I have 8TB of disk and as per the storj dashboard its filling very slow therefore I thought to start another container with different port for more connections.
I want to know that is this think works with same node id which i am using with multiple containers on different storage locations.
One more thing I attach multiple drives or 8TB and without clustering them should I run multiple containers on different drives.
Don’t!
This may have worked on v2, but on v3 this will definitely not help you. IP filtering is already in use to ensure that multiple nodes on the same IP or IP range don’t get more transfers. Instead the same amount of traffic will be spread out over the nodes.
The only reason to add a second node would be if your first node is filling up, you have no more room on that HDD and you have another HDD to connect. But I wouldn’t even consider this until the first node is about 75% full.
When you do want to spin up the second node, you will need a new invite token, a new identity and a separate container using different ports. But you won’t have to worry about this for a while.
1 Like
OK one more thing, so if I attach another HDD and used different node ID for that hdd there will be no issues with IP filtering right.
You would be sharing the traffic with the other node. So the each node will get half the amount of traffic as the single node on the same IP.
If you set it up right there will be no issues, but there will be no advantage either unless you actually start having more data stored than the single node could handle. Hence why I say you should only do this when the first node is running out of space.
1 Like
Actually I am running storj on GCP instance and for every new machine I need to pay for the CPU RAM and Windows Licence, I have lightning fast internet speed. And the problem is Storage is much more cheeper on GCP then CPU and RAM as such I can even get Petabyte of storage but adding multiple virtual machines is quite expensive.
Therefore thought to attach multiple drives with multiple nodes on single virtual machine to increase profits. Now please answer my last question, that IP filtering mechanism don’t block multiple Node IDs on same machine with same IP address.
Thanks for the help !!
Well, I have my doubts about the profitability of hosting a machine on a competing platform that storj is trying to undercut in pricing. Somethings gotta give.
That said, this makes it very simple. There is absolutely no benefit of running more than one node on a single IP/VM. It will just add more CPU/RAM overhead and won’t increase your traffic or data stored. Just increase the size of a single node when it’s running out of space. There is no other setup possible that will perform better than that.
I understand what you’re trying to do, but storj is a decentralized platform and the IP filtering in place is actively fighting what you are trying to set up for a reason. It ensures that nodes aren’t hosted in large clusters on centralized data centers.
tl;dr > You can’t scale up like that on a single machine on purpose.
4 Likes