Scalability in Private and Public Blockchain Systems

In many of my posts, we have greatly discussed the blockchain systems potential for a number of industries. We also discussed factors like cost, privacy, data quality and an elimination of the need of third party trusted intermediaries. In this post, which is a continuation in a blockchain article series on[1], we will focus the factor of scalability both for private and public blockchain systems.

Since last year, the idea of private blockchains is rising whenever there is a discussion regarding blockchain technology. Essentially, rather than having an uncontrolled and fully public state machine and network secured by the cryptocurrencies, it is also quite possible to develop a system where access is more strictly controlled, with rights to change or even read the blockchain state limited to a few users, while maintaining a number of partial guarantees of decentralization and authenticity that is provided by blockchain[2]. These systems are the major focus of internet from the financial institutions, and up to some extent, have led to the backlash from those seeing such developments as either compromising the entire point of decentralization or being a hopeless act of middlemen attempting to be relevant.

A public blockchain is the one that anyone in the world can have an access to, send transactions and see them as a valid added component. Anyone can participate in this process, which is the process of identifying which blocks are added to the chain and what the present state is. Being an alternative to quasi-centralized or centralized trust, the public blockchains are secured by the cryptoeconomics, the combination of cryptographic verification and economic incentives using mechanisms like proof of stake or proof of work, following a common principle that the level to which anyone can have an influence in the process is directly proportional to the amount of economic resources that they can take to bear[3]. These blockchains are usually though as fully decentralized.

The problem of blockchain scalability is often seen in a number of discussions regarding the bitcoin protocol. Some important questions seem to arise out of such discussions such as will the needs of recording each bitcoin transaction in the blockchain sacrifice its security (since only fewer users will have a copy of the entire blockchain) or its capability to manage many transactions (since new blocks on which the transactions are to be recorded are just generated at limited intervals?

There are three major tentative blocks to the blockchain scalability, which are[4]:

1.     The tendency for centralization with an increasing blockchain: the bigger the blockchain grows, the more requirements arise for bandwidth, storage and computational power that must be consumed by full nodes in the network, which leads to the risk of much higher centralization if the blockchain turns out to be big enough that just a few nodes are able to process the block.

2.     The bitcoin oriented issue that blockchain has an inherent hard limit of one megabyte per block (around 10 minutes), and eliminating this limit needs a hard fork, i.e. backward-incompatible change) to the bitcoin protocol.

3.     The high processing fees at present paid for bitcoin transactions, and the potential for these fees is to increase with the network’s growth.

Though the internet is already massively big, it still shows a scaling limit to the bitcoin and other decentralized networks. Bandwidth is becoming better with each year, and though it is expected to have around 250 Tbps by 2016 end, as stated by Cisco[5], most of the bandwidth is spoken for already. This makes it quite critical to keep an eye on the scaling horizon. There seems still more room for growth; Jupiter accomplishes 1300 Tbps in the datacenter, and the global internet seems to finally reach its capacity[6].

Though bitcoin and the related ecosystem is an emerging and novel technology, it is critical to keep in mind that it extends to the other communities. Just like Bitcoin gave the first decentralized append-only database supporting transactions, there are other new solutions and systems that hold interesting properties. We must work to extend bitcoin technology with those advances with their maturity.