top of page

Feature 25 February 2022 (ComputerWeekly)

In the drive to reduce costs and fund growth, the tide could be turning on blockchain and crypto: Gartner predicts that 20% of large organisations will have adopted digital currencies for payments, stored value and the like in less than three years.

Most organisations will not need to develop their own customised blockchain application stack because many large financial services and digital-asset firms have already done the heavy lifting. But what about data management?

Daniel Bizo, research director for Uptime Institute Intelligence, says that energy-intensive crypto and digital currency networks typically do not run in “normal” datacentres, requiring special hardware at the node for economic viability.

The “immediate global problem” is power usage, which remains difficult to track and attribute.

For crypto, the energy gap to fiat currencies is still “vast, up to five orders of magnitude”, says Bizo.

This might entail changes in power capacity, cooling infrastructure, equipment form factors and serviceability in facilities, compared with mainstream IT infrastructure. Those differences could indeed be large enough to create operational challenges for facility operations staff, he says.

So what’s driving the shift? Gartner says blockchain and digital currencies retain theoretical appeal among chief financial officers. The hope is for lower costs, faster transaction processing, continuous accounting and auditing, as well as global reach in a low-error environment.

Rising demand

That’s why central banks are likely to quickly follow China’s lead and launch their own digital currencies. Many more organisations, for one thing, will look to reap benefits from decentralised finance applications, including a chance to hedge against high inflation, according to Gartner.

Kaj Burchardi, managing director and blockchain lead at consultancy BCG Platinion, sees substantial growth on the horizon, especially in Stablecoin, central bank digital currency (CBDC) and the tokenisation of assets.

“The demands of running nodes and the required hardware and services in datacentres will increase as well,” he says.

However, third-generation and emerging blockchain platforms are primarily proof-of-stake, with the latter’s blockchain transaction validations determined in a way that consumes less energy than Bitcoin’s proof-of-work consensus protocol, says Burchardi.

“Only one person or participant is actually validating a transaction,” he says. “That’s randomly allocated, and that’s why you don’t need a lot of parallel computing. The impact on datacentres with proof of stake is minimal – or no more than any other general application you’ve built.”

Negative carbon footprint

Some platforms may already technically have a negative carbon footprint, he suggests, combining minimal consumption with carbon offsets.

Beyond currencies and trading are the blockchain use cases around trade finance and goods transfers, using small software tokens for tracking and supply chain transparency. Platform developer Minexx announced the export of the first kilogram of blockchain tracked gold from Burkina Faso in December 2021, and BCG Platinion built the Tracr platform for the track-and-trace of diamonds from mine to customer.

“It’s supported by blockchain technology because the bill of lading and the approval of what’s in the container versus what the document is saying is continuously tracked,” says Burchardi. “Therefore, you achieve a high degree of transparency.”

The main use of computing power is the validators at the nodes, but these don’t need to be numerous – assuming participants do not need to have the entire blockchain on their end as well as a copy to gain full transparency. But, as ever, use case is key.

“You need a reasonable number of transactions, with only one transaction per week, and you don’t need big software support,” says Burchardi. “If you have millions of transactions per second with a super low-latency requirement, such as high-frequency traders, blockchain probably wouldn’t be the best technology either.”

Network effects might flow into IT infrastructure and datacentre management. Multiple questions and transactions, as well as platform interconnections, can exacerbate complications and consequences alike.

Connections and connectivity

Where there might be a business opportunity for providers is for enabling linkages to different blockchain platforms, smoothing adoption for customers. However, the more scattered the blockchain landscape in question, the less value it might deliver, says Burchardi.

Simon Riggs, PostgreSQL fellow at managed database provider EDB, says blockchain can be thought of essentially as a set of data store requirements.

“It just so happens that those requirements are not so strange,” he says. “OK, blockchain stuff is a little more ‘extreme’, but there are other ways of doing it. I don’t foresee that many changes in the infrastructure as a result of the emergence of blockchain.”

Someone has already written a blockchain extension for PostgreSQL and published the details – with the choice also dependent on what’s in-house already, he notes.

Steven Rackham, senior solutions engineering manager of global finance for Europe, the Middle East and Africa at storage giant Netapp, says one issue can be transaction processing bottlenecks – with blockchains currently able to process far fewer transactions a second than, say, credit cards.

Blockchain, which can be used to trigger certain activities based on certain criteria, may affect automation practices, too, he notes, and then there are regulations to worry about and whether it will work with what you have already.

“And how do you get what you’ve got into a blockchain environment? One of the biggest challenges is going to be incorporation into existing use cases and work models,” says Rackham.

Availability and disaster recovery

Rackham foresees potential challenges around availability and disaster recovery, because blockchain is a distributed ledger. Then there’s a competing driver, for greater flexibility of provision – turning stuff off more often to reduce consumption, for example.

“This could put more pressure on the sustainability story,” he agrees. “It might be a reason to go to cloud.”

Techniques such as sharding are being looked at to help manage the distributed network by segmenting and working on certain areas or applications at a time, says Rackham.

Chris Royles, field chief technology officer at data platform provider Cloudera, confirms customers are interested in distributed ledgers, including for tracking and tracing, including high-value assets but also, for example, in construction sites.

However, problems certainly exist, if the reportedly high prevalence of non-fungible token (NFT) fraud – when blockchain was meant to be about guaranteeing trust across networks – is any guide.

Users will need to take care to ensure they get that desired trust across collaborative networks – despite a near-viral level of hype.

“Physics always applies,” says Royles. “Transactions on the ledger can take time. One aspect of distributed ledger is that there needs to be a network of machines connected in such a way that they can process problems at scale.”

Moving to more distributed networks could be paralysing, since each contributor can in principle do its own processing. This suggests issues when it comes to actually realising holy-grail blockchain concepts such as smart contracts.

“If you trust across a set of organisations, those organisations have to deploy the distributed ledger into their own datacentres,” says Royles.



bottom of page