Solid State Drive (SSD), as new data storage media with low power consumption, high shock resistance and lightweight form, has been widely used in laptops during the past several years. The most attractive characteristic of SSD is its high random read speed because of no mechanical latency. However, poor random write performance becomes the bottle neck in wider applications. Random write is almost two orders of magnitude slower than both random read and sequential access. In our database prototype based on Oracle Berkeley DB (BDB), we propose to insert unmodified data into random write sequence in order to convert random writes into sequential writes, and then data sequence can be flushed at the speed of sequential write. Further, we optimize the write performance by reducing quantity of unmodified data to be inserted. In this demo, we visualize the progress and results when a write sequence is run on three database system. Besides this, we can select three kinds of benchmarks to run. After getting runtime of one benchmark on three database systems, we can replay their run progress in the same time. In this way, we can intuitively show the high performance of our database system.
展开▼
