Serverless Timestamp Assignment for High-Volume Transaction Ordering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing servers face challenges in distinguishing between transactions processed in a short period due to limitations in time resolution, making it difficult to manage high volumes of transactions per second, especially during peak times.
Innovation Solution
A serverless architecture system with modules for receiving, batch processing, and timestamp assignment, capable of generating unique timestamps with microsecond or nanosecond precision, allowing for efficient distribution and differentiation of transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing servers use standard time resolution, then they can process transactions with simpler timestamping, but they cannot distinguish between transactions processed in a short period with microsecond or nanosecond precision
Solution Approach 1:
The patent segments the timestamp generation process into multiple independent modules: receiving module for transaction requests, batch processing module for grouping transactions, timestamp assignment module for generating unique timestamps, and load balancing module for distribution. This segmentation allows each module to specialize in one function, enabling the system to achieve both high time resolution (nanosecond precision) and high productivity (1,000,000+ transactions per second) by parallelizing operations across multiple modules.
2Productivity
If the system processes a large number of transactions in parallel, then throughput increases, but maintaining unique timestamps for each transaction becomes difficult
Solution Approach 1:
The patent implements feedback mechanisms where the timestamp assignment module continuously monitors transaction batches and adjusts timestamp generation accordingly. The load balancing module receives feedback about system state and dynamically distributes batches to available timestamp assignment modules. This feedback loop ensures that even with parallel processing of millions of transactions per second, each transaction receives a unique timestamp by detecting and responding to timestamp conflicts or duplications in real-time.
3Productivity
If the system uses a single timestamp assignment module, then timestamp generation is simpler, but the system cannot handle 1,000,000 transactions per second
Solution Approach 1:
The system divides the single timestamp assignment function into multiple parallel timestamp assignment modules, each capable of independently generating unique timestamps. The load balancing module distributes transaction batches across these modules, enabling the system to scale to 1,000,000+ transactions per second. While this increases architectural complexity, the modular design allows each individual module to remain relatively simple while the collective system achieves high throughput.
Solution Approach 2:
The load balancing module serves multiple functions: it distributes batches to timestamp assignment modules, monitors system state, handles failover scenarios, and optimizes resource utilization. This multi-functional approach consolidates several responsibilities into a single coordinating module, reducing overall system complexity despite the presence of multiple timestamp assignment modules.
Data Source
AI summary
A method for assigning unique timestamps to transaction requests in a serverless architecture is provided. The method comprises receiving a plurality of transaction requests each having an initial timestamp; batching the initial transaction requests into initial batches using a batch processing module; distributing the initial batches to at least one timestamp assignment module using a load balancing module; and replacing the initial timestamp with a unique timestamp for each transaction request in each initial batch to generate an advanced batch according to an order in which the transaction requests are ordered in the initial batches using the timestamp assignment module, wherein the unique timestamp has an advanced precision of at least 1 nanosecond, the advanced precision of the unique timestamp being higher than an initial precision of the initial timestamp. A system having a serverless architecture for assigning unique timestamps is provided.