Silicon reliability has reemerged as a very important problem in digital system design. As voltage and device dimensions shrink, combinational logic is becoming sensitive to temporary errors caused by single event upsets, transistor and interconnect aging and circuit variability. In particular, computational functional units are very challenging to protect because current redundant execution techniques have a high power and area overhead, cannot guarantee detection of some errors and cause a substantial performance degradation. As traditional worst-case design rules that guarantee error avoidance become too conservative to be practical, new microarchitectures need to be investigated to address this problem. To this end, this dissertation introduces Self-Imposed Temporal Redundancy (SITR), a speculative microarchitectural temporal redundancy technique suitable for pipelined computational functional units. SITR is able to detect most temporary errors, is area and energy-efficient and can be easily incorporated in an out-of-order microprocessor. SITR can also be used as a throttling mechanism against thermal viruses and, in some cases, allows designers to design very aggressive bypass networks capable of achieving high instruction throughput, by tolerating timing violations. To address the performance degradation caused by redundant execution, this dissertation proposes using a tiled-data ow model of computation because it enables the design of scalable, resource-rich computational substrates. Starting with the WaveScalar tiled-data flow architecture, we enhance the reliability of its datapath, including computational logic, interconnection network and storage structures. Computations are performed speculatively using SITR while traditional information redundancy techniques are used to protect data transmission and storage. Once a value has been verified, confirmation messages are transmitted to consumer instructions. Upon error detection, nullification messages are sent to the instructions affected by the error. Our experimental results demonstrate that the slowdown due to redundant computation and error recovery on the tiled-data flow machine is consistently smaller than on a superscalar von Neumann architecture. However, the number of additional messages required to support SITR execution is substantial, increasing power consumption. To reduce this overhead without significantly affecting performance, we introduce wave-based speculation, a mechanism targeted for data flow architectures that enables speculation only when it is likely to benefit performance. / text

Natural Hazards are a risk to buried gas pipeline infrastructure, but these risks are difficult to assess and quantify. This can often lead to the risks not being properly identified by pipeline owners. The risk to pipelines within landslide areas are particularly difficult to assess given the complex nature of landslide movements and the soil-pipeline interaction mechanisms imposing loads on a pipeline. This thesis research examines the relationship between ground movements and strains/stresses in buried pipelines through field measured ground movements and in-situ measured pipe strains/stresses. The pipe stresses and strains are then used to estimate probability of pipeline failure and risk based on RBDA limit states approaches. Within Manitoba Hydro’s pipeline network, three at-risk landslide areas (riverbank and deep river valleys) were selected for detailed studies. A field investigation and monitoring program was undertaken to assess possible sources of load and stresses on pipelines. Soil, ground, and pipe instrumentation were installed at the sites and monitored over a four year period. Monitoring results identified soil near the pipeline does not freeze, and ground movements at valley sites are slow moving (<50 mm/year) landslides. The monitoring results also showed pipe stresses and behaviour were affected by backfilling, changes in river levels, thermal affects, soil-pipe relaxation, and ground movements. Pipe push tests were conducted in conjunction with FEM modelling to examine pipe adhesion and to possible explain the pipe behaviour observed. Several ultimate and serviceability limit states pipe failure modes were assessed using the measured pipe stresses. Statistical analysis was undertaken to calculate the probability of pipeline failure for the various limit states failure modes and compared against limit states targets for several scenarios (backfill loads, initial stress-state of the pipeline, other pipelines within Manitoba Hydro network). Overall, the probability of failure estimates were generally insignificant or low due to a postulated soil-pipe relaxation mechanism which is causing a repeated release in longitudinal pipe stresses as the landslide continues to accumulate ongoing ground movements. Three mechanisms are presented and discussed. The statistical analysis indicate pipelines within Manitoba Hydro’s network may exceed limit states targets for yielding and local buckling depending on the loading scenario and the class of the pipeline within the landslide area. The outcome of the research was used to develop a risk managements system to examine geotechnical hazards within Manitoba Hydro’s pipeline network. Specifically, risks associated with ground movements along natural slopes and at river crossings are examined within the system. / October 2016