Graphene chain good choice for large scale web hosting?

[xiahui135]

this is what Storj 、Sia、Maidsafe doing.  They have been go much further than Bitshares in this direction.

[seocuenta]

I agree with you xerox, I do not think it will be hard to find a link between them

[xeroc]

Isn't that the whole point of Bitshares 2.0? A P2P backend for front end "exchanges"? The "exchange" would basically just be a front end.

This.

The witness application is essentially a full-node and may be allowed to sign blocks ...
ANY wallet would connect to a witness node eventually ..

[BunkerChainLabs-DataSecurityNode]

This has been done for a really long time already with block level syncing techs.

One of the more popular and simpler tools you might consider is DRBD

This happens at the block level of storage which ensures whatever method of data storage being used is compatible.

It's not decentralization. it's just replication.

[luckybit]

Isn't that the whole point of Bitshares 2.0? A P2P backend for front end "exchanges"? The "exchange" would basically just be a front end.

[speedy]

Here's an interesting idea Ive been pondering:

Large scale web applications typically have a web server that receives requests and does load balancing to the backend application servers. Requests are handled by as many backend servers as are needed to satisfy user demand.

There is the constant problem of how to sync up data across the servers, especially when one user's request can effect the validity/outcome of another user's request (such as an exchange website). So are there advantages to doing this using a private Graphene blockchain? Or does the extra redundancy of a blockchain make this inefficient relative to standard databases, with no real benefits?

It seems that running a p2p network as the backend of a website would be ideal in terms of taking servers down for maintenance, which is essential.