Happy Friday @everyone, please forgive my delay in sharing our team weekly update with you - I’ve been deep in ‘business’ development, and the guys have been deep in improvements.
Jumping in then …
nodes filling up - a good thing? No.
We need data to flow and spread happily across the archive, and not get stuck because nodes are reaching capacity - noting here that, by design, the price of uploads rises as nodes fill, rewarding those storing and encouraging more nodes to come online and earn.
The archive is for sure taking on more data now. From the Foundation-run node fleet we can see 11.37TB stored on disk, up from 7.21TB - an increase of 57.8% within a week. I would caveat: that is raw on-disk data, so it includes replication and is not the same as unique user data uploaded. Even with that caveat, the storage need is clearly rising. 
We have a few strategies and routes to look at on capacity and node performance. The immediate step we can take is to ask all supporters to please, please look at the size of your nodes.
Lots of very small node-services fill quickly, so it is better for archive performance to run fewer, larger nodes - ideally with a capacity of 20GB at a minimum. This should also help keep your nodes in good standing as we tighten health and shunning rules for nodes that are too small to keep up.
Pruning helps here too: once a node has verified that data is safely stored elsewhere, it can drop its own copy and free up space. Connected to this, we’re working on an update that clears the timeouts which were stopping nodes from releasing data they no longer need (this will also help earning).
Staying with our node theme - we have been diving deep on close-group agreement: making sure clients and nodes consistently agree where data should live, even as the network changes, fills, and distrusts unhealthy peers. Worth flagging for the record that this is connected to performance (data flow and node capacity), not permanence.
In regards to permanence in this start-up small scale phase, the Foundation will deploy nodes and/or increase node sizes to address any potential capacity issues - fine as a backstop for now, but we want to get to the root, not just tinker at the surface. Reliable and boring is what we’re aiming for.
In other news - the Autonomi Chrome extension is now live. The public store listing can take a week or two to appear, but you can access it directly now via the link below. We’d all be very grateful if you’d be kind enough to install it, as it also helps with the listing process:
I look forward to giving you all an ‘on’ time and more comprehensive update next week, but do please ping us in the general chat or support channel if you have any questions or need any clarification 
Thanks so much 
Bux
