A data grouping model based on cache transaction for unstructured data storage systems

Cache prefetching technology has become the mainstream data access optimization strategy in Industrial Intelligent Systems (IIS) and centers. However, rapidly increasing of unstructured generates massive pairwise relationships. Therefore, researchers have to make a choice between spatial locality temporal ensure an acceptable computational complexity. We propose cache-transaction-based grouping model (CTDGM) solve problems described above by optimizing feature representation method efficiency. First, we provide definition cache transaction for extracting (CTF). Second, design chunking algorithm based on CTF spatiotemporal optimize relationship calculation Third, CTDGM constructing relation graph that groups into independent according strength relation. Based results experiment, compared with state-of-the-art traditional methods, our achieves average increase hit rate 5%–20% MSR, VDI-LUN, KC set, which turn reduces number I/O accesses 30%–60%.