Zone-Compartment Cloud Architecture for Workload Isolation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Monolithic cloud systems face issues with high availability, scalability, and resource management, leading to potential application failures, increased costs, and inefficient workload distribution, as they lack component isolation and efficient resource scaling.

Innovation Solution

Implementing a zone-compartment architecture with a routing engine that directs requests to appropriate compartments based on identifiers, replicates databases across zones for high availability, and uses a recommendation engine to split or merge compartments based on workload variations, allowing for independent updates and scaling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a monolithic cloud system architecture is used, then system simplicity is maintained, but high availability and fault isolation are compromised

Engineering Contradiction:

Improvehigh availabilityVSAvoidsystem architecture complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The cloud system is divided into multiple independent compartments, each containing isolated applications and databases. This segmentation allows individual compartments to fail without affecting others, achieving high availability through fault isolation while maintaining manageable complexity through modular design.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A routing engine acts as an intermediary between external requests and internal compartments. It receives incoming requests, determines the target compartment based on identifiers, and routes requests appropriately. This mediator layer abstracts the complexity of the compartmented architecture while enabling reliable request handling and failover.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If resources are scaled up for high availability, then system reliability improves, but cost increases

Engineering Contradiction:

Improvesystem reliabilityVSAvoidresource quantity

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

Databases are replicated across multiple availability zones within the compartmented architecture. Instead of over-provisioning single-zone resources, the system creates copies of databases in different zones, providing high availability through redundancy rather than excessive resource allocation in a single location.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system implements selective replication of databases across availability zones rather than duplicating entire system infrastructure. This partial action approach achieves the necessary reliability for data persistence without the excessive resource consumption of full system replication.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If compartments are kept static for simplicity, then management ease is maintained, but adaptability to workload variations is reduced

Engineering Contradiction:

Improveworkload adaptabilityVSAvoidcompartment management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The compartment architecture transitions from static to dynamic through automated splitting and merging operations. When workload increases, compartments can be split into smaller units; when workload decreases, compartments can be merged. This dynamic behavior enables adaptability to varying workloads while the automation reduces management complexity.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system monitors workload metrics and automatically adjusts compartment structure based on feedback. When performance thresholds are exceeded or resource utilization patterns change, the system triggers splitting or merging operations. This feedback mechanism enables automatic adaptation to workload variations without manual intervention, balancing adaptability with operational simplicity.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11570110B2Cloud system architecture and workload management

Publication Date: 2023.01.31 SAP SE

AI summary

Embodiments offer a cloud system allowing workload management through the use of a zone-compartment architecture. A routing engine extracts an identifier from an incoming request to a cloud system. The routing engine then determines from a mapping in a service mesh, a target compartment corresponding to the identifier. Based upon the target compartment, the routing engine receives an IP address from a registry of the service mesh. Using the IP address, the routing engine routes the request to the target compartment for processing by an application. A database of the target compartment is replicated within another compartment within a different zone, thereby ensuring high availability. Based upon a recommendation provided by the zone, the target compartment may be split or merged to accommodate variations in workflow. The mapping and registry are updated according to changes in the compartments.