Constellation Insights

There are more than 700 cryptocurrencies available for trading online, and many are derivations of Bitcoin, which in turn is based on the blockchain distributed ledger technology. (Go here for Constellation Research VP and principal analyst Steve Wilson's concise definition of these terms.)

IOTA took a different approach entirely when it began working on a DLT for the Internet of Things two years ago. Rather than use blockchain, its Tangle architecture uses a DAG (directed acyclic graph). IOTA co-founder Dominik Schiener summed up the differences in a recent blog post:

At its core, the Tangle still has the same underlying principles as a Blockchain: it’s still a distributed database, it’s still a P2P Network and it still relies on a consensus and validation mechanism.

But, if we are to summarize the main differences between the Tangle and the Blockchain, the two most apparent ones are how the Tangle is structured (a DAG), and how we achieve consensus.

In IOTA there are no “blocks” in the classical sense. Instead, a single transaction references two past transactions. This referencing of transactions is seen as an attestation: with your transaction you attest directly that two transactions, and indirectly that a subsection of the Tangle are valid and conform to the protocols rules.

Instead of a smaller subset of the network being responsible for the overall consensus (miners / stakers), the entire network of active participants (i.e. devices making transactions), are directly involved in the approval of transactions. As such, consensus in IOTA is no longer decoupled from the transaction making process: it’s an intrinsic part of it, and it’s what enables IOTA to scale without any transaction fees.

Transaction fees are no joke in the cryptocurrency world. On average, Bitcoin transactions now cost more than $1. That may not be much if the amount being moved is significantly large, but IOTA envisions a world where many millions of IoT devices will not only exchange data but make purchases from other machines, as Schiener writes:

We envision a future where Machines trade resources (computation, electricity, storage, bandwidth, data etc.) and services with each other without the involvement of any third party — purely Machine-to-Machine. As the Internet of Things starts unleashing itself, the need for “Smart Decentralization” is apparent.

Transaction fees are, of course, anathema to that vision, given that in many cases a machine-to-machine purchase could be measured in cents. Tangle's architecture is intended to make transaction fees a non-issue. Tangle also takes aim at massively increasing scalability and transaction throughput, factors that have plagued previous cryptocurrencies, particularly Bitcoin. In turn, the question of scalability is inextricably intertwined with the notion of a massive network of machines making autonomous purchases from one another. 

Tangle's architecture is described in much greater depth through IOTA's well-written whitepaper, which is available here. The startup's work on Tangle dates to 2015, although its founding members have been working with blockchain and distributed ledger technology much longer than that. 

Its cryptocurrency began trading earlier this month and quickly accumulated a market value of more than $1 billion. As for Tangle, IOTA is conducting case studies for the platform in energy, mobility, smart cities and infrastructure, and expects real-world projects to emerge this year. 

While it is early days for IOTA, the startup's work looks promising and speaks to critical pain points for IoT, says Constellation Research VP and principal analyst Andy Mulholland.

"It has been apparent for some time that the requirement definition for IoT, and the requirement definition for fintech are very different, and that the terms blockchain and distributed ledger are all too often used with too little in-depth understanding," he says. "For those faced with the challenge of managing the commercial business exchange of the very, very small amounts of data that IoT endpoints will contribute individually—though collectively there are millions of endpoints,—will be very interested in the approach that IOTA brings."

"With transmission costs at the extreme micro level, and a different approach to security, one more in keeping with the nature of dumb sensors, IOTA is bringing a useful alternative in the distributed ledger space to Hyperledger," he adds.

Hyperledger is a fast-growing industry consortium working on distributed ledger technologies, counting heavyweights such as IBM among its members. 

This week, the group announced the formation of a new performance and scalability working group that will focus on both software and hardware-related issues:

The PSWG is starting at the base of the mountain in terms of performance and scalability. The group will identify multiple key use cases across different vertical markets. For each of the use cases key metrics will be identified. Each metric will be defined along with how it will be measured. This work will lead to defining a framework and measurement criteria.

The key will be how to define and implement a consistent and fair test suite for each use case. This is important to the industry and to the end user, as it will provide a fair measure of how a specific implementation will perform in different use cases. This will allow the user (customer) to select the proper implementation for their use case.

"Hyperledger is seen by many as the most significant industry initiative to deliver commercial decentralised transaction management, but the question as to scalability, and indeed suitability, continues to be the big issue," Mulholland says. "This new working group is a key move."

24/7 Access to Constellation Insights
Subscribe today for unrestricted access to expert analyst views on breaking news.