Memory Vulnerability Evaluation Using Liveness Factor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining architectural vulnerability factor (AVF) for memory to soft errors are impractical due to the need for extensive fault injection simulations, which are time-consuming and resource-intensive, especially in large designs with many memory structures.
Innovation Solution
The method employs liveness as a factor to evaluate memory vulnerability to soft errors by setting counters for each memory entry, simulating write and read operations over a preconfigured time window, and calculating a liveness factor representing the vulnerability, which significantly reduces computation resources and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fault injection simulations are used to determine AVF for memory, then measurement precision of memory vulnerability is improved, but loss of time and computational resources increase significantly
Solution Approach 1:
The patent changes the measurement parameter from traditional AVF (architectural vulnerability factor) to Liveness Factor, which measures the probability that data in memory is read after being written. This parameter transformation enables faster simulation by focusing on actual data access patterns rather than comprehensive fault injection, thereby reducing simulation time while maintaining measurement relevance for soft error vulnerability assessment
Solution Approach 2:
The patent extracts the essential characteristic of memory vulnerability related to soft errors, which is the liveness of data (whether data is written and subsequently read), and separates this from the complex fault injection process. By taking out only the critical liveness information, the system can determine vulnerability without performing exhaustive fault simulations, thus reducing computational time and resources
2Measurement precision
If fault injection simulations are used to determine AVF for memory, then measurement precision of memory vulnerability is improved, but use of energy and computational resources increase significantly
Solution Approach 1:
The patent transforms the vulnerability measurement approach by using Liveness Factor instead of traditional AVF. The liveness factor is calculated by tracking write and read operations in the memory hierarchy, which requires significantly less computational energy than comprehensive fault injection simulations. This parameter change maintains the ability to assess soft error vulnerability while dramatically reducing energy consumption
Solution Approach 2:
The patent extracts only the essential liveness information from memory operations (whether data is written and subsequently read) and uses this extracted information to assess vulnerability. By taking out only the critical liveness characteristic rather than performing complete fault simulations, the system achieves accurate vulnerability measurement with much lower computational energy requirements
3Measurement precision
If comprehensive fault injection is performed for all memory structures, then measurement precision is improved, but device complexity and simulation setup complexity increase
Solution Approach 1:
The patent extracts the essential liveness characteristic from complex memory operations and uses this simplified metric to assess vulnerability. By focusing only on whether data is written and subsequently read, the system avoids the complexity of comprehensive fault injection setups while maintaining measurement precision for soft error vulnerability assessment
Solution Approach 2:
The patent segments the memory access analysis into discrete liveness intervals, tracking individual write-read sequences rather than performing comprehensive fault simulations across all memory structures. This segmentation approach simplifies the simulation setup and execution while maintaining the ability to measure vulnerability precision for each memory location
Data Source
AI summary
Memory, used by a computer to store data, is generally prone to faults, including permanent faults (i.e. relating to a lifetime of the memory hardware), and also transient faults (i.e. relating to some external cause) which are otherwise known as soft errors. Since soft errors can change the state of the data in the memory and thus cause errors in applications reading and processing the data, there is a desire to characterize the degree of vulnerability of the memory to soft errors. In particular, once the vulnerability for a particular memory to soft errors has been characterized, cost/reliability trade-offs can be determined, or soft error detection mechanisms (e.g. parity) may be selectively employed for the memory. In some cases, memory faults can be diagnosed by redundant execution and a diagnostic coverage may be determined.