As a part of our Enterprise Hub plan, we recently released support for Storage Regions.
Regions let you select where your org’s models and datasets can be stored. This has two predominant advantages, which we’ll briefly go over on this blog post:
- Regulatory and legal compliance, and more generally, higher digital sovereignty
- Performance (improved download and upload speeds and latency)
Currently we support the next regions:
- US 🇺🇸
- EU 🇪🇺
- coming soon: Asia-Pacific 🌏
But first, let’s examine easy methods to setup this feature in your organization’s settings 🔥
Org settings
In case your organization isn’t an Enterprise Hub org yet, you will notice the next screen:
As soon as you subscribe, you’ll have the option to see the Regions settings page:
On that page you possibly can see:
- an audit of where your orgs’ repos are currently positioned
- dropdowns to pick where your repos can be created
Repository Tag
Any repo (model or dataset) stored in a non-default location will display its Region directly as a tag. That way your organization’s members can see at a look where repos are positioned.
Regulatory and legal compliance
In lots of regulated industries, you will have a requirement to store your data in a selected area.
For firms within the EU, meaning you should utilize the Hub to construct ML in a GDPR compliant way: with datasets, models and inference endpoints all stored inside EU data centers.
In the event you are an Enterprise Hub customer and have further questions on this, please get in contact!
Performance
Storing your models or your datasets closer to your team and infrastructure also means significantly improved performance, for each uploads and downloads.
This makes a giant difference considering model weights and dataset files are often very large.
For example, if you happen to are positioned in Europe and store your repositories within the EU region, you possibly can expect to see ~4-5x faster upload and download speeds vs. in the event that they were stored within the US.
