Support Vector Machine Approach to Downscale Precipitation in Climate Change Scenarios

Concern over changes in global climate has increased in recent years with improvement in understanding of atmospheric dynamics and growth in evidence of climate link to long‐term variability in hydrologic records. Climate impact studies rely on climate change information at fine spatial resolution. Towards this, the past decade has witnessed significant progress in development of downscaling models to cascade the climate information provided by General Circulation Models (GCMs) at coarse spatial resolution to the scale relevant for hydrologic studies. While a plethora of downscaling models have been applied successfully to mid‐latitude regions, a few studies are available on tropical regions where the atmosphere is known to have more complex behavior. In this paper, a support vector machine (SVM) approach is proposed for statistical downscaling to interpret climate change signals provided by GCMs over tropical regions of India. Climate variables affecting spatio‐temporal variation of precipitation at each meteorological sub‐division of India are identified. Following this, cluster analysis is applied on climate data to identify the wet and dry seasons in each year. The data pertaining to climate variables and precipitation of each meteorological sub‐division is then used to develop SVM based downscaling model for each season. Subsequently, the SVM based downscaling model is applied to future climate predictions from the second generation Coupled Global Climate Model (CGCM2) to assess the impact of climate change on hydrological inputs to the meteorological sub‐divisions. The results obtained from the SVM downscaling model are then analyzed to assess the impact of climate change on precipitation over India.

AdFT: An Adaptive Framework for Fault Tolerance on Large Scale Systems using Application Malleability

Exascale systems of the future are predicted to have mean time between failures (MTBF) of less than one hour. Malleable applications, where the number of processors on which the applications execute can be changed during executions, can make use of their malleability to better tolerate high failure rates. We present AdFT, an adaptive fault tolerance framework for long running malleable applications to maximize application performance in the presence of failures. AdFT framework includes cost models for evaluating the benefits of various fault tolerance actions including checkpointing, live-migration and rescheduling, and runtime decisions for dynamically selecting the fault tolerance actions at different points of application execution to maximize performance. Simulations with real and synthetic failure traces show that our approach outperforms existing fault tolerance mechanisms for malleable applications yielding up to 23% improvement in application performance, and is effective even for petascale systems and beyond.