You are viewing a single comment's thread from:
RE: Introduction to VIVA, part 5 : VIVA CAN? You bet!
Interesting concept and I shall take my time digesting from part 1 ...I missed the first 4 posts. Do you perhaps know which S3 availability zone or region was down?
News was saying us-east 1 and us-west 1 depending on the source.
Thanks, I found it on Techcrunch. We have our Datalake in AWS ... but not in the US....thankfully this time.
You should consider moving it to VIVA when we launch. You don't pay for storage, just retrieval and it gets globally sharded and distributed.
It's 2.3 petabytes and grows in excess of 10 terabytes a day .... would this be feasible? I have not yet read about VIVA except today's post.
Yes it's designed to handle a use case exactly like that. The CAN as internet content cache is just an example of one way to use it.
But the thing is, the entire VIVA data store is using an index size of 256 bits, so a petabyte is really nothing, we're not running out of space there.
Now, the real question is do we have the capacity and can it be provided cheaply and reliably.
The MedicAxess data store which will run on top VIVA, must store any type of medical modality on demand and has retrieval guarantees of 30 mins or less for aged data and 5 mins or less for newer data.
With MedicAxess you're talking about the complete medical records of the entire nation of Mexico, (MedicAxess is a joint venture between Imaxess, IMSS, & VIVA), a nation that is quickly shifting to digital records across the board including 4k live streams of surgery and other telemedicine applications. All of which must be constantly archived. We projected 1 PB per month growth on that.
So 310 TB per month shouldn't be an issue. It will mostly depend on how many people contribute storage resources to the network, which will mostly be a function of how well we get our message out. But network wise, yes we can most definitely handle the capacity.
But I'll bet we can reduce your storage expenses an order of magnitude or more. Here's how this would work.
I just checked S3 pricing, looks like it's $24 per TB / per mo, not counting access charges. With VIVA it would be about $5.50 per TB per mo and that includes access charges.
That pricing assumes you're a Crown Holder, meaning you bought VIVA Crowns during one of our auctions, or during the ICO.
If you're not a Crown Holder, you'd need to buy VIVA on the open market to pay to access the data stored there, and the math on the back of my napkin here is saying that would be around $15/TB per mo.
Keep in mind, you only pay for retrieval of data, never storage. Your data lives on the network indefinitely, but the longer you go without accessing it, the more expensive it becomes to access.
I would love to discuss this with you if you would like to go to https://chat.vivaco.in/ sometime tomorrow, I should be there all day.