Log Agent Packet Sequencing for Reliable Transmission

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for collecting and disseminating log information from hosted services and multi-tenant systems face challenges such as data loss, duplication, and inefficiency due to network disruptions and downtime, particularly in continuous log stream transmission.

Innovation Solution

A system employing a log agent that collects and transmits log information as ordered sequences with unique identifiers, ensuring reliable transmission by resending packets upon acknowledgment and using sequence identifiers to prevent duplication, while adjusting packet size and transmission rate to match processing capacity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If log information is transmitted continuously over the network, then real-time monitoring capability is improved, but data loss and duplication occur due to network disruptions

Engineering Contradiction:
Improvelog transmission speedVSAvoiddata transmission reliability
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The log stream is divided into discrete packets, each containing a sequence identifier. This segmentation allows individual packets to be tracked and retransmitted independently if lost, resolving the contradiction by enabling continuous transmission while maintaining reliability through packet-level error handling.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Sequence identifiers are assigned to log packets before transmission. This preliminary action enables the receiving end to detect duplicates and losses by checking sequence numbers, allowing reliable reconstruction of the log stream even when network disruptions occur during continuous transmission.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If packet size is increased to reduce transmission overhead, then transmission efficiency is improved, but processing capacity is exceeded causing data loss

Engineering Contradiction:
Improvetransmission efficiencyVSAvoiddata processing reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The packet size and transmission rate are dynamically adjusted based on the processing capacity of the log service. When processing capacity is available, larger packets are transmitted to improve efficiency; when capacity is constrained, packet size and rate are reduced to prevent data loss, thus resolving the contradiction between transmission efficiency and processing reliability.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system monitors processing capacity and uses this feedback to adjust transmission parameters. This closed-loop control ensures that transmission efficiency is maximized without exceeding processing capacity, preventing data loss while maintaining high productivity.

Inventive Principle:
Principle #23Feedback

3Loss of time

If transmission rate is increased to improve real-time monitoring, then response time is reduced, but data duplication occurs during retransmission

Engineering Contradiction:
Improvelog transmission delayVSAvoidlog data duplication
Core Design Contradiction:
Loss of timeVSLoss of information

Solution Approach 1:

Sequence identifiers are embedded in each packet before transmission at high speed. This preliminary tagging enables the receiving end to quickly identify and filter duplicate packets during retransmission, allowing fast transmission without information loss.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Sequence identifiers act as disposable markers that are used once to identify packets and then discarded. This cheap mechanism allows rapid duplicate detection during high-speed transmission without consuming significant resources, resolving the contradiction between speed and information integrity.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

4Quantity of substance

If log data is retained in large files for analysis, then comprehensive system behavior tracking is improved, but file handling and viewing become difficult

Engineering Contradiction:
Improvelog data volumeVSAvoidlog file handling ease
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

Instead of retaining all log data in single large files, the system segments logs into manageable packets distributed across multiple storage locations. This segmentation maintains comprehensive log data volume while improving ease of operation by allowing selective access and processing of individual packets rather than handling entire large files.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from a single-dimension large file structure to a multi-dimensional packet-based organization. Logs are distributed across multiple files and storage locations with metadata indexing, enabling efficient retrieval and handling of specific log segments without moving or copying entire large files.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10511656B1Log information transmission integrity
Publication Date: 2019.12.17 AMAZON TECH INC
  • US10511656B1 patent drawing
  • US10511656B1 patent drawing
  • US10511656B1 patent drawing

AI summary

A system for collecting and disseminating log information may comprise a log agent and a log service. A log agent may transmit a log stream to a log service using non-overlapping adjacent subsets of log entries. Each subset may be associated with a unique and monotonically increasing sequence number, which is not incremented until the subset is received and processed by a log service.