Dynamic Containment Abstraction Binding for Cloud Workload Optimization

Solution Overview

Problem

Cloud computing systems face inefficiencies in resource allocation, leading to poor performance during workload increases and idle resources when workloads decrease, resulting in unnecessary costs for users and suboptimal resource utilization.

Innovation Solution

The system dynamically converts partially hardware-bound containment abstractions to entirely CPU-bound ones by placing them in a memory store, allocating workload across a data center to optimize resource usage, and reverting back when necessary, thereby arbitrating CPU resources and synchronizing asynchronous output for improved performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If containment abstractions are kept hardware-bound to ensure fast access and performance, then response time is improved, but resource utilization deteriorates during low workload periods

Engineering Contradiction:

Improveresponse timeVSAvoidresource utilization

Core Design Contradiction:

SpeedVSProductivity

Solution Approach 1:

The system dynamically converts containment abstractions between hardware-bound and CPU-bound states based on workload conditions. During high workload, abstractions remain hardware-bound for fast access; during low workload, they are converted to CPU-bound states and placed in memory stores, allowing hardware resources to be reused and improving overall resource utilization while maintaining acceptable performance

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The invention changes the binding parameter of containment abstractions from hardware-bound to CPU-bound based on workload metrics. This parameter change allows the same abstraction to be served from different locations (hardware device vs. memory store) depending on conditions, optimizing the trade-off between access speed and resource utilization

Inventive Principle:

Principle #35Parameter changes

2Productivity

If containment abstractions are converted to CPU-bound and placed in memory stores to improve resource utilization, then resource utilization is improved, but access speed deteriorates

Engineering Contradiction:

Improveresource utilizationVSAvoidaccess speed

Core Design Contradiction:

ProductivityVSSpeed

Solution Approach 1:

The system dynamically converts containment abstractions between hardware-bound and CPU-bound states based on workload conditions. During high workload, abstractions remain hardware-bound for fast access; during low workload, they are converted to CPU-bound states and placed in memory stores, allowing hardware resources to be reused and improving overall resource utilization while maintaining acceptable performance

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary conversion of containment abstractions to CPU-bound states and places them in memory stores during low workload periods before peak demand occurs. This allows the system to have pre-positioned abstractions ready for rapid allocation when workload increases, reducing the impact of access speed deterioration

Inventive Principle:

Principle #10Preliminary action

3Reliability

If hardware resources are allocated to containment abstractions to ensure performance during workload increases, then performance is improved, but idle resources increase when workloads decrease

Engineering Contradiction:

ImproveperformanceVSAvoididle resources

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system dynamically converts containment abstractions between hardware-bound and CPU-bound states based on workload conditions. During high workload, abstractions remain hardware-bound for fast access; during low workload, they are converted to CPU-bound states and placed in memory stores, allowing hardware resources to be reused and improving overall resource utilization while maintaining acceptable performance

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system temporarily discards hardware binding for containment abstractions during low workload periods by converting them to CPU-bound states and placing them in memory stores. This allows hardware resources to be recovered and reused for other purposes, reducing idle resources and energy waste while performance is maintained when needed

Inventive Principle:

Principle #34Discarding and recovering

Data Source

PatentUS8789050B2Systems and methods for transparently optimizing workloads

Publication Date: 2014.07.22 JOYENT INC

AI summary

Systems, methods, and media for transparently optimizing a workload of a containment abstraction are provided herein. Methods may include monitoring a workload of the containment abstraction, the containment abstraction being at least partially hardware bound, the workload corresponding to resource utilization of the containment abstraction, converting the containment abstraction from being at least partially hardware bound to being entirely central processing unit (CPU) bound by placing the containment abstraction in a memory store, based upon the workload, and allocating the workload of the containment abstraction across at least a portion of a data center to optimize the workload of the containment abstraction.