Bitcoin Cash Reveals Plans to Begin Sharding
Shammah Chancellor, one of BitcoinABC’s developers, revealed plans on Monday to initiate vertical scaling within BCH. They aim to do so by utilizing extra CPU cores to begin parallel processing. BitcoinABC is one of BCH’s primary clients.
The developer stated that the most efficient way to use the extra cores requires localized data processing. Sharding is the organization of the data for use in the local process. For the time being, Bitcoin utilizes structures of data to compute the Merkle code of the head block, which that prevents it from undergoing the localization process. If the root computation is changed through canonicalization, data can then begin to be sharded.
At the moment, blocks can be processed using only a single server. The lead developer at BitcoinABC, Amaury Sechet, stated that a few factors are able to undergo parallelization, however the order is still unable to be verified.
He said that they aim to process a sole block using multiple CPUs, perhaps on multiple machines. In this case, as the block expands, more machines can reduce, or maintain, the time needed to complete the processes.
Limitations on ordering is detrimental to the processes, so for the moment, it is unlikely that this will be possible. By removing these constraints, the problem may be solved, he added.
Canonicalization consists of the re-ordering of various transactions within Bitcoin Cash, and has been the topic of heated debate in recent months after Bitcoin Classic’s Tom Zander said that the strategy can be implemented without the need to change transaction ordering.
Sechet slammed Zander’s remarks by saying that he is incorrect, or at the very least, straining the truth. By withholding the algorithm, he believes Zander is playing around with the concept.
Tensions have mounted amongst the Bitcoin Cash community as debates persist over each proposal by ABC in anticipation of the upcoming hardfork in November.
The BCH proposal for sharding, dubbed simple or private sharding, will allow for more “power” through the addition of an increased number of processing units.
The most challenging aspect is the actual sharding of the data nodes, as only ETH are attempting to produce “full” sharding.
Scalability continues to be an issue with many aspects, and various ways in which it can be handled. Each gain will assist, however Sechet is not sure when it can be implemented if the structures currently being used will not be able to accommodate it.
Sorting out transactions into a range (canonical ordering) can allow machines to be assigned specific sets of calculations they can compute. As a result, by combining multiple machines with various capabilities per machine, when combined the process can become simpler and more efficient.
Sechet stated that this innovation has been on BitcoinABC’s mind since the very beginning, however it is yet to be put into practice. It is still unclear when the process will be initiated for the first time.