DOH! Speaking of, I guess I need to edit this post. Should we journey to 2 PB or should we add a zero and make it a real journey?!?
I guess you could start a new topic for the 2PiB milestone.
On the side note, I’m also aiming for 1PiB. Currently at ~6900 plots. Long way to go. It took me about 5 months plotting to get here.
Currently using consumer hardware for plotting up to 75/day. Not sure I can scale beyond 1PiB without Bladebit. Just can’t find any auction deal on a server, and I’d rather buy HDDs than a plotter at eBay prices
I think that’s a great idea but I’m gonna shoot for the stars and go for 10pb. I’ve already got over 2pb in storage, just need to fill it. With a full blade chassis, 1000 plots a day seems easily attainable. That’s ~30,000 plots/month so about 4-5 months at that rate would do it.
I’m taking delivery on all those blade servers tomorrow and still need to get some more chassis, but depending on what I find maybe we can come up with a deal, especially if we can skip ebay fees and maybe even transact in XCH? 
1 Like
That’s some industrial scale stuff! Good luck.
And sure, feel free to DM if you want to discuss a deal
1 Like
What do you know that we don’t?! 
You have all that fast plotting hardware… let’s test this. Create 5 plots with a new set of keys and put them in a pool. PM me the seed phrase, and I’ll try to move them to a pool that pays out to me. I’m interested to see if it works this way also!
Edit - Hell, I’m willing to create the plots and set it up I’d someone else will test this with me.
Probably not much! Obviously, I’m disappointed with the market right now but so is everybody else, and I’ve never made any money doing exactly what everybody else is doing. I was very conservative during the “gold rush” back in May and took that time to learn and study the code and see how things turn out. As a developer, I’m impressed with the project and the goals and the progress. I’m learning a huge amount of knowledge about hardware when most of my career has all been software, so my skillset is improving. I truly believe the future of the cloud is decentralized data-centers-in-a-box running in millions of garages instead of at AWS or Azure. This is a huge head-start on that future. Worse case, I’m trading fiat cash for refurbished/auctioned hardware at pennies on the dollar in the hopes that it retains its value as inflation continues to nibble away at that cash. And since I’m self-employed, I either need to spend my earnings on company equipment or the government will spend my money for me at tax time. 
If Chia moons, that’s a bonus!!
You might check out this thread where @xkredr59 has already done this testing - very interesting!
1 Like
Congratulations 
Who is WE? You and the blades? Did you name them already? Talking to them like people talk to their pets? Careful now 
2 Likes
Congrats!
Next Goal 10PiB? 
1 Like
Where’s your victory speech?
''I’d like to thank all my blades for their combined effort getting us here, as a team! thank you Jack most specifically you for plotting through the night solo from day one before you were joined by your siblings.
Also like to thank the local power company for keeping the lights on consecutively since the last outtage, without you guys we would be in the dark ages eating squirrels by the fire!
And finally like to thank all the Naysayyers out there who said I couldn’t achieve this, You gave me my drive, you gave me my persistence, you were like an absent dad that Drove Steve Jobs to new undiscovered peeks, I thank you all for making me whole!
Now off to eat a bowl of…none other than Chia!’’
2 Likes
Congratulations! To the next 9 Pb!
When does the “Estimated Time to Win ?” change to TODAY??
1 Like
When you are HPOOL, of course.
Looks like you and I hit 1 PiB about the same time! Kudos. Now on to 2…
I ended up with 10 Intel NUCs, one farmer and nine (now 7) harvester/plotters, and topped out around 200 plots/day. After the first electric bill, I slowed down a bit, especially when the network leveled off around ~33 EiB. Storage similar to yours, though I still have some smaller drives from my humble start on top of some 24-drive servers.
Some thoughts you may appreciate:
- Most if not all of my wins came during a pause in my plotting.
- In my experience, plots need to mature for a while before they can rack up a win. Proof of time stuff.
- After laying so many plots so quickly, I am only just now a little positive on luck.
I had gone over a month dry, and decided to stop plotting for a week to see what happened…starting Christmas Day I won three times inside of a week. A different win happened when the harvesters were down for a week, think I had 700 plots at the time.
So it’s a long-haul message of vision and patience…
5 Likes
Nice! Congrats to you too, it is definitely a big achievement!
It sounds like you were plotting on your harvesters? Definitely don’t recommend that - the disk I/O and proc usage might be what was causing your unlucky streak if the harvesters couldn’t get the proofs to the farmer in time…Especially if using BladeBit, which absolutely totally saturates disk I/O so much that it is hard to get another plot copied off the disk while it is writing, much less also harvesting!
I’ve got my workflow up to 400 plots a day now after some BladeBit tuning and some scripting for simultaneous network file copies to different destination disks on the farmer. I’ve got 4 blade servers each with dual 14 core procs and 448gb RAM and they can each pump out a plot every 15 minutes (including network file copy over 20gb/s network). I’ve still got one farmer controlling all disks and even with lots of file copies coming in over the network from those 4 plotters, my lookup times are all less than 2 seconds. I’ve not yet had an “unlucky streak” - my block wins are still happening every 6-7 days as expected. I’m only up to about 1.1PB now cause I spent a lot of time tweaking the workflow over the holiday, but I expect to be at 2PB by end of January or sooner!
2 Likes
Very interesting to hear about the pause of plotting on your harvesters and your winnings happening.
I did think performance would matter over time and which is why I question those who become frustrated with raspberry Pi winnings and then end up quitting. Is this a thing or am I looking too much into this @enderTown @BigSilverOrb ?
2 Likes
I definitely think that could be a possibility - in my humble opinion, it doesn’t make much sense to use a Pi beyond maybe a few disks. I get that it is possible and it is the most power-efficient setup, but I’d rather have more power available to arguably the most important part of the whole thing - the farmer/harvesters! Besides, once you scale past 20 or 30 disks, the power usage of your farmer becomes less and less important relative to your watt per tb of all your drives. We’ve already seen that more powerful procs dealt with the dust storms better, so having that extra power in reserve is better for me and my setup anyway. I’m starting to get to the point where downtime is costing me real “money” 
5 Likes