Key-Value Store Using Ranged Keys in Skip Lists
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional key-value stores are limited by their single arbitrary mapping between keys and values, restricting comparison and operation types, which hampers throughput performance.
Innovation Solution
Implementing a key-value store using skip list data structures with ranged keys, allowing multiple layers with nodes storing corresponding key ranges, and utilizing a burst buffer appliance for enhanced processing and storage operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional key-value stores use single arbitrary mapping between keys and values, then the data structure is simple, but the throughput performance is limited and operation flexibility is restricted
Solution Approach 1:
The patent segments the key space into multiple ranges, with each node in the skip list storing a ranged key representing a key range rather than a single key. This segmentation allows parallel processing of different key ranges, improving throughput performance while maintaining the skip list's relatively simple structure
Solution Approach 2:
The patent adds a new dimension to the key storage by using multi-layer skip lists where each layer can store different ranged keys. This dimensional expansion enables more operations to be performed simultaneously across different layers, enhancing throughput without fundamentally complicating the underlying data structure
2Adaptability or versatility
If conventional key-value stores provide single arbitrary mapping, then the data format is simple, but the flexibility of comparison and operations is limited
Solution Approach 1:
The patent implements multi-functional ranged keys that can represent different types of comparisons (less than, greater than, equal to, range queries) within a unified data structure. Each ranged key can support multiple operation types, making the key-value store more versatile without requiring separate specialized structures for each operation type
Solution Approach 2:
The patent introduces dynamic ranged keys that can be adjusted and modified based on operation requirements. The skip list structure allows for dynamic insertion, deletion, and modification of ranged keys, enabling flexible adaptation to different query patterns and operation types while maintaining structural integrity
Data Source
AI summary
A processing platform of an information processing system comprises one or more processing devices and implements a key-value store that utilizes a skip list data structure having a plurality of layers each having two or more nodes, with each of at least a subset of the nodes of the skip list data structure storing a corresponding ranged key having a range of keys associated therewith. A given one of the ranged keys corresponds to a single value stored in the key-value store and comprises a lower key and an upper key of the range of keys but does not explicitly include intermediate keys in the range of keys. A burst buffer appliance may be associated with the key-value store and configured to process ranged keys of one or more skip list data structures for storage in or retrieval from the key-value store.