Inter-DC WAN Capacity Provisioning via Deferrable Transfer Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud providers face high costs due to the exponential growth of inter-data center WAN traffic, driven by first-party applications, leading to excessive provisioning of network bandwidth capacity, which is not efficiently managed by existing methods.
Innovation Solution
Provisioning inter-DC WAN capacity based on network failure statistics and bandwidth demands, allowing for dynamic deferral of non-critical transfers during link failures, and optimizing bandwidth usage by scheduling transfers during off-peak hours, thereby reducing the need for redundant capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If additional inter-DC WAN capacity is provisioned to handle increased first-party application traffic, then network bandwidth availability is improved, but network provisioning cost increases exponentially
Solution Approach 1:
The patent implements dynamic bandwidth allocation where deferrable transfers are paused during link failures and resumed when capacity is available. This dynamic adjustment allows the system to optimize bandwidth usage in real-time, reducing the need for static over-provisioning while maintaining service quality.
Solution Approach 2:
The system changes the operational parameters of network transfers by introducing deferrable transfers with pause/resume capabilities. This parameter change enables flexible bandwidth consumption patterns that adapt to network conditions, allowing cost-effective provisioning without sacrificing reliability.
2Reliability
If redundant network capacity is provisioned to ensure reliability during link failures, then network reliability is improved, but device complexity and provisioning overhead increase
Solution Approach 1:
The patent extracts deferrable transfers from the critical path of network operations, separating them from immediate transfers that require guaranteed bandwidth. This extraction allows the system to maintain reliability for critical operations while using a simpler, more cost-effective approach for non-critical deferrable traffic.
Solution Approach 2:
The system introduces an intermediary mechanism (the deferrable transfer management layer) that coordinates between network capacity availability and application requirements. This intermediary optimizes resource allocation by pausing and resuming transfers based on real-time network conditions, reducing the need for complex redundant provisioning.
3Productivity
If all transfers are performed immediately to meet application demands, then application performance is improved, but peak bandwidth demand increases requiring excessive provisioning
Solution Approach 1:
The patent implements periodic action by scheduling deferrable transfers during off-peak hours when network capacity is underutilized. This periodic execution pattern smooths bandwidth demand across time, allowing immediate transfers to complete without contention while deferrable transfers utilize otherwise idle capacity, eliminating the need for excessive peak provisioning.
Solution Approach 2:
The system performs preliminary scheduling of deferrable transfers during low-demand periods before peak demand occurs. This preliminary action optimizes bandwidth utilization by pre-positioning data transfers during off-peak hours, reducing peak demand requirements while ensuring application performance is maintained.
Data Source
AI summary
The techniques disclosed herein provision inter-DC WAN capacity based on network failure statistics and bandwidth demands of a cloud-hosted application. Network capacity is provisioned based on an assumption of runtime cooperation between the application and the network. For example, if the network detects that a link has failed, the application may cooperate with the network to pause a deferrable transfer, reserving bandwidth for non-deferrable transfers. With knowledge that deferrable transfers will be dynamically paused when a primary link fails, backup links may be provisioned with less capacity than the primary link. The ability to dynamically defer transfers also enables a greater degree of bandwidth smoothing, e.g. reducing peak demand by scheduling deferrable transfers for off-peak hours. This allows network links to be provisioned with less capacity than if all transfers were performed immediately.