Image Retrieval System Using Concurrent Key Splitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for processing checks are inefficient, requiring excessive use of storage space, computer resources, and time due to the need to copy, transmit, and upload electronic check images between multiple databases, which slows down the generation of electronic cash letters.
Innovation Solution
A system and method that retrieve electronic check images directly from local databases and incorporate them into cash letter job streams, eliminating the need for intermediate storage and transmission, and utilizing a splitter program to sort keys into individual files for concurrent retrieval via TCP/IP protocol, allowing for simultaneous connection to multiple servers and databases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods are used to process checks by copying and transmitting electronic check images between multiple databases, then the check processing can be completed, but the storage space consumption increases excessively
Solution Approach 1:
The patent extracts the electronic check images directly from the originating bank's database and incorporates them into the cash letter job stream without copying to intermediate storage locations. This eliminates the need for excessive storage space while ensuring check processing completion.
Solution Approach 2:
The system segments the check processing workflow into direct database-to-job-stream retrieval, eliminating intermediate storage steps. By dividing the process into direct extraction and immediate incorporation, the system reduces storage consumption while maintaining processing reliability.
2Reliability
If conventional methods are used to process checks by copying and transmitting electronic check images between multiple databases, then the check processing can be completed, but the computer resources consumption increases excessively
Solution Approach 1:
The system extracts check images directly from the source database and incorporates them into the job stream without intermediate copying operations. This eliminates unnecessary computer resource consumption associated with multiple copy and transmit operations while ensuring processing completion.
3Reliability
If conventional methods are used to process checks by copying and transmitting electronic check images between multiple databases, then the check processing can be completed, but the time required increases excessively
Solution Approach 1:
The patent extracts electronic check images directly from the originating bank's database and incorporates them immediately into the cash letter job stream, eliminating the time-consuming intermediate steps of copying to and transmitting through multiple databases. This reduces processing time while ensuring complete check processing.
Solution Approach 2:
The system performs preliminary extraction of check images directly at the source database before they would need to be transmitted through intermediate systems. This preliminary action eliminates subsequent transmission delays and accelerates the overall processing timeline.
4Ease of operation
If a single processor retrieves images from multiple databases sequentially, then the system is simple to operate, but the image retrieval speed is slow
Solution Approach 1:
The patent segments the image retrieval operation by dividing the set of databases into multiple groups, with each processor group assigned to retrieve images from specific databases concurrently. This segmentation enables parallel processing that increases retrieval speed while maintaining operational simplicity through organized task distribution.
Solution Approach 2:
The system merges multiple processor operations into concurrent execution, where multiple processors simultaneously retrieve images from different databases. This combination of parallel operations achieves high retrieval speeds while the systematic assignment of databases to processors maintains ease of operation.
5Productivity
If multiple processors retrieve images concurrently from multiple databases, then the image retrieval speed increases to 1000-2500 images per second, but the system complexity increases
Solution Approach 1:
The patent organizes the concurrent processing system by segmenting databases into groups assigned to specific processors. This structured segmentation enables high retrieval rates of 1000-2500 images per second while managing system complexity through systematic task allocation and organized database-processor mappings.
Data Source
AI summary
Aspects of this disclosure relate a system for retrieving images that may include one or more mainframe computer sites or Logical Partitions (LPARS) one or more servers operatively connected to the LPARS, and one or more databases operatively connected one or more of the LPARS. The system may also include a processor that may be configured to perform a method for generating electronic cash letters. The method for generating electronic cash letters may include reading a primary input key file assigned to the cash letter job, wherein the primary input key file includes a set of keys and each key corresponds to an image stored in one of the databases. The method may further include splitting the keys in the primary input key file into individual key files. The method may further include attaching a key processor subtask to each of the individual key files and running the key processor subtasks concurrently and independently of one another, wherein each of the key processor subtasks connects to one of the servers via TCP/IP protocol and retrieves the images from the particular database to which the keys in individual key files correspond.