Dynamic Cache Allocation for Cloud IOPS Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In public cloud computing environments, tenants face performance issues due to fixed storage infrastructure capabilities, particularly in Input/Output Operations per Second (IOPS), which require procuring new volumes and migrating data when additional performance is needed, leading to inefficiencies and increased costs.

Innovation Solution

A dynamic cache allocation system that allows tenants to automatically adjust storage infrastructure cache allocation based on performance needs and cost, using a spot price mechanism that adjusts according to supply and demand, enabling programmatic increase or decrease of cache allocation through a RESTful API, with options for Absolute Price or Spot Price purchase.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If tenants procure a new volume and migrate data to achieve additional IOPS performance, then storage performance is improved, but system complexity and time loss increase

Engineering Contradiction:

ImproveIOPS performanceVSAvoiddata migration time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent implements dynamic cache allocation that allows tenants to adjust cache allocation in real-time based on performance needs. The system dynamically provisions additional cache resources without requiring data migration, enabling IOPS performance improvement while avoiding time loss associated with traditional volume procurement and data copying approaches.

Inventive Principle:

Principle #15Dynamics

2Productivity

If tenants procure a new volume and migrate data to achieve additional IOPS performance, then storage performance is improved, but device complexity increases

Engineering Contradiction:

ImproveIOPS performanceVSAvoidstorage infrastructure complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent merges the cache allocation functionality directly into the existing storage infrastructure, allowing tenants to dynamically adjust cache resources through a unified interface. This integration eliminates the need for separate volume procurement and data migration processes, reducing device complexity while maintaining improved IOPS performance.

Inventive Principle:

Principle #5Merging (Combining)

3Reliability

If fixed cache allocation is used to ensure stable performance, then reliability is improved, but adaptability worsens

Engineering Contradiction:

Improveperformance stabilityVSAvoidcache allocation flexibility

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic cache allocation that allows tenants to adjust cache resources in real-time based on changing performance requirements. This dynamic approach maintains reliability by ensuring sufficient cache is always available when needed, while simultaneously improving adaptability by allowing flexible adjustment of cache allocation without fixed constraints.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms that monitor storage performance metrics and automatically adjust cache allocation accordingly. This feedback loop ensures performance stability is maintained while adapting to changing workloads, resolving the contradiction between reliability and adaptability.

Inventive Principle:

Principle #23Feedback

4Adaptability or versatility

If dynamic cache allocation is implemented to improve adaptability, then resource utilization improves, but device complexity increases

Engineering Contradiction:

Improvecache allocation flexibilityVSAvoidcache management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a self-service cache allocation system where tenants can automatically adjust their own cache resources through programmable interfaces. This self-service approach improves resource utilization and adaptability while minimizing the complexity burden on system operators, as the system manages cache dynamics automatically based on tenant requirements.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS9992076B2Dynamic cache allocating techniques for cloud computing systems

Publication Date: 2018.06.05 CISCO TECHNOLOGY INC

AI summary

Techniques are presented herein to allow a tenant application in a cloud computing environment to automatically adjust storage infrastructure cache allocation according to performance needs and cost associated with use of the cache. Information is stored for a spot price of a cache of a storage infrastructure used for data storage in a cloud computing environment. A request is received to allocate additional cache for use by one or more applications associated with a first tenant running in the cloud computing environment. Information is sent to the one or more applications indicating a current spot price of the cache of the storage infrastructure. An instruction is received from the one or more applications to purchase the additional allocation of the cache based on the current spot price or based on a predetermined fixed price. An allocation request is sent to the storage infrastructure for additional allocation of the cache.