The era of petascale computing brought machines with hundreds of thousands of processors. The next generation of exascale supercomputers will make available clusters with millions of processors. In those machines, mean time between failures will range from a few minutes to few tens of minutes, making the crash of a processor the common case, instead of a rarity. Parallel applications running on those large machines will need to simultaneously survive crashes and maintain high productivity. To achieve that, fault tolerance techniques will have to go beyond checkpoint/restart, which requires all processors to roll back in case of a failure. Incorporating some form of message logging will provide a framework where only a subset of processors are rolled back after a crash. In this paper, we discuss why a simple causal message logging protocol seems a promising alternative to provide fault tolerance in large supercomputers. As opposed to pessimistic message logging, it has low latency overhead, especially in collective communication operations. Besides, it saves messages when more than one thread is running per processor. Finally, we demonstrate that a simple causal message logging protocol has a faster recovery and a low performance penalty when compared to checkpoint/restart. Running NAS Parallel Benchmarks …

We have conducted electronic-structure calculations for Am metal under pressure to investigate the behavior of the 5f-electron states. Density-functional theory (DFT) does not reproduce the experimental photoemission spectra for the ground-state phase where the 5f electrons are localized, but the theory is expected to be correct when 5f delocalization occurs under pressure. The DFT prediction is that peak structures of the 5f valence band will merge closer to the Fermi level during compression indicating presence of itinerant 5f electrons. Existence of such 5f bands is argued to be a prerequisite for the phase transitions, particularly to the primitive orthorhombic AmIV phase, but does not agree with modern dynamical-mean-field theory (DMFT) results. Our DFT model further suggests insignificant changes of the 5f valence under pressure in agreement with recent resonant x-ray emission spectroscopy, but in contradiction to the DMFT predictions. The influence of pressure on the 5f valency in the actinides is discussed and is shown to depend in a non-trivial fashion on 5f band position and occupation relative to the spd valence bands.