Balancing Registration Server Load via Dynamic Lifetime Dispersion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The processing of user registration requests in a network can lead to server overload due to non-uniform arrivals, resulting in delays or loss of registration requests, as the initial access requests are outside the server's control and peak at different times.
Innovation Solution
A method is introduced to balance registration server load by calculating initial registration lifetimes to generate a uniform distribution of re-registration requests and assigning fixed lifetimes for re-registration requests, using the cumulative distribution function of previous registration history to manage load across epochs, while accommodating traffic spikes and dynamic user sets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a maximum prescribed lifetime is set for all users, then user access is ensured, but server load peaks occur during renewal periods
Solution Approach 1:
The patent segments the uniform maximum lifetime into variable lifetimes based on user registration history. By dividing users into different groups with different lifetime assignments, the system spreads out renewal requests over time rather than concentrating them all at once, thus reducing server load peaks while maintaining reliable user access.
Solution Approach 2:
The patent introduces dynamic lifetime assignment where the registration lifetime is not fixed but varies based on the user's registration history and arrival time. This dynamic adjustment allows the system to adaptively manage server load by assigning shorter lifetimes to users who registered recently and longer lifetimes to those who registered earlier, balancing accessibility with load management.
2Device complexity
If initial registration requests are processed with uniform lifetime, then implementation is simple, but re-registration requests create disproportionate load
Solution Approach 1:
The patent applies local quality by treating initial and re-registration requests differently. For initial requests, a simplified uniform lifetime is used. For re-registration requests, the system calculates customized lifetimes based on the user's specific registration history and arrival time, thereby addressing the local complexity only where needed to balance re-registration load.
Solution Approach 2:
The patent performs preliminary calculation of cumulative distribution function values during the registration process. By pre-calculating these values based on registration history, the system prepares the necessary data structures that enable efficient lifetime assignment for subsequent re-registration requests, reducing the computational burden during peak load periods.
3Duration of action of stationary object
If registration lifetime is extended, then user access is maintained, but server load for processing renewals increases
Solution Approach 1:
The patent changes the parameter of registration lifetime from a fixed maximum value to a variable value that depends on the user's registration history and arrival time. By dynamically adjusting this parameter, the system optimizes the balance between maintaining user access and managing server processing capacity, assigning shorter lifetimes when needed to reduce renewal load.
Data Source
AI summary
A method of balancing user registration lifetimes in a network is disclosed for smoothing the load of the registration server. The method of balancing user registration lifetimes in a network includes providing an initial registration lifetime for a registration request that is calculated using the time of the request and the previous history of registration requests, and in a particular embodiment, using the cumulative distribution function of the history of registration requests. The method of balancing user registration lifetimes in a network is particularly useful for handling spikes in registration traffic, and can accommodate user-specified variable lifetimes, as well dynamic user arrivals and departures. It provides effective load balancing over time for a single server, at the expense of a constant factor additive load per user, which can be amortized to near zero over the long run.