With the ever-growing need of data in HPC applications, the congestion at the I/O level becomes critical in super-computers. Architectural enhancement such as burst-buffers and pre-fetching are added to machines, but are not sufficient to prevent congestion. Recent online I/O scheduling strategies have been put in place, but they add an additional congestion point and overheads in the computation of applications.In this work, we show how to take advantage of the periodic nature of HPC applications in order to develop efficient periodic scheduling strategies for their I/O transfers. Our strategy computes once during the job scheduling phase a pattern where it defines the I/O behavior for each application, after which the applications run independently, transferring their I/O at the specified times. Our strategy limits the amount of I/O congestion at the I/O node level and can be easily integrated into current job schedulers. We validate this model through extensive simulations and experiments by comparing it to state-of-the-art online solutions, showing that not only our scheduler has the advantage of being de-centralized and thus overcoming the overhead of online schedulers, but also that it performs better than these solutions, improving the application dilation up to 13% and the maximum system efficiency up to 18%.

We provide a framework to analyze multi-level checkpointing protocols, by formally defining a k-level checkpointing pattern. We provide a first-order approximation to the optimal checkpointing period, and show that the corresponding overhead is in the order of k =1 √ 2λ C , where λ is the error rate at level , and C the checkpointing cost at level. This nicely extends the classical Young/Daly formula on single-level checkpointing. Furthermore, we are able to fully characterize the shape of the optimal pattern (number and positions of checkpoints), and we provide a dynamic programming algorithm to determine the optimal subset of levels to be used. Finally, we perform simulations to check the accuracy of the theoretical study and to confirm the optimality of the subset of levels returned by the dynamic programming algorithm. The results nicely corroborate the theoretical study, and demonstrate the usefulness of multi-level checkpointing with the optimal subset of levels.

We study the resilient scheduling of moldable parallel jobs on highperformance computing (HPC) platforms. Moldable jobs allow for choosing a processor allocation before execution, and their execution time obeys various speedup models. The objective is to minimize the overall completion time of the jobs, or the makespan, when jobs can fail due to silent errors and hence may need to be re-executed after each failure until successful completion. Our work generalizes the classical scheduling framework for failure-free jobs. To cope with silent errors, we introduce two resilient scheduling algorithms, Lpa-List and Batch-List, both of which use the List strategy to schedule the jobs. Without knowing a priori how many times each job will fail, Lpa-List relies on a local strategy to allocate processors to the jobs, while Batch-List schedules the jobs in batches and allows only a restricted number of failures per job in each batch. We prove new approximation ratios for the two algorithms under several prominent speedup models (e.g., roofline, communication, Amdahl, power, monotonic, and a mixed model). An extensive set of simulations is conducted to evaluate different variants of the two algorithms, and the results show that they consistently outperform some baseline heuristics. Overall, our best algorithm is within a factor of 1.6 of a lower bound on average over the entire set of experiments, and within a factor of 4.2 in the worst case.

-Conservation Agriculture (CA) has been promoted during the last 15 years in Madagascar to develop a sustainable rainfed agriculture in order to cope with low fertility upland soils, soil erosion, low productivity and erratic rainfall. If CA does provide a better sustainability with adapted cropping patterns when adopted, a question is, whether CA is able to alleviate the impact of climatic variations thanks to the mulching effect and to which extent. We took the example of medium altitude zones in the Lake Alaotra area to illustrate farmers' perceptions. If climate change is not proven in these areas, climatic variations are very high and erratic rainfall patterns at the beginning of the rainy season induce serious risks of crop failure. Two surveys were implemented in 2013: (i) on CA changing practices of 92 farmers (during the 10 years project duration) and (ii) on the evolution of behavior among 28 farmers who have recently adopted CA (less than 5 years). In CA systems, the mulch contributes to better cropping systems resilience and helps alleviating the effect of climatic variations. If a limited number of farmers have effectively adopted CA practices, in the long run, most farmers have developed innovative cropping systems between conventional agriculture and CA, to address their own constraints, leading to a widespread range of agro-ecological practices. A typology of behavior is presented to identify farmers' strategies regarding risks in a context characterized by multiple uncertainties.Keywords: conservation agriculture / climatic impact alleviation / adaptation to climate change / Lake Alaotra / Madagascar Résumé -L'agriculture de conservation peut-elle réduire l'impact des variations climatiques à Madagascar ? Perceptions paysannes. L'agriculture de conservation (AC) a été promue au cours des 15 dernières années à Madagascar pour développer une agriculture pluviale durable sous fortes contraintes : sols à faible fertilité, érosion des sols, faible productivité et pluies irrégulières. Si l'AC offre une meilleure durabilité avec des modes de culture adaptés, la question est de savoir si elle peut réduire l'impact des variations climatiques grâce à l'effet de paillage (mulch). Nous avons pris l'exemple des zones d'altitude moyenne dans la région du lac Alaotra pour illustrer les perceptions des agriculteurs. Si les changements climatiques ne sont pas prouvés dans ces zones, les variations climatiques sont très importantes et les précipitations irrégulières au début de la saison des pluies induisent de forts risques de mauvaises récoltes. Deux enquêtes ont été mises en oeuvre en 2013 : sur l'évolution des pratiques d'AC de 92 agriculteurs (sur la durée du projet de 10 ans) et sur l'évolution du comportement de 28 agriculteurs qui ont récemment adopté l'AC (moins de 5 ans). Dans les systèmes en AC, le mulch contribue à améliorer la résilience des systèmes de culture et aide à réduire l'effet des variations climatiques. Si un nombre limité d'agriculteurs ont effectivement adopté des pratiques en A...