for a TOM¶
If a TOM needs to store a large amount of data, like images or spectra, it may eventually become impractical to do so on a local hard drive or network share. This is where cloud storage services like Amazon S3 come in handy. These services allow you to store large quantities of data at a low cost, while providing high reliability and feature rich services. In most cases using a cloud storage system also provides performance and speed increases to your application.
Configuring the TOM toolkit to store data on Amazon S3 is fairly straightforward. Once enabled, data product downloads, uploads, and static assets (images, stylesheets, etc) will be stored in Amazon S3 instead of the local filesystem where your TOM is run.
Sign up for an AWS Account¶
To use S3, you’ll first need to sign up for an Amazon Web Services account. New accounts get access to one year of free tier access which includes a year of S3 at a max of 5GB. If you’re interested in the cost beyond 5Gb, try out the Amazon cost calculator.
Once you have created an account, you’ll need your access key id and secret access key. These can be found under your profile settings -> “My security credentials”. Make sure you save these in a safe place, you’ll need them later.
Create a bucket¶
A bucket is like the highest level folder you can store data in S3. You should create one for your TOM. Name it whatever you’d like. Most of the default settings should be fine.
<?xml version="1.0" encoding="UTF-8"?> <CORSConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/"> <CORSRule> <AllowedOrigin>*</AllowedOrigin> <AllowedMethod>GET</AllowedMethod> <AllowedHeader>*</AllowedHeader> </CORSRule> </CORSConfiguration>
This policy allows GET requests from anywhere. Feel free to edit it to match your particular use case specifically.
Configure S3 Storage backend for your TOM¶
Now that we have a bucket set up, let’s configure our TOM to use it. First we need
to install two additional python packages. You should add these to your project’s
Next, we’ll edit our TOM’s
settings.py to use S3 instead of local storage. Place
the following lines somewhere around the existing static files configuration
DEFAULT_FILE_STORAGE = 'storages.backends.s3boto3.S3Boto3Storage' STATICFILES_STORAGE = 'storages.backends.s3boto3.S3Boto3Storage' AWS_ACCESS_KEY_ID = os.getenv('AWS_ACCESS_KEY_ID', '') AWS_SECRET_ACCESS_KEY = os.getenv('AWS_SECRECT_ACCESS_KEY', '') AWS_STORAGE_BUCKET_NAME = os.getenv('AWS_STORAGE_BUCKET_NAME', '') AWS_S3_REGION_NAME = os.getenv('AWS_S3_REGION_NAME', '') AWS_DEFAULT_ACL = None
Notice that these settings get their values via environmental variables. Depending on how you deploy your TOM, you can set these in a variety of ways. For example: export AWS_ACCESS_KEY_ID=MyAccessKey would be one way to set them using Bash.
AWS_ACCESS_KEY_ID is your access key id from your security credentials.
AWS_SECRECT_ACCESS_KEY is your secret access key from your security credentials.
AWS_STORAGE_BUCKET_NAME is the name you gave to the bucket you created.
AWS_S3_REGION_NAME is the name of th region you created your bucket in.
Once these settings are filled out, your TOM should store all future data in S3. If you had existing data in your TOM, you should copy it over to your bucket in the exact same way it was stored locally.
For Heroku Users¶
If you are using Heroku (perhaps by following the Heroku deployment
guide) there is one more
additional step. At the very bottom of
settings.py change the line:
This instructs the
django-heroku package to not automatically configure static
files for your TOM (since we are explicitly using S3 now).
Additionally, Heroku makes it easy to set environmental variables. See Configuration and Config Vars.