the end user would put the bootstrap into their datadir, start the client, the client processes the blocks, you save a shit ton of time vs syncing a virgin client. it is not the fastest thing in the world, but i would guess of the time savings at the very least 75%.
also, super sexy pi client just cooked for distro from sluppy can be found on the
release page check it out.
So you put it in the same directory as the Blockchain? If you have an existing Blockchain do you delete that? What is the reason of renaming it bootstrap.dat? Would it just not be easier to replace the Blockchain with another one and not have to rename it?
Congrats on the new pi client!!!Yes, you put it in (in this case) .hyperstake (or the windows/Mac equivalent) and then delete everything but wallet.dat, hyperstake.conf and bootstrap.dat
The reason for renaming it is that the core code looks for bootstrap.dat to import a static blockchain, and it imports the whole freaking thing before indexing it. That saves a shitload of time. The reason for not just copying the whole .hyperstake (or whatever coin) data directory is that the data may be corrupt, have transactions only specific to your wallet, etc. And of course, the size. Everything that is needed for the client to extract all that information is in the blk000* files. There are several of them as the blockchain grows, so if you're making a bootstrap AND there is more than one, you need to combine them in serial order. It's easy to do, and it will save you endless headaches when synching a new install or recovering a messed up wallet install.
Another thing I do, and highly suggest if you're not running on a Pi or something tiny like that is add txindex=1 to your configuration file. This pulls ALL the blockchain data, like a block explorer. so you can pull up the data from any valid block in the chain in your own client, not just the ones you had a transaction in