Sadly, I have been extremely occupied lately. And with little time for the crypto world to be honest. But, on top of the minimalist time I have reserved for it, I have been using it to mainly:
- Explore the news about current and emerging technologies
- Improve security of my own experiences
- Learn and gather new learning materials to teach or distribute to others
- Test new hardware and play with new software (crypto related)
About my node
To be honest it's been harder than I anticipated to maintain it under a NAS solution (XPenology). But it's been helping the network once and then. I always learn a new bit, throughout the process, but now I completely understand the advice some great witnesses posted, when they advised on a more step-by-step path. There is a reason for it...
I will keep it running, but only as long as it does not cause me huge costs. Since I am not a big known person, with enough support to call it profitable (I never mined a block), I am not very worried about my node "poor performance". But until the experience keeps stacking, I will keep the effort.
Which are my road-blocks?
Mainly less disk latency and a bit more of CPU power. The STEEM blockchain does not need many cores, instead, I would use more powerful pipelines that can sort out the "bursts" (this is my little 0.10$ experience). Having an 8-core low power Intel CPU, for example, does not help here. Although, do not interpret my comment as CPU being a problem... not at all! The IO latency is the major requirement for now (and that need cores)... secondly the memory requirements (about to change with the next version).
My mistake...
I have tried to increase the IO Bandwith (that I did), by growing the RAID from 6 disks to 10... but at the same time, the latency went up the pipe. In my particular case, using SSD caching is not working because of the solution (XPenology) not being smart enough, to distribute the IO (the IO manager is not great). On new OSes, like RHEL 7 or even Ubuntu, you can choose these and basically customize everything... Of course, you lose something, which is usually TIME!
Future of running NASes!
I think this will be something I will be looking soon. And it might involve NVMes, Nx 10Gbps links, plus wirelesses antennas, tones of Hard disk Spindles, and some awesome distributed filesystems I can't tell you about, of which are the TOP of the TOP stuff you can find around the world!
Then maybe, I will be able to run stuff like I imagine... All on the same Bone! BlackBoxBone! BBB! That's how I will call it, my next NAS.
Anyway, keep tuned...
RESTEEM if you wish to show others what you have learned... and,
VOTE for Witnesses
I am one of them, forykw!