Private Service Instances on Single-Tenant Hardware

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In large-scale data centers, providing scalable and isolated computing and storage resources to customers while maintaining regulatory, security, and performance requirements is challenging due to the sharing of physical resources among multiple tenants, which can compromise security and performance.

Innovation Solution

Implementing a service provider network that allows customers to request private instances of virtualization services on single-tenant hardware, with a private API endpoint, enabling physical isolation and scalable resource allocation based on demand, using IP tunneling technology for network management and virtualization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If physical computing resources are shared among multiple tenants to improve resource utilization and reduce costs, then productivity and cost efficiency are improved, but security and performance isolation are compromised

Engineering Contradiction:
Improveresource utilizationVSAvoidsecurity isolation
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent segments physical computing resources into multiple virtual machines, each isolated and dedicated to a specific tenant. Virtualization technology divides physical hardware (servers, storage, networking) into separate virtual instances, ensuring that each tenant's workload runs in an isolated environment while sharing the underlying physical infrastructure. This resolves the contradiction by maintaining security isolation through virtual boundaries while enabling resource sharing for improved productivity.

Inventive Principle:
Principle #1Segmentation

2Productivity

If virtualization technologies are used to share computing resources among multiple customers, then resource efficiency is improved, but managing and provisioning physical computing resources becomes increasingly complicated

Engineering Contradiction:
Improveresource efficiencyVSAvoidprovisioning complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements self-service provisioning capabilities that allow tenants to automatically request, allocate, and manage their own virtual computing resources through automated workflows. The system provides self-contained virtual machines with pre-configured resources, enabling tenants to provision infrastructure without manual intervention from service providers. This reduces provisioning complexity while maintaining high resource efficiency through automated virtualization management.

Inventive Principle:
Principle #25Self-service

3Reliability

If single-tenant hardware is provided to ensure security and performance isolation, then reliability is improved, but resource utilization and cost efficiency deteriorate

Engineering Contradiction:
Improveperformance isolationVSAvoidresource utilization
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent merges multiple virtual machines from different tenants onto shared physical hardware infrastructure. By consolidating workloads from multiple tenants onto the same physical servers, storage systems, and networking equipment, the system achieves high resource utilization while maintaining performance isolation through virtualization. Each tenant experiences dedicated performance characteristics despite sharing physical resources with others, resolving the contradiction between reliability and productivity.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS9397909B2Methods and apparatus for scalable private services
Publication Date: 2016.07.19 AMAZON TECH INC
  • US9397909B2 patent drawing
  • US9397909B2 patent drawing
  • US9397909B2 patent drawing

AI summary

Methods and apparatus for providing scalable private services in service provider networking environments. A service provider that provides a large, public, multi-tenant implementation of a web service to multiple customers via a public API endpoint may allow a customer to request the establishment of a private implementation of the service. In response, a service private instance may be automatically and/or manually established for the customer that provides a private API endpoint to the service and that is at least in part implemented on single-tenant hardware that is not shared with other customers. The service private instance may initially be implemented as a relatively small scale and possibly limited implementation of the service when compared to the service public instance. As the needs of the customer grow, the service private instance may be automatically and/or manually scaled up from the initial implementation.