In-Memory Data Compression for Low-Latency Capacity Expansion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for memory in computer systems drives up costs despite decreasing memory prices per unit, necessitating a more efficient use of memory resources.
Innovation Solution
Implementing pooled memory systems that utilize compression techniques with low overhead and low latency, combined with non-volatile storage as a backup, to increase effective memory capacity beyond the stated capacity of individual memory appliances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data compression is applied in memory, then storage capacity is increased, but processing time and latency increase
Solution Approach 1:
The patent applies preliminary compression action by compressing data before it is written to memory storage. The compression manager proactively compresses incoming data streams and stores the compressed versions in memory, so that when data is retrieved, it is already in a compact form. This preliminary compression reduces the amount of data that needs to be processed during read operations, thereby minimizing the time penalty associated with compression/decompression cycles.
Solution Approach 2:
The system implements self-service compression by integrating the compression manager directly within the memory subsystem. The compression and decompression operations are performed automatically by the memory controller without requiring external processing resources. This self-contained approach allows compression to occur in-line with memory operations, reducing the overhead impact on processing time while maximizing storage capacity utilization.
2Speed
If compression overhead is reduced, then processing speed increases, but compression ratio decreases
Solution Approach 1:
The patent applies local quality by using different compression algorithms or parameters for different data types or access patterns. The compression manager analyzes incoming data characteristics and applies appropriate compression techniques - using higher compression ratios for data that is accessed less frequently or is less time-critical, while using faster, lower-ratio compression for data that requires quick access. This localized approach to compression quality allows the system to optimize both speed and compression ratio for different portions of the memory subsystem.
3Quantity of substance
If more memory is allocated, then storage capacity increases, but system cost increases
Solution Approach 1:
The patent fundamentally changes the parameter of data representation by storing compressed data instead of uncompressed data in memory. This parameter change allows the same physical memory capacity to store effectively more data - potentially doubling or tripling the usable storage capacity depending on data compressibility. By changing how data is encoded in memory rather than adding more physical memory modules, the system achieves increased storage capacity without the linear cost increase that would result from purchasing additional memory hardware.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An appliance is described. A write engine may process a write request from an application to write a first data into a memory. The write request may including the first data and an address. A compression engine may compress the first data to produce a first compressed data. A write module may store the first compressed data in the memory. The first data may be a first part of a page that may further include a second data as a second part. The first compressed data may be a first part of a compressed page that may further include a second compressed data as a second part.