Detecting Deadzones in Cloud Streaming VMs

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Cloud-based game streaming services face performance issues due to resource contention and deadzones in heavily virtualized environments, leading to delays and frame rate problems for users.

Innovation Solution

A method and system for identifying deadzones in a host computing device by analyzing event data from multiple virtual machines, determining the root cause, and adjusting operations such as migrating virtual machines or changing resource allocation to mitigate these issues.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If multiple virtual machines are executed in parallel on a single host to increase service capacity, then the productivity and service capacity are improved, but resource contention and deadzones occur leading to delays and frame rate problems

Engineering Contradiction:

Improveservice capacityVSAvoidframe rate consistency

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system continuously monitors pipeline stage completion timestamps and calculates frame delivery intervals to detect deadzones. This feedback mechanism identifies when resource contention causes frame rate drops, enabling dynamic adjustments to VM scheduling and resource allocation to maintain service reliability

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts VM scheduling and resource allocation based on real-time detection of deadzones and resource contention events. By making the scheduling policy adaptive rather than static, the system can respond to changing workload conditions and prevent frame rate inconsistencies while maintaining high service capacity

Inventive Principle:

Principle #15Dynamics

2Use of energy by moving object

If physical resources are shared among multiple virtual machines through multiplexing to optimize resource utilization, then the efficiency of resource use is improved, but performance issues arise due to resource contention and delays

Engineering Contradiction:

Improveresource utilization efficiencyVSAvoidframe delivery delay

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The system proactively identifies resource contention patterns and potential deadzones by analyzing pipeline stage timestamps before they cause significant frame delivery delays. By detecting issues early, the system can preemptively adjust resource allocation and VM scheduling to prevent time loss while maintaining efficient resource utilization

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The analytics engine acts as an intermediary between multiple VMs and the physical host resources, monitoring resource usage patterns and mediating resource allocation to prevent contention. This intermediary layer enables efficient resource sharing while maintaining frame delivery timing through coordinated resource management

Inventive Principle:

Principle #24Intermediary (Mediator)

3Device complexity

If heavily virtualized environments are used to consolidate multiple services on fewer hosts to reduce infrastructure complexity, then the device complexity is reduced, but deadzones and resource contention cause noticeable performance issues at the client side

Engineering Contradiction:

Improveinfrastructure complexityVSAvoidclient-side performance

Core Design Contradiction:

Device complexityVSEase of operation

Solution Approach 1:

The system implements feedback loops that monitor pipeline performance metrics and correlate them with host resource usage and events from other VMs. This feedback enables the system to identify the root causes of deadzones and adjust virtualization resource allocation to maintain client-side performance while preserving infrastructure consolidation benefits

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11498007B2Measuring and detecting idle processing periods and identifying root causes thereof in cloud-based, streaming applications

Publication Date: 2022.11.15 NVIDIA CORP

AI summary

A technique for analyzing data in order to detect issues within a cloud-based service is disclosed. Host computing devices in a data center launch virtual machines, where at least some virtual machines run a pipelined stack for a streaming service. Virtual machines in the host computing devices generate event data including timestamps. Metadata generated by the pipelined stack during each streaming session is analyzed to identify deadzones in the corresponding host computing device, and the event data is processed to identify potential root causes of the corresponding deadzones. The event data can be generated by the virtual machine hosting the streaming service or by different virtual machines on the same host computing device. A distribution of events of each event type relative to the identified deadzones is determined and an operation of the host computing device can be adjusted based on the distribution.