“Clouds” in a Cloudless Future
“Clouds” in a Cloudless Future
With the immense growth of cloud computing as a storage and compute platform, check out where blockchain can come into play.
Join the DZone community and get the full member experience.Join For Free
Learn how to migrate and modernize stateless applications and run them in a Kubernetes cluster.
How blockchain could overcome a major stumbling block to Dropbox, ICloud and the like.
Cloud services have been growing at such a rapid pace that analysts refrain from providing even short-term forecasts. Rare estimates are understated relative to reality. For the time being, the lack of data security guarantees is a major stumbling block to “clouds.” Blockchain is the solution, the Royal Bank of Canada’s analysts conclude. Indeed! The capabilities offered by a distributed ledger are the segment’s perfect match, though still leveraged to a very limited extent. CREDITS editors find out the reasons and possible ways of closer cooperation, if any.
Since the early 2010s, the cloud services market development has come on leaps and bounds. More and more people prefer cloud applications to hard drives and flash memory cards for their data storage.
However, accurate market analysis is hampered by the lack of a generally accepted calculation method. Every agency adheres to their own calculation approach, therefore, results differ drastically. The only thing they really have in common is that forecasts that looked overly optimistic early in the year seem to be excessively conservative by the end of the year.
In 2016, Gartner projected the “cloud” market to grow at 16% and was blamed for bias and too much optimism. And look how well that turned out. The market showed +20% growth rate! Furthermore, it was in 2016 that “price war” was launched (and is still going on), urging “clouds” to continuously lower prices. Growth rates persisted in 2017, the trends in 2018 have remained purely positive, too.
Here is another example of an erroneously modest forecast. In 2014, SAP expected 70% of IT companies to invest in cloud services in one form or another three years later. In fact, 94% of companies did by 2017. According to a forecast by Oracle CEO Mark Hurd, 80% of the IT industry’s budgets will be spent on “cloud innovation” by 2025.
“That IT transformation will have far-reaching implications. 80% of corporate data centers will be gone. These DPCs will only be kept to support legacy IT systems,” Hurd added.
Are Clouds and Blockchain Made for Each Other?
On the face of it, blockchain may seem to offer an optimal solution for cloud-based file storage. Decentralization, no option for unauthorized changes to saved files, full data encryption capabilities are not merely a technology — it all sounds like a “cloud” dream.
Recently, the Royal Bank of Canada’s analysts have published a publicly available report entitled “Crypto Currency & Blockchain Technology: A Decentralized Future — A Potential Multi-Trillion Dollar Opportunity.” The key conclusion is that cryptocurrencies are mistakenly treated as the store of value only, whereas in reality they also make the Internet more secure. For example, the blockchain-based tokens make it possible to create decentralized versions of cloud services, such as iCloud, Box, or Dropbox.
“In centralized environments, your data is owned and controlled by a third-party server,” analysts argue. “Moreover, clouds are vulnerable to hacks.” Think of 2016 when the SCO Online reported that roughly 40 million iCloud accounts had been hacked.
Analysts touched on the existing cloud characteristics, as exemplified by BOX. Overstaffing: ~ 1,500 employees. Outdated technology: Centralized data sharing. Poor security: Not-trustless. Central and encryption based, vulnerable to hacks on multiple points of potential failure. Subjective cloud management: governance through a board of directors.
Decentralized cloud can manage with a few dozens of employees. Technology security — Decentralized Storage Network — is supported by cryptography. The cloud is governed through the unbiased Consensus Protocol.
Still No Clicking, Though
However, the two extremely innovative industries — clouds and blockchain — are not friends yet. The relevant startups are scarce (hugely less than the clones of “CryptoKitties”), and the existing ones can hardly be called a success.
Why so? Igor Chugunov, CEO & Founder at CREDITS, explains:
“Conventional blockchain platforms face a number of problems that significantly hinder the handling of large data volumes. It’s not just about the number of transactions per second — there are many sides to the problem. How long it takes to process one transaction is equally important — it directly affects the storage operation speed, the ease of access to files, and the block recording time. You must admit that people would be unlikely to feel like using a “cloud” where it can take them a couple of minutes to simply rename files.”
In addition, the consensus algorithms employed on most blockchain platforms (Proof-of-Work and its modifications), as well as the algorithms for approval of the previous correct status of blockchain (Merkle Trees) are characterized by extremely low energy efficiency, high power consumption, and requirements for the computational performance of equipment, resulting in high commission rates and rendering “blockchain-based cloud” non-competitive amid the “price war” which is still pending in the market — or else the price rates for services will be too high if at least the minimum rate of return is to be reached.
Luckily, most of these problems can be avoided or at least eased to a great extent, if not remedied. This can be achieved by working on several fronts at a time. First, the file ID search algorithms now in place are faster and less costly than the habitual Merkle Trees.
“Instead, we use a transaction register where each entry consists of the block of transactions’ hash code and contains the transaction direction, the initial and final wallets, the type of debiting, the number of debiting units, the type of crediting, and the number of crediting units. This approach enables us to revolutionize the transaction processing speed and eliminate the risks of unvalidated changes to the register,” says CTO at CREDITS Eugeniy Butyaev, describing his alternative solution to the problem.
The second important front is the development of data compression methods. On the cusp of cryptocurrencies, the blockchain was a hugely “cumbersome environment” (look at the size of the Bitcoin blockchain), whereas today state-of-the-art projects allow files on the blockchain to be compressed by some 90%.
Finally, the third front is throughput improvement. However, making it meaningful for the “cloud” necessitates an increase of tens of thousands of percentage points and not tens of percentage points, like with most promising projects.
The only practical way to achieve this is “asynchronous data processing,” ensuring the largest possible number of transactions processed at the same time. It is of paramount importance to the so-called Internet of Things projects as these networks should have numerous transactions. Yet, similarly to Merkle Trees, this requires non-conventional solutions — the employment of Proof-of-Work and most of its variations is unable to procure the necessary operating speed.
How Can the “Blockchain-Based Cloud” Be Used?
CREDITS blockchain platform is among the few projects to leverage all of the above-listed approaches at the same time. It is primarily designated to create decentralized applications.
“Many projects are limited to minor improvements vs. the existing products. They find an approach like this to be more secure and at lower technical risk. Yet customers have no interest in these products — insignificant improvements do not justify the transition to a new platform. We, on the contrary, have decided to propose something really new,” Igor Chugunov explains.
Our platform will be capable of processing one million transactions per second. The processing time for one transaction is planned to be reduced to the hundredths of a second; the completion cost, almost to zero. We tested our system in March. Its alpha version has already processed a pool/block of 488,403 transactions per second during periods of peak load.
Although initially the finance industry was the primary target for CREDITS, achieved characteristics render the platform attractive in a plethora of fields. The clouds included.
“Decentralized storages can’t be hacked as they’ll have to hack all network nodes and not just one server. You will be allowed to access your information using Private Keys. File sharing is done with a Public Key. Users contributing and serving storage capacity receive tokens. And vice versa, customers spend tokens for receiving extra storage capacity,” Igor summarized, drawing on the cloud of the future.
Opinions expressed by DZone contributors are their own.