Sorry. I read The response wrong! Looking forward to testing and i am Loving The progress on ttl and garabage cleaning 
6 Likes
3 posts were merged into an existing topic: When will āUncollected Garbageā be deleted?
My Nodes where full (30TB) and now half of it went to trash. Iām ready for new data 
4 Likes
First Trash hast to be cleaned, then you are ready for now Data. I have to wait 5 more Days too 
4 Likes
I added another hdd to bridge the gap
5 Likes
In the meanwhile test data will be totally stopped?
1 Like
I have a question because there is a lot of trash.
There are 4 nodes on the server sharing one IP, and more than half of the capacity has been removed this time. When I receive ingress again, instead of all 4 receiving it equally, each one will receive 1/4 of the incoming amountā¦ right?
Then, since I think the ingress speed will decrease compared to when running a single node, I am considering reducing the capacity of 3 nodes until one node is full. Iām sure there are people like me who use multi-node. Can I know how you plan to operate it? 
each node will get 1/4 of max possible capacity, total will be 1, but because you have 4 nodes your nodes will handle it better than 1 node and win more races.
3 Likes
8 posts were split to a new topic: Is data uploaded/downloaded through a satellite?
Previously, there were about 400 TB of data on my nodes (out of about 500 TB allocated and another 300 TB ready to be added)
Now, after the not yet completed deletion of the basket, about 180 TB.
For all this time after the deletion and the beginning of the test data flow, I had about 200 TB of data.
I optimized my nodes as best I could and now there is no long queue, delay, base on the NVME ā¦
But if this trend continues, then there will be less and less sense in maintaining nodes, and there is no need to expand the volume at all, since more than 60% is now free.
Location Ukraine, perhaps due to the distance from the USA (those who are closer win your races) and the situation in the region, it has become unprofitable for me to maintain nodes given the need to maintain online in conditions of constant problems with the power supply. I really hope that the volume of comparison will begin to grow, and not just overload my Internet channels, which I also expanded to increase productivity (but without results)
2 Likes
The new node selection algorithm seems to prefer the closer nodes a lot and most customers are from the US. So if I ever wanted to run storj nodes as a business I would do it from a north america location. Also Ukraine is a country at war and some customers might exclude it for that reason.
I know of no request from any customer to exclude Ukraine nodes from storing their data, on the other hand, there have been several who want to exclude Russia nodes.
2 Likes
I understand what you are saying, but I also want to say that this algorithm, which does not have rebalancing, is focused exclusively on clients who, as we all understand, are in the US, will not allow the network to expand, and such dedicated operators as I cannot influence our nodes to increase in volume.
For my part, I provide uninterruptible power supply, optimize, but it all turns out to be pointless(
How about russian customers, if such exist?
If Ukraine is not excluded, then thank you.
We really have territorial problems with power supply and I canāt fully and constantly maintain all nodes in working order, but there are nodes that even in such a time of war work constantly but at the same time do not run into volume.
In any case, THANKS STORJ, but recently I have not seen filling in my own, there are deletions, there is a replenishment that is trying to keep the volume at half of what it was before, but deletions to the trash this week cleared significantly more than were replenished.
Which part of no request from ANY customer was unclear?
2 Likes
Please note that most of was was recently deleted should have been data your nodes were holding but not getting paid for, clearing space so you could receive more new paid data.
4 Likes
In theory yes. In practice that doesnāt seem to happen. I am running a pi5 node in Europe. That should be already 150ms disadvantage and my success rate is 98%-100%. It looks like there are way too many slow nodes to the point that your location doesnāt matter.
3 Likes
Is there any chance to share some info about the customer? I mean, for example their business type, or anything else which would not disclose business sensitive info. Just out of curiosity.
1 Like