Nikita Ivanov - Founder & CTO, GridGain
A 2016 Deutsche Bank survey of 200 participants in the global financial industry found a very high degree of enthusiasm for blockchain, the digital-ledger technology behind Bitcoin. A remarkable 87 percent of respondents expect it to have a major impact on the securities services market, with 75 percent looking for widespread adoption within three to six years, and nearly two-thirds anticipating substantial cost savings in the range of 11-25%.
Implementing blockchain technology, however, requires significant infrastructure upgrades because legacy systems cannot easily provide the security and performance needed to support the decentralised blockchain architecture. As organisations develop their blockchain strategies, they must also create a scalable, distributed and very well-protected computing platform. Doing this cost-effectively will likely require an in-memory computing solution. A new white paper from GridGain Systems, Enabling Blockchain and Bitcoin with In-Memory Computing, provides an in-depth discussion on this topic.
The underlying blockchain technology was originally designed to allow small payments to be made without incurring credit card fees. Today, bitcoins are increasingly accepted for purchases, and Bitcoin virtual exchanges exist for buying and selling bitcoins, which are stored in electronic wallets. Whenever a Bitcoin transaction occurs, a secure transaction message goes from sender to receiver. Only the sender and receiver can view the contents of the message, but a record of the transaction is added to the blockchain and all blockchain subscribers can see the source and destination.
Because all subscribers receive the blockchain update simultaneously, no costly intermediaries or time-consuming reconciliation steps are required. The blockchain design also ensures integrity and security because the set of sequential “blocks” of encrypted transaction information can never be removed or deleted, only added to, in chronological order. Further, each block of transactions contains both a timestamp and a hash of the previous block, making the blockchain highly tamper-resistant.
With its simplicity, speed, transparency and security, blockchain has the potential to dramatically accelerate post-trade processing, collateral management, payment processing, trade finance, security servicing and processing, and transfer agency registrar functions. It can also enable same-day settlement and create more equilibrium and decentralization among players in the financial services industry. Organisations also have the ability to create private blockchains and allow only valid participants to use it – a scenario that might appeal to any financial services company providing the above services.
A high performance infrastructure is essential
The greatest challenge to participating in a blockchain is performance. Since blockchains have no central data repository, each party must collect the blocks of transaction information sent out to all subscribers, and these blocks constitute an ever-growing amount of data that must be consumed in real time. For Bitcoin, for example, there is just one ever-increasing blockchain of all transactions since the beginning. Financial services firms not only need to store the incoming blocks, but also translate and validate the information. They need to maintain tables of security identifiers, validation information, cross-references, and so on. They also need real-time transaction processing and analytical capabilities to enhance existing process with blockchain technology.
As a result, any application involving storing and processing of blockchain data will require a high performance, scalable architecture. Interoperability is also a must in order to integrate blockchain into existing financial systems and processes. Key requirements of a blockchain-ready architecture include distributed parallel processing capabilities to speed the calculations needed to build and update blockchains. Some applications, such as Bitcoin mining, also require especially strong analytical capabilities and processing power. A blockchain-friendly architecture should also include streaming support, enterprise-level security features, and a high level of transactional consistency.
In-memory computing is the most cost-effective technology to deliver these capabilities. Because in-memory computing involves keeping data in RAM for extremely fast access, with no disk-related slowdowns, it is faster than any other storage-based computing method. In fact, an in-memory computing cluster platform has been shown to process transactions roughly 1,000 times faster than disk-based solutions. An in-memory computing cluster also delivers high availability, disaster recovery, and concurrency across systems – key requirements for a blockchain application. Further, to support the full range of blockchain use cases, in-memory computing cluster solutions are available that deliver a combination of high-volume ACID transactions, real-time analytics and hybrid transactional/online analytical processing (OLAP).
While in-memory computing used to be cost-prohibitive, the cost of memory has dropped roughly 30% per year since the 1960s, and it is now relatively inexpensive to equip clusters of machines with terabytes of RAM to support very large and very fast data projects. This in turn has led to a rapid maturation of the in-memory computing market. It used to be that to deploy even a bare-bones in-memory computing solution, organisations had to configure and cobble together multiple products. Today, full-featured in-memory computing platforms featuring all the capabilities required for blockchain applications are readily available. Organisations with an existing code based can even find an in-memory computing solution that includes SQL support and a full range of APIs, leading to much faster deployments.
As blockchain interest continues to grow, so does the interest in in-memory computing, and many organisations are already at the proof-of-concept stage or close to moving a blockchain application powered by in-memory computing into production. Companies that delay developing a blockchain strategy based on in-memory computing will soon find themselves at a critical technical and competitive disadvantage.