| 摘要 |
<p>Data compression logic of the invention compresses data from a host to a connected cache memory for storage within a storage device. A compressor within the logic implements a compression algorithm of the data; and a plurality of Chunk RAMs store compressed and uncompressed data chunks from the compressor. Predictive digital logic assesses memory usage within the RAMs while uncompressed and compressed chunks are stored within the RAMs. The predictive digital logic selects one RAM to drain its data into the cache memory based upon the compression efficiency of the compressed data as compared to the uncompressed data and prior to the complete loading of the chunks within the RAMs, resulting in reduced latency in writing compressed data to the connected cache memory. The predictive digital logic can include calculation logic for dynamically calculating compression efficiency, used in selecting which RAM to drain.</p> |
