[bitcoin-dev] Scaling by Partitioning
loi.luuthe at gmail.com
Wed Dec 9 06:30:24 UTC 2015
Its Loi Luu, one of the authors of the SCP protocol (
Before SCP, we had been thinking hard about how to do sharding efficiently
without degrading any security guarantee. A simple solution which splits
the coins, or TXs in to several partitions will just not work. You have to
answer more questions to have a good solutions. For example, I wonder in
your proposal, if a transaction spends a "coin" that ends in "1" and
creates a new coin that ends in "1", which partition should process the
transaction? What is the prior data needed to validate that kind of TXs?
The problem with other proposals, and probably yours as well, that we see
is that the amount of data that you need to broadcast immediately to the
network increases linearly with the number of TXs that the network can
process. Thus, sharding does not bring any advantage than simply using
other techniques to publish more blocks in one epoch (like Bitcoin-NG,
Ghost). The whole point of using sharding/ partition is to localize
the bandwidth used, and only broadcast only a minimal data to the network.
Clearly we are able to localize the bandwidth used with our SCP protocol.
The cost is that now recipients need to themselves verify whether a
transaction is double spending. However, we think that it is a reasonable
tradeoff, given the potential scalability that SCP can provides.
On Wed, Dec 9, 2015 at 12:27 AM, Akiva Lichtner via bitcoin-dev <
bitcoin-dev at lists.linuxfoundation.org> wrote:
> I am seeking some expert feedback on an idea for scaling Bitcoin. As a
> brief introduction: I work in the payment industry and I have twenty years'
> experience in development. I have some experience with process groups and
> ordering protocols too. I think I understand Satoshi's paper but I admit I
> have not read the source code.
> The idea is to run more than one simultaneous chain, each chain defeating
> double spending on only part of the coin. The coin would be partitioned by
> radix (or modulus, not sure what to call it.) For example in order to
> multiply throughput by a factor of ten you could run ten parallel chains,
> one would work on coin that ends in "0", one on coin that ends in "1", and
> so on up to "9".
> The number of chains could increase automatically over time based on the
> moving average of transaction volume.
> Blocks would have to contain the number of the partition they belong to,
> and miners would have to round-robin through partitions so that an attacker
> would not have an unfair advantage working on just one partition.
> I don't think there is much impact to miners, but clients would have to
> send more than one message in order to spend money. Client messages will
> need to enumerate coin using some sort of compression, to save space. This
> seems okay to me since often in computing client software does have to
> break things up in equal parts (e.g. memory pages, file system blocks,) and
> the client software could hide the details.
> Best wishes for continued success to the project.
> P.S. I found a funny anagram for SATOSHI NAKAMOTO: "NSA IS OOOK AT MATH"
> bitcoin-dev mailing list
> bitcoin-dev at lists.linuxfoundation.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the bitcoin-dev