The distributed systems are highly fault prone environment to work with. Thus, fault tolerance is a major concern in distributed computing. The available failure resiliency techniques mitigate only the failures that are limited in some way. However, it is extremely difficult to limit the failures because the autonomous nodes may suffer autonomous failures. Nevertheless, in mission critical applications, the uncontrolled failures are unaffordable. Hence, it is highly desirable that the system survives such failures. The survivability of computer networks is a popular research topic. However, the survivability of computing systems has not been in focus. In this illustration, we have attempted to raise this issue and bring out certain requirements and features related to survivable computing systems. To the best of my knowledge, it is the first attempt in the direction.