论文部分内容阅读
本文较系统地介绍了一台大型机的高速大容量MOS 主存贮器的工程可靠性设计。从工程可靠性出发,对主存贮器的系统结构、逻辑设计、容错技术、工艺结构、元器件的测试筛选等方面做了论述;以该主存贮器实体为基础,对每个影响可靠性的基本元素(如元器件、金属化孔等)进行逻辑功能和出错机理分析;特别是利用失效率求平均无故障时间的公式,分析和计算了在有海明校验部件的系统中,产生一位错与多位错的比率,从而定量地计算出海明纠错的效率。该机运行两年多以来,证明了这些分析与实际是基本相符的。主存贮器经过几次连续考机和近二年运行,其平均无故障时间已经超过1400小时。
This article systematically introduces the engineering reliability design of a mainframe high speed and high capacity MOS main memory. Based on the reliability of the project, the system structure, logic design, fault-tolerant technology, process structure and components testing and screening of the main memory are discussed. Based on the main memory entity, The basic elements (such as components, metallized holes, etc.) for logic functions and error mechanism analysis; in particular, the use of the failure rate and mean time to failure of the formula, analysis and calculation of a system with Hamming calibration components, Produce a bit error rate and a number of dislocations, and thus quantitatively calculate the efficiency of error correction. The aircraft has been in operation for more than two years, proving that these analyzes are basically in line with the actual situation. Main memory after several consecutive test machine and nearly two years of operation, the MTBF has more than 1400 hours.