[quote="hapsody, post:1, topic:5949, full:true"]
I try to build a service with IPFS. It’s a sort of data sharing market service.
I would like to make user to share their files on browser. so I’ve written some code with js-ipfs.
it works between browser to browser. once a browser upload some files, the other can download it
But the problem is speed. it was so slow (200MB, it took about 1 hour).
I wanna boost up this speed. so I’m gonna test with more ‘browser node’ which has same file to raise up the speed.
In this situation, Does have the each node same file at the same time?
If someone requests the file on ipfs,
in order to be shared faster, I think that all nodes should share fragmented file to one person at the same time.
On the other hands, as I know, IPFS is not allowed duplicated file.
In other words, Does it mean that IPFS has a single file even if several people share it? if so, isn’t it impossible to make faster?
which one is the fact?
I think there is confusion about what “deduplication” means.
It doesn’t mean that there is only one copy ever of a file circulating. Actually, there is no way to prevent someone from copying anything, any OS can dumbly copy byte per byte, even without understanding what it copies.
“Deduplication” only means that if 2 people independently upload the same file, IPFS will know that there are 2 copies of the exact same file. So, if someone asks for a copy of this file, IPFS can download it from the original uploader, or the second uploader, or both.
Another point is that IPFS doesn’t really “have” any file in a sense that the files live on the computers of the users, and never leave the hard drive unless someone requests it.
Actually, IPFS is “just” like a book-lovers social network for computers (the books being files) where they say:
- “Hey guy, I have a copy of this book (a file), if you want it, just ping me, and if I’m online I’ll send you a copy”.
- “Hey guys, I’m looking for this book? Do you have it? Or do you know who might have it?”