The primary goal of the R&D Programme is to develop Bitcoin mining tools and techniques which will significantly outperform existing practices.
It is an extremely ambitious project, but the radically advanced technologies being utilised render the goal achievable.
QBT’s R&D team is composed of a number of sector experts selected across the UK and Italy, and includes highly skilled professionals, Ph.D students and university professors, with expertise in Quantum Computing, Machine Learning, Cryptography and Algorithms Optimisation Theory.
The main research areas are:
I. Quantum Computing
The quantum computing R&D focus is on exploration of the simultaneous solution of very large optimisations tasks of the Bitcoin mining algorithm, using the quantum annealing basic computation mechanism provided by D-Wave. Several aspects of the overall Bitcoin mining process are potential targets for a quantum approach. These are being explored and their performance is being measured, in terms of execution time and quality of the solutions.
II. Cryptographic Optimisation
While SHA256, the heart of the Bitcoin mining method, is a general-purpose cryptographic algorithm, the restrictions on the input message and output hash strings, are such that a number of specific features can be identified and exploited by QBT’s R&D team for designing optimised versions of the SHA256 algorithm.
III. Deep Learning and Artificial Intelligence (“AI”)
This research area is key for QBT, since the analysis of very large collection of mining data is key for the optimisation of the mining process. Advanced neural networks and the associated monitoring tools are not only knowledge extraction systems, but they play a fundamental role in supporting the human bitcoin mining algorithm experts, allowing them to focus on specific areas to improve their analysis capabilities.
IV. FPGA / ASIC Design
Performance of QBT’s proprietary Bitcoin mining optimisations is first assessed on an FPGA chip, initially by simulation and then tested on real chips. This provides valuable input to QBT’s team for the ASIC design. Performance simulation of the ASIC architecture of the optimised SHA256 cryptographic algorithm is carried out before small batch prototype production, using 12 nm technologies which will be used for the first version of QBT’s ASIC Bitcoin mining chip.
V. Algebraic and Boolean Equation Reduction
The ability to transform a sequence of logical and algebraic complex expression in an equivalent one, but shorter and easier to be calculated, is a key element of the achieving material optimisation of the Bitcoin mining algorithms. The combined use of our human experts and automatic expression reductions tools, is a very promising mathematical approach to achieve alternative methods to optimise the Bitcoin mining algorithm. QBT’s R&D team’s expertise on the subject is key to this alternative mining approach.
VI. Very Large Big Data
A very large big data repository is a key element to analyse the properties of the Bitcoin mining algorithm; several Exa computations are being generated and stored in the Epa memory hardware, using advanced compression techniques, for fast storage and retrieval, making available to the other research areas a unique wealth of data for their research activities.
VII. High performance computing architectures
The design of a high-performance IT supporting infrastructure has been guided by the adoption of the best-in-class solutions for the internal communication of the development and production equipment, connected via optical fibres, with storage devices buffered via specific servers and several RAM on board Terabytes for the individual processing units. On the fly pre-processing of data fed to on-site and off-site miners, represents a key competitive advantage which QBT’s R&D team regards as a strategic factor. Remote and on-site general purpose, FPGA and ASIC hardware integration with a quantum processor is one of the topics of focus of QBT’s R&D team.