Summary form only given. The progress in data-rate and access time of the hard disk drive has not kept pace with the tremendous increase in data density. This has a deteriorating effect on the total performance of desktop computer systems. The underlying cause of this problem is that only a few heads are responsible for all data transfer. It would be better if the number of heads scales proportionally with the capacity. Probe recording systems, which use thousands of heads in parallel, might offer a solution. The disadvantage of the probe array architectures that have been suggested up to now, is that the total storage area, and therefore the capacity, is low. For areas of several tens of cm/sup 2/, the probe array would be difficult to realize and too expensive. In order to reach high capacity, we suggest to break up the array into smaller subarrays of about 10 probes each, and distribute them over a large area (a tile). This requires that the sub-arrays have actuators with which they can move over the medium surface, like miniature data storage robots, or StoBots, with dimensions of a few 100 /spl mu/m.
展开▼
机译：仅提供摘要表格。硬盘驱动器的数据速率和访问时间的进步跟不上数据密度的巨大增长。这对台式计算机系统的总体性能有恶化的影响。造成此问题的根本原因是，只有几个头负责所有数据传输。如果水头的数量与容量成正比，那会更好。并行使用数千个探头的探针记录系统可能会提供解决方案。到目前为止，已经提出的探针阵列架构的缺点是总存储区域以及容量都很低。对于几十cm / sup 2 /的面积，探针阵列将难以实现并且太昂贵。为了达到高容量，我们建议将阵列分成每个约10个探针的较小子阵列，然后将它们分布在较大的区域（一块瓷砖）。这就要求子阵列具有执行器，使它们可以在介质表面上移动，例如微型数据存储机器人或StoBots，尺寸为几百个/ spl mu / m。
展开▼
