Two-Level Indexing for Compressed Transaction Record Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Telecommunications operators face significant financial and operational challenges in storing and retrieving large volumes of transaction records, such as call records, due to the need for rapid search and retrieval capabilities over extended periods, which existing methods struggle to efficiently address.
Innovation Solution
A method involving the collation of transaction records into groups based on predetermined criteria, followed by first-level indexing and compression, allowing for efficient storage and subsequent rapid retrieval through the use of first and second-level indexes, significantly reducing data storage requirements while enabling detailed search and retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If transaction records are stored in uncompressed form with detailed indexing for rapid retrieval, then retrieval speed is improved, but data storage volume increases significantly
Solution Approach 1:
The patent divides the index structure into two levels: first-level indexes that provide broad navigation to compressed record groups, and second-level indexes that enable detailed search within decompressed groups. This segmentation allows the system to maintain fast retrieval capability while working with compressed data, resolving the contradiction between retrieval speed and storage volume by organizing indexing operations hierarchically rather than requiring all indexes to be maintained on uncompressed data
Solution Approach 2:
The patent changes the physical state of the transaction records from uncompressed to compressed form for storage, while maintaining the ability to retrieve and decompress specific groups on-demand. This parameter change (compression ratio) dramatically reduces storage volume while the two-level index structure ensures that retrieval speed is preserved by enabling selective decompression of only relevant data groups rather than requiring the entire dataset to be uncompressed
2Quantity of substance
If compression algorithms are applied to reduce storage volume, then data storage requirements are reduced, but retrieval and search capability deteriorates
Solution Approach 1:
The patent performs preliminary indexing operations at two levels before compression is applied to the actual record data. First-level indexes are created to map compressed group identifiers to their locations, and second-level indexes are prepared to enable search within groups. This preliminary action ensures that even though the data is compressed, the retrieval and search operations can proceed efficiently by using the pre-computed index structures to guide selective decompression and search operations
Solution Approach 2:
The patent introduces a two-level index structure as an intermediary between the compressed data storage and the search/retrieval operations. The first-level indexes act as a mediator to locate relevant compressed groups, and the second-level indexes mediate the search process within decompressed groups. This intermediary index structure allows the system to work with compressed data while maintaining full search capability, as the indexes translate search queries into efficient access patterns without requiring the underlying data to be uncompressed
3Adaptability or versatility
If all transaction records are kept in accessible form for any combination of search criteria, then search flexibility is improved, but storage cost increases significantly
Solution Approach 1:
The patent segments the transaction records into multiple compressed groups, each with its own second-level index, while maintaining a first-level index that provides overall navigation. This segmentation allows the system to maintain search flexibility across all records by organizing them in manageable units, while simultaneously reducing storage costs through compression. The segmented structure enables the system to handle any combination of search criteria by navigating through the hierarchical index structure and decompressing only the relevant segments rather than requiring all data to be continuously accessible
Solution Approach 2:
The patent applies compression as a parameter change to the stored data, transforming it from an uncompressed accessible state to a compressed state that reduces storage cost by a factor of ten or more. The two-level index structure compensates for this parameter change by enabling efficient navigation and selective decompression, thereby maintaining search flexibility across all transaction records despite their compressed state. This allows the system to support any combination of search criteria while storing data in a space-efficient compressed format
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and apparatus are provided to store transaction records in a retrievable form and to enable subsequent search and retrieval of stored transaction records. Transaction records are captured and then grouped according to predetermined grouping criteria such that they may be indexed to a first level and then efficiently compressed for bulk storage. In the event that records need to be retrieved subsequently, the first level index may be used to select one or more groups of records satisfying first level search criteria and, following retrieval of the selected groups from storage and de-compression, a second level index may be created to enable a more detailed record-level search for matching records in the retrieved groups. Preferably, the same indexing technique is used for both the first and second level of indexing.