Storage Cluster Bandwidth Allocation Under NAT Request Redirection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge of efficiently allocating bandwidth to individual nodes in a cluster while avoiding overloading or underutilization due to unknown node-to-port mapping and unmapped nodes in data replication requests across clusters, which can exceed or fall short of cluster-wide bandwidth limits.
Innovation Solution
Implementing network throttling techniques that allocate bandwidth to nodes mapped directly to a NAT gateway and redirect requests to unmapped nodes, while tracking request handling to avoid overloading, ensuring efficient distribution across all nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data replication requests are forwarded to source cluster nodes via NAT gateway, then network bandwidth utilization is improved, but node-to-port mapping information becomes unknown and nodes may be overloaded or underutilized
Solution Approach 1:
The patent implements a feedback mechanism where the NAT gateway monitors and reports node-to-port mapping information to the replication manager. This allows the system to dynamically adjust request routing based on current mapping states, resolving the contradiction by maintaining bandwidth efficiency while eliminating mapping uncertainty through continuous information feedback.
Solution Approach 2:
The replication manager acts as an intermediary between the NAT gateway and source cluster nodes. It receives replication requests, determines appropriate node-to-port mappings using reported information, and forwards requests accordingly. This intermediary resolves the mapping complexity by centralizing the decision-making process while maintaining efficient bandwidth utilization.
2Productivity
If replication requests are distributed across all nodes, then bandwidth efficiency is improved, but individual nodes may be overloaded or underutilized
Solution Approach 1:
The patent applies local quality by assigning different roles to different nodes based on their current capacity and mapping information. The replication manager evaluates node-specific characteristics (such as available bandwidth, current load, and mapping status) and routes requests to appropriately qualified nodes, ensuring both efficient bandwidth utilization and balanced load distribution across the cluster.
Solution Approach 2:
The system dynamically adjusts request routing decisions based on real-time node status and mapping information. The replication manager continuously evaluates which nodes are currently suitable for handling replication requests and adapts its routing strategy accordingly, maintaining both bandwidth efficiency and load balance as system conditions change.
3Productivity
If cluster-wide bandwidth limits are enforced, then network resource management is improved, but individual node bandwidth allocation becomes difficult to control
Solution Approach 1:
The patent segments the cluster-wide bandwidth management into two hierarchical levels: (1) the replication manager enforces cluster-wide bandwidth limits and policies, and (2) individual nodes receive specific bandwidth allocations based on their roles and current mappings. This segmentation allows simultaneous control at both the aggregate and individual levels, resolving the contradiction between resource management and operational ease.
Data Source
AI summary
Methods, systems, and devices for data management are described. A first computing node within a first set of computing nodes may receive, from a network gateway, a first replication request that is associated with a first replication job for the first cluster of computing nodes to provide copies of a first set of one or more computing snapshots to a second cluster of computing nodes. The first set of computing nodes may determine that a second computing node different from the first computing node will service the data replication request. A network bandwidth allocation may be increased for the first computing node that received the replication request directly from the network gateway while the network bandwidth allocation for the second computing node is maintained for the second computing node based on the second computing node receiving the request via an internal redirection of the request.