SaaS Historical Trending via Dedicated Database Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional SaaS applications face limitations in historical trending capabilities due to the multi-tenancy data architecture, where complex queries for one customer can adversely affect performance for others, leading to restricted access to historical data analysis.
Innovation Solution
Implementing a virtual tenant model where each business account has a dedicated server instance and database, with a historical table that periodically collects snapshots of data from database tables, allowing for complex queries and flexible data management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a multi-tenant model is used where multiple business customers share a common database, then resource utilization and cost efficiency are improved, but complex historical queries for one customer adversely affect performance for other customers
Solution Approach 1:
The patent segments the shared database into separate database instances for different customers. Each customer's data is isolated in their own database instance, allowing complex historical queries to be executed without affecting other customers' performance. This segmentation resolves the contradiction by maintaining resource efficiency through shared infrastructure while ensuring query reliability through data isolation.
Solution Approach 2:
The patent introduces an intermediary layer (separate database instances) between the shared infrastructure and individual customer queries. This intermediary allows the system to maintain the benefits of multi-tenancy while preventing performance interference, as each database instance acts as a buffer that isolates query workloads.
2Loss of information
If complex historical queries are allowed in a multi-tenant SaaS platform, then valuable business insights can be obtained, but the speed and performance of the entire database is adversely affected
Solution Approach 1:
By segmenting the database into separate instances per customer, the patent enables complex historical queries to run without interfering with other customers' operations. Each customer can execute comprehensive analytical queries on their full historical data without causing performance degradation for others, thus preserving business insights while minimizing time loss for other users.
Solution Approach 2:
The patent implements preliminary actions by pre-separating customer data into dedicated database instances before queries are executed. This proactive segmentation ensures that when complex historical queries are needed, the infrastructure is already prepared to handle them without affecting other customers, eliminating the need for query optimization workarounds.
3Reliability
If simple query limitations are imposed in multi-tenant SaaS systems, then database performance for all customers is maintained, but customers are unable to successfully leverage data collected over time to analyze historical trends
Solution Approach 1:
The patent uses segmentation to create separate database instances that enable customers to perform complex historical trend analysis without compromising overall database performance. Each customer's dedicated instance allows versatile analytical queries while maintaining system-wide reliability, as the workload is isolated to individual instances rather than affecting the shared infrastructure.
Solution Approach 2:
The patent changes the architectural parameter from a shared database model to a segmented multi-instance model. This parameter change fundamentally alters the system's capabilities, allowing customers to execute complex historical queries with full adaptability while maintaining performance reliability through the isolated instance architecture.
Data Source
AI summary
Disclosed is an improved method, system, and program product to implement a business platform that assigns a server instance and dedicated database to each business customer. A snapshot of data is periodically copied from one or more database tables to a historical table of the dedicated database. Data is retrieved in response to a query from the historical table, enabling historical trending capabilities without affecting a performance of other business customers of the business platform.