Expanding on existing nodes

Storage Node Operators
,
#1

HI guys, Recently I acquired a 12tb HDD. In addition I bought a super power efficient rack mount server, The plan is to populate it with 12tb hdd and another one a bit later. The beauty of Datacenter is that they area able to provide me 4 different public IP addresses. So the plan is to create 4x VM`s that will be veted at the same time and also each will have approximately 3tb of storage space.

The idea is that once I get Additional HDD I will just add it to the server and half of the VMs will get migrated over to it and all of the VMs will gat an Storage upgrade.

So my question is IS it ok to increase the size of storage available to storj after creating will the network use the newel available storage space too or will it be capped by the maximum that was created at the setup.

PS. I know about the config and that i can change its a question for those that have done it.

#2

It is no problem to change the size any time.
You can even reduce it, but it takes a lot of time until the data really gets deleted, as this depends on real user deletions.

1 Like
#3

I would not recommend to use a one VM per node. The one node per HDD is ok, but you can run all of them in one VM or bare-metal.
Also, please, calculate your potential earnings and your costs for VPS. It could be not worth it

#4

For the cost and profit analysis I found out I will be able to cover the running costs in 5 months because the providers of the datacenter are actually really cost efficient compared to lots others i found out. Additionally it is located on the just above the Primary Prague Internet Exchange, so quite high bandwidth for the price payed.

I want to run them in VMs as In future once the nodes outgrow the initial case I would get another and migrate then over so they can grow. I would like to grow my nodes to a final capacity of around 20-25tb, I know that will take a long time but I’m committed, I’m willing to support the initial node financially for the first 12months and if after 12 months I found out the running costs are covered I will continue, or f the VMs Get full I will expand on the available storage space

#5

That’s interesting only if they are in different /24 subnets.

And what is this mystery hosting service that supposedly have very low prices? :slight_smile:

#6

yes on different subnets

#7

Yep, definitely agree with jammerdan. I have changed various nodes both ways, a few times. First, with increasing node size, I’ve migrated my first node a couple of times to larger and larger HDDs. I started the node on a 4TB HDD, then migrated it to a 10TB, and then subsequently to a 12TB HDD. No issues with increasing available space.

Then just recently I’ve been playing around with my three nodes to see how traffic changes or node performance changes (if at all) when I lower the available space below the actual space used. So in the graph below you can see the ingress over the past two days, where I did the following:

  1. lowered available space on Node 1 to simulate it being full by only about 100-500MB and you can see ingress stopped on that node for a period of time, then in between #1 and #2, ingress started again on Node 1 for a short period and then stopped again, because there must have been enough deletes or trash emptied to free up space below the threshold, but then after it reached the top limit again it stopped.
  2. Set back the available space on Node 1 and ingress started back up again.
  3. lowered available space on Nodes 1 and 2 to again simulate being full but this time by a few GBs. I wanted to see how Node 3 performed handling all of the ingress traffic since that node is running on a slightly different machine setup/configuration. Looks like it did alright.
  4. Set back the available space on Nodes 1 and 2.

image
image1794×655 175 KB

1 Like
#8

I must say this is very interesting