Fabric Controller Link Power Management for Variable Oversubscription
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers with disaggregated resources face challenges in dynamically managing bandwidth and resource allocation, requiring manual intervention for additional bandwidth, which is inefficient and inflexible in responding to changing workload needs.
Innovation Solution
Implementing a system with disaggregated resources, such as compute, memory, and storage sleds, interconnected through high-speed networks, allowing for dynamic allocation and deallocation of resources to form managed nodes, and using a fabric controller to manage network links and power states based on bandwidth demand, enabling flexible and automated resource management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If network equipment is fixed in terms of links and bandwidth, then device complexity is reduced and stability is improved, but adaptability to changing workload needs deteriorates
Solution Approach 1:
The system enables self-service through automated fabric controller that monitors bandwidth utilization and dynamically adjusts link power states without manual intervention. The fabric controller autonomously determines when to transition links between sleep and active states based on real-time telemetry data, eliminating the need for manual network reconfiguration while adapting to workload changes.
Solution Approach 2:
The patent implements dynamics by making network links dynamically adjustable between different power states (sleep and active) based on bandwidth demand. The fabric controller continuously monitors network conditions and adjusts link configurations in real-time, transforming the previously static network infrastructure into a dynamic system that adapts to changing workload requirements.
2Productivity
If additional bandwidth is allocated, then productivity and bandwidth capacity are improved, but device complexity and manual intervention requirements increase
Solution Approach 1:
The fabric controller implements self-service by automatically monitoring bandwidth utilization metrics and dynamically allocating additional bandwidth through link aggregation when demand increases. The system autonomously determines when to activate additional links or aggregate bandwidth without requiring manual configuration, thereby improving productivity while eliminating manual intervention.
Solution Approach 2:
The system employs feedback mechanisms where the fabric controller continuously collects telemetry data on bandwidth utilization and uses this information to automatically adjust network资源配置. The feedback loop enables the system to respond to increasing workload demands by dynamically allocating additional bandwidth through automated link activation or aggregation, improving productivity without manual configuration.
3Reliability
If links are kept in active state, then bandwidth availability is improved, but energy consumption increases
Solution Approach 1:
The patent applies dynamics by implementing variable link power states that transition between sleep and active modes based on real-time bandwidth demand. The fabric controller dynamically adjusts link states, keeping links active only when bandwidth is needed and transitioning them to sleep state during low-utilization periods, thereby maintaining reliability while reducing energy consumption.
Solution Approach 2:
The system changes the operational parameters of network links by introducing multiple power states (sleep and active) instead of a single fixed state. The fabric controller modifies link parameters dynamically based on bandwidth utilization metrics, enabling the system to optimize the balance between bandwidth availability and power consumption by adjusting link power states according to actual demand.
4Productivity
If manual intervention is used for bandwidth allocation, then configuration precision is improved, but productivity and response time deteriorate
Solution Approach 1:
The fabric controller implements self-service by automatically monitoring bandwidth utilization metrics and making precise allocation decisions without manual intervention. The system autonomously determines when additional bandwidth is needed and configures appropriate link aggregation or activation, maintaining allocation accuracy while dramatically improving productivity and response time to workload changes.
Solution Approach 2:
The system employs feedback mechanisms where the fabric controller continuously monitors network telemetry data and uses this information to make precise bandwidth allocation decisions. The feedback loop enables automated, accurate resource allocation that responds rapidly to changing workload demands, achieving both high productivity and precise bandwidth management without manual configuration.
Data Source
AI summary
Technologies for switch link and ply management for variable oversubscription ratios include powering up and down links of one or more network plys according to bandwidth demand, desired oversubscription ratio and/or other parameters. Telemetry data representing one or more network traffic metrics of one or more switch plies is monitored to determine respective power states of the plurality of links associated with the one or more switch plies as a function of a desired oversubscription ratio calculated based on the telemetry data. The respective power state of the plurality of links is set accordingly.