Cryptocurrency Miner Compute Engine Series Arrangement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The winner-takes-all compensation scheme in cryptocurrency mining leads to an arms race among miners, resulting in high costs and the formation of mining pools to share risks, but this approach is inefficient due to the all-or-nothing nature of block awards, which can lead to increased computational capacity and subsequent adjustments in difficulty, affecting the stability of block creation times.
Innovation Solution
A cryptocurrency mining system where compute engines are organized in series within compute boards, reducing the overall current required to power them, thereby minimizing IR drops and electromigration, and allowing for more efficient operation by distributing computational tasks across multiple compute engines, each powered in series rather than in parallel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If compute engines are powered in parallel to increase computational capacity, then processing power is improved, but current magnitude increases causing higher IR drops and electromigration
Solution Approach 1:
The patent segments the compute engines into multiple series-connected groups rather than connecting all engines in parallel. This segmentation allows the system to maintain high computational capacity through multiple groups while reducing the current burden on any single group, thereby minimizing IR drops and electromigration effects.
Solution Approach 2:
The patent transitions from a single-dimensional parallel architecture to a multi-dimensional architecture combining series and parallel connections. By organizing compute engines in series-parallel configurations, the system achieves both high computational capacity (through parallel groups) and reduced current stress (through series connections within groups), effectively adding a new dimensional approach to system design.
2Productivity
If more compute engines are added in parallel to handle increased difficulty, then computational capacity is improved, but system complexity and power distribution challenges increase
Solution Approach 1:
The patent divides the compute engine array into multiple series-connected groups that can be independently powered. This segmentation simplifies power distribution by allowing each group to be managed separately, reducing the overall complexity compared to distributing power to numerous parallel-connected engines.
Solution Approach 2:
Instead of the conventional approach of connecting compute engines in parallel to simplify power distribution, the patent inverts the approach by using series connections. This inversion, while counterintuitive, actually simplifies the power distribution architecture by reducing the total current required and enabling more manageable power delivery to each series group.
3Productivity
If compute engines operate at high current to maintain hash rate, then productivity is improved, but IR drops and electromigration reduce system reliability
Solution Approach 1:
The patent introduces a series-parallel architectural dimension that allows compute engines to operate at high power levels while distributing the current load across multiple series groups. This dimensional change enables maintaining high hash rates without subjecting individual engine groups to excessive current stress, thereby preserving system reliability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This configuration reduces the magnitude of current supplied to compute engines, enhancing their reliability and performance, allowing for more stable and efficient cryptocurrency mining operations while maintaining the required computational capacity within the network.
Implementation Method 1
A cryptocurrency miner with current reducing compute engine arrangement... reducing the overall current required to power them, thereby minimizing IR drops and electromigration
Data Source
AI summary
A cryptocurrency miner includes a power supply, a network interface, a compute module, and a controller. The compute module includes compute engine stories coupled in series with the power supply. Each compute engine story includes compute engines coupled in parallel between a voltage input node and voltage output node of the respective compute engine story. The controller receives, via the network interface, a job from a pool server of a mining pool and distributes aspects of the job to the compute engine stories.