Flexible Data Access Layer for Cloud Query Performance
Solution Overview
Problem
Existing data access layers in computer systems, particularly in cloud platforms, are complex and lead to slow query operations due to their complexity and dependencies, affecting productivity and performance, especially as data volumes grow.
Innovation Solution
A simple and flexible data access layer is implemented using object-oriented mapping APIs that allow users to persist and query data in key/value pairs, sorted sets, maps, and fat events, reducing complexity and enhancing performance by eliminating the need for joins, constraints, and stored procedures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional DAO implementations (JPA/Hibernate, jOOQ, JDBC) are used to provide data access layer, then data access functionality is achieved, but system complexity increases and query operation speed decreases
Solution Approach 1:
The patent extracts and eliminates complex dependencies (joins constructions, stored procedures, and other heavyweight features) from the data access layer, retaining only the essential data access functionality. This is achieved by using a simplified architecture that directly maps database tables to in-memory data structures without requiring complex SQL operations or external dependencies.
Solution Approach 2:
Instead of following the traditional approach of mapping objects to relational databases with complex joins and constraints, the patent inverts the approach by mapping database tables directly to in-memory data structures. This inversion simplifies the data access layer by eliminating the need for complex object-relational mapping and allowing direct data manipulation in memory.
2Adaptability or versatility
If traditional DAO implementations with complex dependencies are used, then comprehensive data access features are provided, but query operation speed decreases due to long latencies
Solution Approach 1:
The patent performs preliminary actions by loading data into in-memory data structures before query operations are needed. This pre-loading approach eliminates the need for complex SQL joins and database queries during actual operations, significantly improving query speed while maintaining comprehensive data access capabilities.
Solution Approach 2:
The patent creates in-memory copies of database data structures that can be manipulated directly without accessing the database. These copies include data tables, indexes, and relationships that mirror the database structure, allowing fast in-memory operations while maintaining the versatility of comprehensive data access features.
3Reliability
If data access layer accounts for considerable portion of code to handle complex dependencies, then data access functionality is maintained, but productivity decreases
Solution Approach 1:
The patent extracts and removes the considerable portion of code dedicated to handling complex dependencies, joins, and stored procedures. By using a simplified in-memory data structure approach, the data access layer code is dramatically reduced while maintaining essential data access functionality, thereby improving development productivity.
4Quantity of substance
If cloud platform uses multi-tenant database system with growing user data, then data storage capacity increases, but data access speed decreases
Solution Approach 1:
The patent introduces a new dimension by implementing an in-memory data layer above the database storage layer. This additional dimension allows data to be accessed from fast in-memory structures for frequently accessed data while maintaining the capacity to store large volumes of data in the database, thus improving access speed without sacrificing storage capacity.
Data Source
AI summary
Systems and methods provide a simple and flexible data access layer for an application. The systems and methods receive, at a data access layer associated with an application, a request for accessing data in memory; identify a pattern of the data of the request; determine whether the data access layer includes a table having a structure of one of key/value pair, sorted set, map, or event that is suitable for accessing the data of the request based on the identified pattern. If the data access layer includes a table with suitable structure, using that table for accessing the data of the request. If the data access layer does not include a table with suitable structure, generating a table having the structure of one of key/value pair, sorted set, map, or event; and using the generated table for accessing the data of the request.



