On the bitcoin reddit today, there was a discussion of storing exabytes of data on the network and then reading the data back from multiple accounts as an attack on maidsafe. The idea being that with that much data, you would overrun the caching and therefore the GETs would actually go all the way to the vaults, allowing the attacker to earn tons of safecoin. Also, the fact that you’re storing exabytes of data means there’s a good chance vaults you control will have some of that data.
Of course, the attacker needs to pay to store such a large amount of data. If the cost per gigabyte were never allowed to dip below some threshold, such attacks would be prohibitively expensive.
The number of exabytes that would be required for such an attack would vary with the size of the network. The larger the network, the more exabytes of data an attacker would need to store. Therefore, the cost/GB of storage needs to be kept higher for the network when it’s small, and only allowed to decrease as the network grows, so that the cost to attack the network is always kept in the “too expensive” category. This would probably happen naturally anyway, but if it doesn’t, perhaps an override function could limit how low the cost could go, based on network size.
Does any of this make sense? Or is this attack not possible anyway?