March 24, 2017
The International Business Machines Corporation recently launched a commercial blockchain platform for enterprises to develop, deploy and test their business applications in a distributed ledger environment. Often referred to as the IBM Blockchain the platform, in turn, sits on top of Hyperledger Blockchain’s open source code. Well before the launch, there had been much talk in the industry about the IBM-Hyperledger consortium that was seen as one among the big three consortia driving enterprise blockchain initiatives, other two being R3CEV and Digital Asset Holdings.
Hyperledger describes itself as a protocol enabling business to business and business to consumer transactions. In this article, we attempt to discuss the underlying hyperledger code fabric in the light of the business functionalities that can be derived from its technical aspects and how it deviates from the most successful consumer side blockchain implementation so far, i.e., bitcoin.
The architecture of the Hyperledger Blockchain as depicted in the following figure, encompasses three key components managing 1) Membership 2) Blockchain & Transactions and 3) Chaincode (smart contracts)
Image courtesy: Hyperledger Whitepaper
The below table summarizes its technical deviations from bitcoin and its business-side impact for anyone looking to evaluate the same:Parameter Bitcoin’s Blockchain Hyperledger Blockchain Business Impact Network Permissionless network without any central authority Permissioned & permissionless networks Businesses can choose to restrict access to their network Consensus Algorithm Rigid consensus algorithm that requires proof-of-work computations, also termed as mining: Modular, plug-and-play consensus algorithms customizable to use-cases. Ex: Paxos, PBFT, Round-robin, et. Mining is expensive; Hyperledger attempts to save on computation cycles at the cost of lesser decentralization & subsequent network effects Smart Contracts Employs a weaker version of smart contracts – Scripting, which enables to replace public key ownership script. Example: multi-sig scripts Hyperledger Chaincode entity provides the capability to define smart contract templating language (similar to Velocity or Jade), restrict the functionality of the execution environment and the degree of computing flexibility to satisfy the legal contractual requirements Businesses can create applications to embed and execute business logic, a schema for the same can be extended to clients and partners to access a common dashboard Identity & Auditability Public keys which are visible to all while private keys which execute transactions, all managed within the network A user registration authority encapsulates confidential user data, which then generates security keys for members to transact The identities of transacting parties can be concealed – one of the basic requirements for any financial use cases Private Transactions and Confidential Contracts For all bitcoin transactions, the public addresses and content is broadcasted on the shared ledger to all the nodes, minimal content confidentiality Content confidentiality is achieved by runtime loading and decrypting at transaction as well as business logic (chaincode) level Transaction patterns can be prevented from being observed and interpreted, shared ledgers doesn’t give away details about business relationships that should not be revealed to competitors Performance & Scalability One block mined every 10 minutes, fixed block size of 1 MB – latency amounts to 1 hour or 6 blocks, peak throughput at 6–7 tps Use-case dependent performance and scalability, trade-off between transaction speed and decentralization, undergoing research to improve on the same Scalability still stands as one of the major roadblocks since many businesses require technical environments with 10k tps and above.
Having said the above, all is not going good for any commercial blockchain offering, the platforms have only recently shifted from testing to pilot phases and majority of enterprises still appear to be in the exploration mode, definitely not looking to go all in.
It appears from our analysis that performance and scalability of blockhain has emerged as one of the pressing pain points in its commercial adoption. Amidst this, there is a lobby of early bitcoin adopters which seeks to scale the most widely adopted blockchain, i.e., bitcoin’s blockchain by increasing block size, targeting a performance improvement of up to 4000 tps (equivalent to performance capability of major card networks) from the current figure of 7 tps. This, some of them plan to achieve by a series of hard forks over 12 years.
If in future, this level of flexibility is evinced by an already well-established blockchain, running $20bn in market capitalization and ripe with evident network effects, then it would not be unreasonable to expect enterprises and financial institutions to switch sides.