Server Load Distribution for Unanticipated Web Service Overloads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing server management configurations fail to effectively handle unexpected overloads in web-based services, leading to service degradation and inefficiencies in resource allocation, particularly when multiple applications share resources and have varying loads.
Innovation Solution
A system and method for distributing load across multiple shared resources, allowing partial but not full overlap between applications, with autonomous provisioning of resources based on measured consumption to maintain peak headroom and minimize exposure to overloads, while throttling processing steps to ensure predictable response times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If multiple applications share the same resources, then resource efficiency is improved, but one application's unexpected overload can impact other applications
Solution Approach 1:
The patent divides the shared resource pool into application-specific resource subsets. Each application is assigned a portion of the resources, and the system tracks which resources are dedicated to which applications. When an application experiences unexpected overload, only its specific resource subset is isolated, allowing other applications to continue using their allocated resources without interruption. This segmentation resolves the contradiction by maintaining resource sharing efficiency while preventing cascade failures.
2Reliability
If dedicated resources are provided to each application, then service reliability is improved, but resource inefficiency increases during non-peak periods
Solution Approach 1:
The patent implements dynamic resource allocation where the system continuously monitors application load and adjusts resource assignments in real-time. During normal operation, applications share resources efficiently. When an application experiences a peak load or unexpected overload, the system dynamically reassigns resources to ensure that application gets the necessary capacity while other applications maintain their service levels. This dynamic adjustment resolves the contradiction by providing reliability only when needed rather than permanently dedicating resources.
3Loss of energy
If resources are shared to level peaks of normal variable loads, then resource efficiency is improved, but the system cannot handle unexpected overloads
Solution Approach 1:
The patent implements beforehand cushioning by pre-allocating resource buffers and establishing isolation mechanisms before unexpected overloads occur. The system maintains a pool of resources that can be rapidly reallocated and creates resource boundaries that prevent unexpected overloads in one application from propagating to other applications. This cushioning approach allows the system to handle unexpected overloads while maintaining overall resource efficiency during normal operation.
Data Source
AI summary
A system and method is disclosed for allocating servers across a large number of applications and for providing a predictable and consistent service response under conditions where the use of the service and associated loads is driven by unknown factors. The invention provides fault tolerance within an application through multiple resources per application and fault tolerance across applications by limiting the overlap in resources between applications. The computational load on the service may include both individual processing time due to the complexity of a single request and the number of requests. Complexity may be unpredictable because the service is self-provisioned and may allow service users to create an arbitrary sequence of compound processing steps. The number of requests may vary due to a variety of events, including daily, seasonal, or holidays, or factors driven more directly by the user of the service, such as sales, advertising, or promotions. The invention throttles server loads to provide managed degradation of application processing. The system has application in personalization, behavioral targeting, Internet retailing, personalized search, email segmentation and ad targeting, to name but a few applications.