Search for: All records

The performance of compute hardware varies: software run repeatedly on the same server (or a different server with supposedly identical parts) can produce performance results that differ with each execution. This variation has important effects on the reproducibility of systems research and ability to quantitatively compare the performance of different systems. It also has implications for commercial computing, where agreements are often made conditioned on meeting specific performance targets. Over a period of 10 months, we conducted a large-scale study capturing nearly 900,000 data points from 835 servers. We examine this data from two perspectives: that of a service provider wishing to offer a consistent environment, and that of a systems researcher who must understand how variability impacts experimental results. From this examination, we draw a number of lessons about the types and magnitudes of performance variability and the effects on confidence in experiment results. We also create a statistical model that can be used to understand how representative an individual server is of the general population. The full dataset and our analysis tools are publicly available, and we have built a system to interactively explore the data and make recommendations for experiment parameters based on statistical analysis of historical data. 

The composition of bioretention soil media (BSM) is among the most critical design attributes contributing to the water quality performance of bioretention systems, as various amendments may increase the capacity for chemical sorption of certain nutrient pollutants. We investigated the spent lime (a calcium-based water treatment residual) as BSM amendments for nutrient retention. The study was conducted in two parts: the first was a field-based mesocosm experiment in which we assessed the effect of spent lime amendments on leachate nutrient concentration for treatments receiving different levels of phosphorus and nitrogen loading (simulated by different levels of compost added to the substrate). The second was a laboratory study comparing various levels of spent lime and coir on leachate nutrient concentration at two different simulated loading rates. Effluent water was collected and analyzed for PO43−, NH4+ and NO3− concentrations in the field and lab. Spent lime significantly reduced leachate PO43− concentrations (upwards of 50%) in both the field and lab mesocosm studies compared to treatments without spent lime. Reductions in NH4+ concentrations were also observed due to spent lime but with variable significance across the different compost levels, whereas NO3− concentrations were higher in plots with spent lime than plots without spent lime. In the lab, columns with coir had significantly higher leachate PO43− concentrations compared to spent lime-treated columns, however, leachate NH4+ and NO3− concentrations did not significantly differ between treatments at the same compost levels. This study shows that spent lime, which is a waste product, is effective in significantly reducing leachate PO43− concentrations from BSM, while be a cost-effective substitute to engineered proprietary media that is expensive to acquire; however, future studies must also evaluate its potential for clogging.