Network Traffic Manager Session Statistics Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Prior technologies lack effective methods to aggregate HTTP session statistics in a network traffic manager, leading to difficulties in determining new versus active sessions, which hampers accurate reporting and resource management.
Innovation Solution
A method within a multi-blade network traffic management device that determines the presence of cookies in HTTP requests to prepare for aggregating session statistics, including creating new cookies for new sessions and updating records for existing ones, and publishing synchronization tables to maintain accurate session tracking, thereby enabling efficient aggregation of per-session statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If HTTP session persistence is implemented across multiple blades, then accurate session tracking is improved, but system complexity increases
Solution Approach 1:
A synchronization table is introduced as an intermediary data structure stored in shared memory across all blades. This table mediates session state information between different blades, enabling them to share session tracking data without direct complex inter-process communication. The synchronization table acts as a central coordination mechanism that simplifies distributed session management.
Solution Approach 2:
The synchronization table serves multiple functions simultaneously: it stores session state information, enables session persistence across blades, provides session identification, and facilitates statistical aggregation. By making this single data structure multi-functional, the system achieves accurate session tracking without proportionally increasing complexity.
2Measurement precision
If comprehensive session statistics aggregation is performed, then reporting accuracy is improved, but processing cycles increase
Solution Approach 1:
Session statistics are aggregated and prepared in advance during the session handling process itself, rather than performing comprehensive aggregation only when reporting is needed. The synchronization table is updated continuously with session state information, so that when statistics are required, the data is already pre-organized and ready for quick aggregation across all blades.
Solution Approach 2:
Instead of maintaining complete session state copies on each blade independently, the system uses a shared synchronization table that all blades can read from. This copying approach allows statistical aggregation by having multiple blades access the same data source, reducing redundant processing while maintaining accurate reporting.
3Measurement precision
If detailed session tracking is maintained on each blade, then session identification accuracy is improved, but memory usage increases
Solution Approach 1:
Session tracking data is merged from individual blade-level data structures into a single shared synchronization table that is accessible by all blades. This consolidation eliminates redundant memory usage while maintaining accurate session identification, as each blade can access the unified session state information without requiring its own complete copy.
Solution Approach 2:
The synchronization table serves as a universal session storage mechanism that all blades can access. Rather than each blade maintaining its own session tracking data, the single multi-functional table provides session identification services to all blades, significantly reducing total memory requirements while preserving identification accuracy.
Data Source
AI summary
A method, non-transitory computer readable medium and an multi-blade network traffic manager device that assists with aggregating per-session statistics on a clustered system includes receiving a request for a HTTP transaction. Presence of a cookie within the received request is determined. One or more actions is performed based on the determination of the presence of the cookie to prepare for aggregating session statistics within a clustered system. Session statistics information is aggregated upon performing the one or more actions and completing the request for the HTTP transaction.