81–90 of 220 posts
For those who don't want to wait, there's DigitalOcean Spaces ( https://www.digitalocean.com/products/spaces/ ). Disclaimer: I haven't used it, but planning to, since I already use their VPS.
Really interesting, will R2 support lifecycle rules like S3 does? We write around 90 million files per month to S3, if we could replace that with R2 and have the files automatically expire after 30 days that'd be a pretty amazing price reduction for us.
Hey, I'm the PM for R2. We support per-object TTLs, so this should work!
What's the censorship policy? Is this going to be content-neutral, like Cloudflare was when fronting ISIS websites? Or is this going to be fine-until-bad-PR, like when Cloudflare decided to stop hosting The Daily Stormer? There is a special kind of lock-in when it comes to object storage, as generally you use something like this when the data is too big to store another copy of locally or at another provider. It's no…
You raise an interesting point regarding object storage being a relatively unique type of lock-in.
Feature request: automatic ipfs integration.
At Taloflow (https://www.taloflow.ai), (disclosure: I'm the CEO/Cofounder) we provide buying insights for cloud object storage (and soon other IaaS/PaaS). We will definitely be adding Cloudflare R2 to the mix.
Strategy Letter V, commoditize your competitor's advantages!
> We’ve gotten rid of complex, manual tiering policies in favor of what developers have always wanted out of object storage: limitless scale at the lowest possible cost.
Cloudflare has a clear strategy: Be the simplest cloud platform to deploy to. It has been a breeze as a small dev shop adopting their tech. AWS started with the startups, but have since long struggled to keep up that simplicity in face of supporting what must be a dizzying array of customer requirements. Remains to be seen how Cloudflare fares in that regard. I like my Golang better than Rust.
> Cloudflare R2 will include automatic migration from other S3-compatible cloud storage services. Migrations are designed to be dead simple.
Taking a leaf out of Amazon Data Migration Service and its free transfers from elsewhere into RedShift/RDS/Aurora/OpenSearch. Niice.
> ...we designed R2 for data durability and resilience at its core. R2 will provide 99.999999999% (eleven 9’s) of annual durability, which describes the likelihood of data loss... R2 is designed with redundancy across a large number of regions for reliability.
S3 goes upto 16 9s with cross-region replication... and so wondering why R2's still at 11 9s? May be the mutli-region tiering is just acceleration (ala S3 Accelerated Buckets) and not replication?
> ...bind a Worker to a specific bucket, dynamically transforming objects as they are written to or read from storage buckets.
This is huge, if we could open objects in append-mode. Something that's expensive to do in S3 (download -> append -> upload) even after all these years.
> For example, streaming data from a large number of IoT devices becomes a breeze with R2. Starting with a Worker to transform and manipulate the data, R2 can ingest large volumes of sensor data and store it at low cost.
Okay, where do I sign-up?
> R2 is currently under development...
Oh.
There is one major reason S3 remains the king of storage for mobile media uploads: bucket notifications. Does R2 implement this feature? If so, I’m going to have to run some experiments with this...
Hey, PM for R2 here! We're fully integrated with Workers, so you can write a Worker that calls additional logic when a request is made to the bucket's url. We have notifications, where a change to the bucket invokes a worker with a specific event, on our roadmap, but have a bunch of other work prioritized ahead of them.