Data-intensive problems challenge conventional computing architectures with demanding CPU, memory, and I/O requirements. Using benchmarks that draw on three data types - scientific imagery, unstructured text, and semantic graphs representing networks of relationships - the authors demonstrate that emerging hardware technologies to augment traditional microprocessor-based computing systems can deliver 2 to 17 times the performance of general-purpose computers on a wide range of data-intensive applications by increasing compute cycles and bandwidth and reducing latency.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据