Dynamic Logical Zone Assignment for Data Center Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customers purchasing computing resources, such as virtual machine instances, are often unaware of the exact location and identity of the underlying data center where their instances are launched, leading to potential issues with data transfer, latency, and redundancy across multiple accounts.
Innovation Solution
A dynamic assignment technique that links multiple accounts together, allowing for the selection of a data center based on usage information and capacity, using either passive or active migration methods to reassign instances to a selected data center, ensuring optimal resource allocation and minimizing disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If logical zones are assigned to different data centers for different accounts, then each account can have customized instance placement, but account owners become unaware of the exact location and identity of the underlying data center
Solution Approach 1:
The system implements feedback by notifying account owners of the specific data center assignment when instances are launched. The notification system provides information about which data center will host the instances, allowing account owners to be aware of the exact location while still allowing the system to make optimized assignments based on current capacity and usage information
2Productivity
If instances are migrated between data centers to optimize resource allocation, then resource utilization improves, but service interruptions may occur
Solution Approach 1:
The system performs preliminary actions by evaluating data center capacity and usage information before making migration decisions. The notification system alerts account owners in advance of planned migrations, allowing them to prepare for potential service interruptions or schedule migrations during maintenance windows, thus balancing resource optimization with service continuity
Solution Approach 2:
The system implements dynamic migration strategies where the timing and execution of instance migrations are adjusted based on current system state and account preferences. Migrations can be scheduled during low-activity periods or coordinated with account owners to minimize impact, making the resource allocation process adaptive rather than static
3Extent of automation
If data center selection is automated based on usage information and capacity, then optimal resource allocation is achieved, but account owners lose control over instance location
Solution Approach 1:
The system implements a multi-functional notification and preference management system that serves multiple purposes: informing account owners of assignments, collecting account owner preferences, and allowing override of automated decisions. This universal interface maintains account owner control while enabling automated optimization for routine decisions
Data Source
AI summary
Techniques are disclosed for dynamic assignment of a logical zone to a data center. A request may be received to link together different accounts. Data center selection information may then be obtained including, for example, an identification of the data centers to which each of the linked accounts is currently assigned, linked account usage information and system capacity information. The obtained information may be used to determine a selected data center to which to assign the logical zone across each of the linked accounts. The selected data center may be determined based on one or more selection priorities, which may be set based on input from any combination of a customer, a service provider and/or another party. Any linked accounts not currently assigned to the selected data center may then have instances migrated to the selected data center using, for example, one or more passive and/or active migration techniques.