Tiered Cache Topology with Anycast Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional caching methods in network communications lead to excessive requests on origin servers due to cache misses, where data centers are unaware of each other's cached resources, resulting in increased latency and resource wastage.
Innovation Solution
A tiered cache topology is determined using a data center selection algorithm that identifies proxying data centers based on measured latencies and whether the origin server's IP address is an anycast IP address, optimizing cache hits and reducing latency by configuring data centers to route requests through the most efficient proxying data centers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional caching methods are used where each data center independently caches resources, then local cache availability is improved, but excessive requests are sent to origin servers due to cache misses
Solution Approach 1:
The patent merges cache resources across multiple data centers by implementing a tiered cache topology where data centers share cached resources. When one data center caches a resource, other data centers can access it through the tiered structure, reducing duplicate cache misses and excessive requests to origin servers while maintaining local cache availability.
Solution Approach 2:
The patent introduces intermediary components including a controller that manages cache topology and proxy servers that mediate between data centers and origin servers. These intermediaries coordinate cache hits across data centers, preventing direct excessive requests to origin servers while preserving local caching benefits.
2Productivity
If data centers are configured to route requests through proxying data centers in a tiered topology, then cache hit ratios are improved, but network latency may increase due to additional routing hops
Solution Approach 1:
The patent applies local quality by implementing a tiered cache topology where data centers closer to the requester serve as lower-tier caches, providing fast local access. Only when local caches miss do requests propagate to higher-tier data centers or origin servers, ensuring that most requests are served locally with minimal latency while still achieving high cache hit ratios.
Solution Approach 2:
The patent segments the cache system into multiple tiers with different responsibilities. Lower-tier data centers handle local caching for reduced latency, while higher-tier data centers handle broader resource sharing for improved cache hit ratios. This segmentation allows the system to optimize for both speed and efficiency simultaneously.
3Productivity
If a tiered cache topology is implemented with anycast networks, then resource retrieval efficiency is improved, but system complexity increases due to anycast network configuration
Solution Approach 1:
The patent implements self-service by enabling data centers to automatically discover and utilize cached resources from other data centers through the anycast network without manual configuration. The anycast IP addresses automatically route requests to the nearest available data center with the cached resource, improving resource retrieval efficiency while minimizing system complexity through automated operation.
Data Source
AI summary
A control server receives probe data from a plurality of data centers indicating measured latencies with a first IP address associated with an origin server. The control server sums the measured latencies of a first data center having a lowest measured latency and a second data center. When the sum is below a threshold value, the control server determines the IP address to be an anycast IP address and selects a proper subset of the plurality of data centers as proxying data centers for other data centers in the plurality of data centers. When the sum is not below the threshold value, the control server determines the IP address to not be an anycast IP address and selects the first data center having the lowest measure latencies as the proxying data center for other data centers in the plurality of data centers.