Network Interface Memory Region Counters for Secure Monitoring

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributed computing systems face performance issues due to hotspotting, where specific nodes become overloaded, leading to potential failure, and existing monitoring methods are either self-reporting, which can increase load, or not well-optimized, resulting in coarse-grained monitoring that is expensive and undesirable.

Innovation Solution

A network interface with host memory region counters that monitor system utilization metrics, allowing for unobtrusive and secure third-party monitoring of memory resources without engaging the host's operating system, enabling low-overhead and secure tracking of memory region activities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If self-reporting monitoring is implemented, then system utilization data can be obtained, but the load on affected nodes increases

Engineering Contradiction:
Improvesystem utilization dataVSAvoidnode load
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent introduces a network interface as an intermediary component that monitors memory region accesses without requiring the monitored nodes to actively report their state. The network interface intercepts and counts accesses to specific memory regions, providing accurate utilization data while eliminating the overhead of self-reporting mechanisms on the monitored nodes.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If coarse-grained monitoring is used, then monitoring overhead is reduced, but measurement precision deteriorates

Engineering Contradiction:
Improvemonitoring overheadVSAvoidmonitoring granularity
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent divides the monitoring scope into fine-grained memory regions, each with its own access counter in the network interface. This segmentation allows precise tracking of individual memory region utilization while maintaining low overhead, as the network interface handles all counting operations centrally without requiring complex coordination or frequent updates from multiple nodes.

Inventive Principle:
Principle #1Segmentation

3Adaptability or versatility

If third-party monitoring is implemented, then external visibility is improved, but security risks increase

Engineering Contradiction:
Improvethird-party monitoring capabilityVSAvoidsecurity risks
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The network interface serves as a secure intermediary that enables third-party monitoring without exposing the monitored system to external threats. It provides controlled access to utilization data through standardized interfaces while maintaining security boundaries, allowing authorized external entities to query monitoring data without gaining access to the underlying system resources or sensitive operations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3596602B1Unobtrusive support for third-party traffic monitoring
Publication Date: 2023.02.22 GOOGLE LLC
  • EP3596602B1 patent drawingFigure 1A
  • EP3596602B1 patent drawingFigure 1B
  • EP3596602B1 patent drawingFigure 2

AI summary

System utilization related to memory usage can be monitored by storing host memory usage information in the corresponding host physical memory. However, retrieving this information can be a high overhead operation because it involves engaging with the operating system of each host. Moreover, storing memory usage information in the host physical memories can pose a security risk if they also store privileged data. Network interfaces according to the present disclosure provide unobtrusive and secure support for monitoring of network and other system resources such as regions of memory within host physical memories. Implementations according to the present disclosure include a plurality of memory region counters stored on a network interface. Each memory region counter corresponds to one of the memory regions located in a physical memory of a host coupled to the network interface. Each of the counters includes a system utilization metric associated with its corresponding memory region.