Techniques to address unreliability and variability of computing systems

As technology feature size continues to shrink, we see two challenging problems in the design of computer systems. One is the hardware unreliability due to increasing chances of transient hardware faults caused by high-energy particles and temperature hot spots. The other is the variability in the semiconductor manufacturing process, which manifests itself as a large variation in gate lengths, threshold voltage, and other parameters within a wafer and even within a die. This variability eventually impacts the frequency and the leakage power dissipation of a chip. In the first part of this thesis, we study the problem of handling I/O in memory-based check-pointing systems. The increasing demand for reliable computers has led to proposals for hardware-assisted rollback of memory state. Such approach promises major reductions in Mean Time To Repair (MTTR). The benefits are especially compelling for database servers, where existing re-covery software typically leads to downtimes of tens of minutes. Unfortunately, adoption of such proposals is hindered by the lack of efficient mechanisms for I/O recovery. We present and evaluate ReViveI/O, a scheme for I/O undo and redo that is compatible with mechanisms for hardware-assisted rollback of memory state. We have fully implemented a Linux