The $28-Billion Challenge: Can Ethereum Scale to Meet Demand?

shutterstock_392885236-e1498508497495
30 June 2017

For all the recent successes, this much is clear: ethereum is still far from becoming the “world computer” its engineers originally envisioned.

The best example of ethereum’s in-progress status arguably occurred earlier this month, when the world’s second largest blockchain by investment saw record-high transaction activity thanks in part to a single popular initial coin offering (ICO). The sale, which saw users compete to purchase a new token for an application called Status, ended up affecting the entire network, now valued at more than $28bn, causing it to run more slowly for all users.

Due in part to how the specific ICO was engineered, users struggled to get transactions and contract updates processed, while a group of exchanges temporarily shuttered ether trades to deal with the congestion.

In response, the cryptocurrency world was abuzz.

On the one hand, enthusiasts had every reason to be excited about the burst in ethereum use. The downside, however, was that ethereum effectively buckled at a level of capacity that wasn’t even close to where its engineers want to take it.

Critics quipped that while the goal is for ethereum to replace today’s internet, the platform couldn’t even handle one fundraiser (if, admittedly, a big one).

Further, the growing pains point to another growing disconnect – the gap between how the platform is advertised by its sometimes zealous users and what its current capabilities are. (The price of ether, for instance, has increased by more than 3,000% since the end of 2016, even despite a recent price dip.)

The issues raise important questions for the ambitious blockchain platform. Namely, will ethereum ever be able to scale to support a large number of users? And, just how many will it be able to accommodate?

Future technology

To put ethereum’s issues into context, all open blockchains, including bitcoin, have capacity problems. In fact, for the last two years, the bitcoin community has been arguing about the best way to increase its limited capacity without compromising decentralization, the characteristic that sets bitcoin apart as a payment network.

While newer ethereum investors might not be familiar with these big-picture problems, its developers have been experimenting with potential solutions since even before the platform launched in 2015.

More recently, ethereum creator Vitalik Buterin put forth an idea for boosting transaction speeds on the network during congestion periods like the one seen last week. If implemented, ethereum would be able to handle eight ICOs at the same time, rather than just one.

But while this could help with transaction throughput in the short term, it’s still a long way from what ethereum enthusiasts envision. (In the future, for example, some have even gone so far as to imagine scenarios where every article on the internet would become a mini-ICO).

The short version of that conclusion is that ethereum can’t support many users or apps right now, but with the help of up-and-coming technologies, the hope is that it will one day be able to.

Off-chain solutions

Of these technologies, “payment channels” are probably the furthest along.

The in-progress Raiden Network, an ethereum implementation inspired by bitcoin’s Lightning Network, adds an extra layer to the ethereum blockchain that could increase transaction throughput from roughly 13 per second to as many as a million. And a minimum viable product of the technology could be ready by the end of the year, according to the most recent update from Consensys engineer Ameen Soleimani.

In order to work as planned, though, ethereum needs to scale not just for payments, but for a range of non-financial smart contracts – the technology that separates it from older public blockchains such as bitcoin.

Payment channels do not help smart contracts to scale, though. So-called “state channels” aim to take the idea further by applying the same off-blockchain payment technology to smart contracts.

At least one project claims to have already created state channel technologies that work for its decentralized applications, although the technology is not open-source. Some developers are also working to standardize this technology to work for a range of use cases, so developers don’t have to build state channels from scratch to work for their specific apps.

Another project, TrueBit, uses a similar idea.

In order to scale ethereum’s computations, the idea is to execute smart contracts off-chain instead of running them directly on the blockchain. Then, smart contracts can be kicked back to the blockchain when necessary to resolve disputes.

But, like state channels, it’s still mostly in the concept stage for now, though the team recently set up the TrueBit Foundation with the goal of chipping away at the code and turning it into “a living, breathing system.”

Sharding

Then there is a longer-term, and more theoretical, scaling idea that borrows from the world of traditional databases.

When asked about scaling, National University of Singapore PhD student Loi Luu was quick to mention sharding – a theory that suggests that each node would only need to store a subset of the data – and how it could be a more powerful way of boosting ethereum’s computational capacity.

Still, Luu said the problem isn’t going to be fixed overnight: he expects it will take another 1–2 years to implement the technology.

“The high-level idea is clear, but how to do it practically? The community needs to work on it for quite some time,” he said. “Everyone knows what they want to achieve, it’s just a matter of how.”

Too much ‘hand-waving’?

Still, while some like Luu are optimistic that engineers will find a way for ethereum to support more users, others are skeptical.

Some in the cryptocurrency community have long been critical of ethereum’s ambitions. The traditional argument is that, because it aims to take blockchain technology beyond currency, it needs to store more data than a less flexible platform like bitcoin. (To that point, despite being years younger than bitcoin, ethereum now requires users to store more data.)

Coinkite CEO and co-founder Rodolfo Novak said, for that reason, he’s skeptical of the platform in general. He even went as far as to remark: “I’m surprised [ethereum] has lasted this long.”

He took issue with the concept of sharding, in particular, noting that the tech is still in the white-boarding stages.

Novak concluded with a warning:

“There is a lot of hand-waving tech and very good marketing. That, to me, is an irresponsible level of irrational exuberance. It won’t end well.”

Still, many users and companies are anticipating that such scaling technologies, will one day be made to work. And advocates argue that it’s still very much the early days for ethereum.

In the Medium post “Scaling Ethereum to Billions of Users,” Fred Ehrsham – who formerly co-founded Coinbase and now advises blockchain startups – argued that a current bottleneck is that very few developers are working on each scalability project right now.

Whatever the case, perhaps last week’s issues will push the network’s current limitations further into the limelight and bring a sharper focus on the issue for those involved.

Disclosure: CoinDesk is a subsidiary of Digital Currency Group, which has an ownership stake in Coinbase. 

TV glitch image via Shutterstock