5 ms·

Why would Chia be the threat here? Filecoin(IPFS) is already a threat, it already has a capacity of 4.81 EiB of data, 4810 PetiBytes! (Is that how you spell it

Why would Chia be the threat here? Filecoin(IPFS) is already a threat, it already has a capacity of 4.81 EiB of data, 4810 PetiBytes! (Is that how you spell it? In contrast to Petabytes) And to store 1 GiB of data per year it only costs the user $0.00082 USD. That shows harddrives are already being gobbled up by miners at an incredible rate. Has this made a noticeable rise in storage prices? Anyone know of any data that can show that?

Dylan16807 5y ago
> And to store 1 GiB of data per year it only costs the user $0.00082 USD. How can filecoin possibly be that cheap? Providers like B2 and wasabi are 6-7 cents. Sia spent a while in the single digits is currently around 10. This is less than a tenth of a cent?
- wmf 5y ago
  I think Filecoin miners are currently being heavily subsidized by mining rewards so they don't need to make a profit on storing data. At those prices, I wonder if storing data is more trouble than it's worth.
john_minsk 5y ago
What is this space used for? Generally curious
KingMachiavelli 5y ago
Supposedly Filecoin is storing 'real' data, so it's not using any more storage than AWS or any cloud provider would. Is that not the case? At this point I wouldn't be surprised if the the number of 'real' data stored was only a fraction of the total network size. Chia is just storing 100% useless data outside of its use by Chia itself. i.e Filecoin's and similar coins should have value directly tied to their use to store data while Chia is just another crypto that has no 'intrinsic' value outside its use as a currency and computation platform.
MrFoof 5y ago
>Filecoin(IPFS) is already a threat, it already has a capacity of 4.81 EiB of data, 4810 PetiBytes! (Is that how you spell it? In contrast to Petabytes) 4.81EiB is ~5.54EB. Filecoin has been around for ~81.5 months. That's an average growth rate of only ~68PB/month, but we can of course assume that the current growth rate is higher. For example, last year there was a period where it grew 250PB in a month. Chiacoin has supposedly added ~200PB to its mainnet in the past 24 hours. That growth rate is also still climbing. Seagate has been quoted as having an current HDD manufacturing capacity of ~350EB per year. The general assumption is that between Seagate, Western Digital and Toshiba there's likely a total manufacturing capacity of close to 1000EB a year. Of that 1000EB, a large percentage goes to OEMs and SIs, another large percentage goes to direct customers (Alibaba, Amazon, Apple, Facebook, Google, Microsoft, etc.), and the remainder goes into the sales channel (retail, VARs, etc.). OEMs, SIs and direct customers have contracted supply and pricing, whereas the sales channels do not. Channel is likely 30-40% of current production volume. So that's 300-400EB a year of capacity across all manufacturers going into the sales channel where end users can buy from. -- ----- ----- ----- At 0.2EB/day of growth, that is 18-24% of all channel volume being consumed just by new Chia farm plots. At that rate, that'll cause channel supply issues very, very soon for certain. Once channel supply becomes exhausted, prices will rise since the demand is there. The issue is... what ends up happening to the sustained growth rate? If it were to reach something like 1.0EB/day in growth, that is basically all of the normal channel capacity being spoken for just by Chia farm plots. That's an actual disaster for end users. Manufacturers can't respond to those demand spikes instantly. They can respond some, for certain, but if some component hits maximum manufacturing capacity, you need to expand or build new factories (whether the manufacturers, or contract manufacturers within the supply chain for certain components). Building new factories is not quick. Not just construction, but ensuring that the demand will be sustained so you don't end up with these huge capital expenditure factories going idle when demand wanes.

Dylan16807 5y ago

> And to store 1 GiB of data per year it only costs the user $0.00082 USD. How can filecoin possibly be that cheap? Providers like B2 and wasabi are 6-7 cents. Sia spent a while in the single digits is currently around 10. This is less than a tenth of a cent?

wmf 5y ago
I think Filecoin miners are currently being heavily subsidized by mining rewards so they don't need to make a profit on storing data. At those prices, I wonder if storing data is more trouble than it's worth.

john_minsk 5y ago

What is this space used for? Generally curious

KingMachiavelli 5y ago

Supposedly Filecoin is storing 'real' data, so it's not using any more storage than AWS or any cloud provider would. Is that not the case? At this point I wouldn't be surprised if the the number of 'real' data stored was only a fraction of the total network size. Chia is just storing 100% useless data outside of its use by Chia itself. i.e Filecoin's and similar coins should have value directly tied to their use to store data while Chia is just another crypto that has no 'intrinsic' value outside its use as a currency and computation platform.

MrFoof 5y ago

>Filecoin(IPFS) is already a threat, it already has a capacity of 4.81 EiB of data, 4810 PetiBytes! (Is that how you spell it? In contrast to Petabytes) 4.81EiB is ~5.54EB. Filecoin has been around for ~81.5 months. That's an average growth rate of only ~68PB/month, but we can of course assume that the current growth rate is higher. For example, last year there was a period where it grew 250PB in a month. Chiacoin has supposedly added ~200PB to its mainnet in the past 24 hours. That growth rate is also still climbing. Seagate has been quoted as having an current HDD manufacturing capacity of ~350EB per year. The general assumption is that between Seagate, Western Digital and Toshiba there's likely a total manufacturing capacity of close to 1000EB a year. Of that 1000EB, a large percentage goes to OEMs and SIs, another large percentage goes to direct customers (Alibaba, Amazon, Apple, Facebook, Google, Microsoft, etc.), and the remainder goes into the sales channel (retail, VARs, etc.). OEMs, SIs and direct customers have contracted supply and pricing, whereas the sales channels do not. Channel is likely 30-40% of current production volume. So that's 300-400EB a year of capacity across all manufacturers going into the sales channel where end users can buy from. -- ----- ----- ----- At 0.2EB/day of growth, that is 18-24% of all channel volume being consumed just by new Chia farm plots. At that rate, that'll cause channel supply issues very, very soon for certain. Once channel supply becomes exhausted, prices will rise since the demand is there. The issue is... what ends up happening to the sustained growth rate? If it were to reach something like 1.0EB/day in growth, that is basically all of the normal channel capacity being spoken for just by Chia farm plots. That's an actual disaster for end users. Manufacturers can't respond to those demand spikes instantly. They can respond some, for certain, but if some component hits maximum manufacturing capacity, you need to expand or build new factories (whether the manufacturers, or contract manufacturers within the supply chain for certain components). Building new factories is not quick. Not just construction, but ensuring that the demand will be sustained so you don't end up with these huge capital expenditure factories going idle when demand wanes.