Compression Accelerator Sampling Circuitry for Early Abort
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data processing systems face performance issues due to the time-consuming nature of data compression and the inefficiency of existing compression algorithms, particularly when dealing with non-compressible data, which can lead to increased latency and reduced system performance.
Innovation Solution
A hardware compression accelerator with early-abort circuitry that analyzes a subset of the source data to determine if further compression is necessary, allowing for early termination of the compression process based on predetermined parameters, thereby reducing latency and improving performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data compression is performed using conventional software tools (e.g., zswap), then data can be compressed and stored in RAM, but the compression process takes significant time and reduces system performance
Solution Approach 1:
The patent replaces software-based compression mechanisms with a hardware compression accelerator. This substitution of the compression engine from software to dedicated hardware circuitry dramatically reduces compression time while maintaining compression effectiveness, directly resolving the time-loss contradiction.
Solution Approach 2:
The patent implements an early-abort mechanism that performs preliminary analysis of data compressibility before committing to full compression. By evaluating a sample of the data first and aborting if compression is deemed ineffective, the system avoids wasting time on incompressible data, thus reducing overall compression time.
2Quantity of substance
If full compression processing is performed on all data, then maximum compression is achieved, but latency increases significantly for incompressible data
Solution Approach 1:
The early-abort mechanism performs preliminary evaluation of data compressibility by analyzing a sample portion before initiating full compression. This preliminary action allows the system to identify incompressible data early and abort processing, preventing excessive latency while maintaining compression effectiveness for compressible data.
Solution Approach 2:
Instead of always performing complete compression processing, the patent applies partial action by processing only a sample of the data to determine compressibility. This partial processing approach avoids the excessive latency of full compression on incompressible data while maintaining effective compression for suitable data.
3Productivity
If compression is performed on incompressible data, then the compression process completes, but system performance deteriorates due to wasted processing time
Solution Approach 1:
The early-abort mechanism performs preliminary assessment of data compressibility before full processing. This preliminary action identifies incompressible data early, allowing the system to abort processing and maintain high throughput by avoiding wasted processing time on unsuitable data.
Solution Approach 2:
The patent implements a discard mechanism for incompressible data detected through early evaluation. By discarding the compression operation for incompressible data rather than completing it, the system recovers processing time and maintains productivity, effectively eliminating the performance deterioration caused by processing incompressible data.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An integrated circuit includes a compression accelerator to process a request from software to compress source data into an output file. The compression accelerator includes early-abort circuitry to provide for early abort of compression operations. In particular, the compression accelerator uses a predetermined sample size to compute an estimated size for a portion of the output file. The sample size specifies how much of the source data is to be analyzed before computing the estimated size. The compression accelerator also determines whether the estimated size reflects an acceptable amount of compression, based on a predetermined early-abort threshold. The compression accelerator aborts the request if the estimated size does not reflect the acceptable amount of compression. The compression accelerator may complete the request if the estimated size reflects the acceptable amount of compression. Other embodiments are described and claimed.