Blockchain Execution Trace Analysis for Cloud Service Failure Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In cloud computing environments, identifying the originating point of failure in composite services is challenging, especially when failures in one atomic service go undetected and are propagated to subsequent services, making it difficult to pinpoint the source in a tamper-proof and non-repudiable manner.
Innovation Solution
Each atomic service in a composite service is associated with a private key and a corresponding public key, with execution traces stored in a blockchain, allowing for tamper-proof and non-repudiable recording of inputs and outputs, enabling reverse analysis to determine the originating point of failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If execution traces are stored in a traditional centralized system, then the system complexity is lower, but the reliability and tamper-proof capability are reduced
Solution Approach 1:
A blockchain-based execution trace storage system is introduced as an intermediary between atomic services and failure analysis. The blockchain acts as a decentralized, immutable ledger that records execution traces from multiple atomic services, providing tamper-proof storage without requiring direct trust relationships between service providers. This resolves the contradiction by offering high reliability through cryptographic verification while maintaining manageable complexity through standardized blockchain interfaces.
Solution Approach 2:
Execution traces are copied and stored in a distributed blockchain network rather than a single centralized location. Each node in the blockchain network maintains a copy of the execution traces, ensuring data integrity and preventing tampering. This approach enhances reliability through redundancy and cryptographic hashing while the modular blockchain architecture keeps system complexity manageable through proven protocols.
2Measurement precision
If execution traces are analyzed in forward order, then the analysis process is simpler, but the ability to identify the originating failure point is reduced
Solution Approach 1:
The patent implements reverse-order analysis of execution traces stored in the blockchain. When a failure is detected in a composite service, the system analyzes execution traces from the last atomic service backward to the first, allowing rapid identification of the originating failure point. This inversion of the traditional forward-analysis approach dramatically improves failure origin identification accuracy while the automated blockchain querying process keeps analysis complexity manageable.
3Reliability
If cryptographic signing is implemented for each execution trace, then the authentication capability is improved, but the processing overhead increases
Solution Approach 1:
Atomic services perform cryptographic signing of their execution traces immediately upon completion, before any failure analysis occurs. This preliminary action ensures that authentication data is already prepared and stored in the blockchain when needed for failure analysis, eliminating delays during critical failure investigation. The signing operation is a one-time cost that enables rapid, trustworthy failure origin identification without recurring processing overhead.
Data Source
AI summary
A method is implemented by one or more network devices to identify an originating point of failure in a composite service executed in a cloud computing environment. The execution of the composite service includes execution of a plurality of atomic services in an ordered sequence, where for each atomic service that is executed, an execution trace for that atomic service is stored in a blockchain to form an ordered sequence of execution traces, where the execution trace for a given atomic service is signed using the private key associated with that atomic service and analyzing one or more of the ordered sequence of execution traces to determine which of the plurality of atomic services originated the failure, where each execution trace that is analyzed is authenticated using the public key that corresponds to the private key associated with the atomic service that generated that execution trace.