Weight-Based Network Traffic Control for Cloud Performance Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies fail to effectively regulate network usage in data centers, leading to disproportionate bandwidth consumption and performance issues due to malicious or greedy tenants, resulting in unpredictable performance and denial-of-service attacks, with existing methods lacking scalable and robust network performance isolation.
Innovation Solution
Implementing a weight-based network traffic control system where each traffic sourcing entity is assigned a weight, and the bandwidth share is allocated proportionally, with a rate controller managing the transmission rate based on network feedback to ensure fair and adaptive usage, independent of specific data center topology or hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If TCP congestion control is used to manage network sharing, then network utilization is improved, but malicious or greedy tenants can achieve unbounded utilization by using many TCP flows, variations of TCP, or protocols such as UDP that do not respond to congestion control
Solution Approach 1:
The patent introduces a network controller as an intermediary component that sits between the tenants and the physical network. This controller enforces bandwidth limits and regulates network access at the tenancy level, preventing malicious tenants from achieving unbounded utilization while maintaining overall network productivity. The controller acts as a mediator that applies policy-based control to ensure fair resource distribution.
Solution Approach 2:
The system dynamically changes network parameters such as bandwidth allocation, rate limits, and traffic shaping parameters at the tenancy level. By adjusting these parameters based on tenant behavior and network conditions, the system prevents disproportionate bandwidth consumption while maintaining high network utilization when resources are available.
2Reliability
If static traffic limits are imposed on traffic sent to and from each VM, then bandwidth consumption is controlled, but malicious users can render target services unreachable by placing trojan receivers and using full rate UDP flows to overflow host server bandwidth
Solution Approach 1:
The network controller pre-configures rate limits and bandwidth constraints for each tenancy before attacks occur. These preliminary controls are enforced at the network level, preventing overflow attacks by limiting the maximum rate at which any single tenancy can transmit traffic, thereby protecting target services from being rendered unreachable.
Solution Approach 2:
The system implements feedback mechanisms that monitor network traffic patterns and tenant behavior in real-time. When abnormal patterns indicative of attacks are detected, the controller dynamically adjusts rate limits and applies additional constraints to prevent denial-of-service conditions while maintaining normal service operation.
3Productivity
If network use is allocated without effective regulation, then network capacity is maximally utilized, but tenants obtain disproportionate use of the network causing unpredictable performance
Solution Approach 1:
The network controller implements dynamic bandwidth allocation that adjusts resource distribution based on current network conditions, tenant priorities, and service level agreements. This dynamic approach allows the system to maintain high overall capacity utilization while ensuring predictable performance for each tenancy through real-time policy enforcement and adaptive rate limiting.
Data Source
AI summary
Traffic in a cloud is controlled by the nodes participating in the cloud. Tenants of the cloud each have a ratio. On any given node, a current transmission rate of the node is allocated among the tenants of the node, or more specifically, their execution units (e.g., virtual machines) on the node. Thus each tenant receives a predefined portion of the transmission capacity of the node. The transmission capacity can vary as conditions on the network change. For example, if congestion occurs, the transmission capacity may be decreased. Nonetheless, each tenant receives, according to its ratio, a same relative portion of the overall transmission capacity.