Dynamic binary translation using run-time feedbacks

Dynamic binary translation is the process of translating, modifying and rewriting executable (binary) code from one machine to another at run-time. This process of low-level re-engineering consists of a reverse engineering phase followed by a forward engineering phase. UQDBT, the University of Queensland Dynamic Binary Translator, is a machine-adaptable translator. Adaptability is provided through the specification of properties of machines and their instruction sets, allowing the support of different pairs of source and target machines. Most binary translators are closely bound to a pair of machines, making analyses and code hard to reuse. Like most virtual machines, UQDBT performs generic optimizations that apply to a variety of machines. Frequently executed code is translated to native code by the use of edge weight instrumentation, which makes UQDBT converge more quickly than systems based on instruction speculation. In this paper, we describe the architecture and run-time feedback optimizations performed by the UQDBT system, and provide results obtained in the x86 and SPARC® platforms.

Dynamic path profile aided recompilation in a JAVA just-in-time compiler

Just-in-Time (JIT) compilers for Java can be augmented by making use of runtime profile information to produce better quality code and hence achieve higher performance. In a JIT compilation environment, the profile information obtained can be readily exploited in the same run to aid recompilation and optimization of frequently executed (hot) methods. This paper discusses a low overhead path profiling scheme for dynamically profiling AT produced native code. The profile information is used in recompilation during a subsequent invocation of the hot method. During recompilation tree regions along the hot paths are enlarged and instruction scheduling at the superblock level is performed. We have used the open source LaTTe AT compiler framework for our implementation. Our results on a SPARC platform for SPEC JVM98 benchmarks indicate that (i) there is a significant reduction in the number of tree regions along the hot paths, and (ii) profile aided recompilation in LaTTe achieves performance comparable to that of adaptive LaTTe in spite of retranslation and profiling overheads.

A layered JavaScript virtual machine supporting dynamic instrumentation

L’observation de l’exécution d’applications JavaScript est habituellement réalisée en instrumentant une machine virtuelle (MV) industrielle ou en effectuant une traduction source-à-source ad hoc et complexe. Ce mémoire présente une alternative basée sur la superposition de machines virtuelles. Notre approche consiste à faire une traduction source-à-source d’un programme pendant son exécution pour exposer ses opérations de bas niveau au travers d’un modèle objet flexible. Ces opérations de bas niveau peuvent ensuite être redéfinies pendant l’exécution pour pouvoir en faire l’observation. Pour limiter la pénalité en performance introduite, notre approche exploite les opérations rapides originales de la MV sous-jacente, lorsque cela est possible, et applique les techniques de compilation à-la-volée dans la MV superposée. Notre implémentation, Photon, est en moyenne 19% plus rapide qu’un interprète moderne, et entre 19× et 56× plus lente en moyenne que les compilateurs à-la-volée utilisés dans les navigateurs web populaires. Ce mémoire montre donc que la superposition de machines virtuelles est une technique alternative compétitive à la modification d’un interprète moderne pour JavaScript lorsqu’appliqué à l’observation à l’exécution des opérations sur les objets et des appels de fonction.

Compilation of quality controlled nutrient profiles from the Mediterranean Sea

In the last decades, a striking amount of hydrographic data, covering the most part of Mediterranean basin, have been generated by the efforts made to characterize the oceanography and ecology of the basin. On the other side, the improvement in technologies, and the consequent perfecting of sampling and analytical techniques, provided data even more reliable than in the past. Nutrient data enter fully in this context, but suffer of the fact of having been produced by a large number of uncoordinated research programs and of being often deficient in quality control, with data bases lacking of intercalibration. In this study we present a computational procedure based on robust statistical parameters and on the physical dynamic properties of the Mediterranean sea and its morphological characteristics, to partially overcome the above limits in the existing data sets. Through a data pre filtering based on the outlier analysis, and thanks to the subsequent shape analysis, the procedure identifies the inconsistent data and for each basin area identifies a characteristic set of shapes (vertical profiles). Rejecting all the profiles that do not follow any of the spotted shapes, the procedure identifies all the reliable profiles and allows us to obtain a data set that can be considered more internally consistent than the existing ones.

Bellingshausen Sea bathymetric compilation

The southern Bellingshausen Sea (SBS) is a rapidly-changing part of West Antarctica, where oceanic and atmospheric warming has led to the recent basal melting and break-up of the Wilkins ice shelf, the dynamic thinning of fringing glaciers, and sea-ice reduction. Accurate sea-floor morphology is vital for understanding the continued effects of each process upon changes within Antarctica's ice sheets. Here we present a new bathymetric grid for the SBS compiled from shipborne echo-sounder, spot-sounding and sub-ice measurements. The 1-km grid is the most detailed compilation for the SBS to-date, revealing large cross-shelf troughs, shallow banks, and deep inner-shelf basins that continue inland of coastal ice shelves. The troughs now serve as pathways which allow warm deep water to access the ice fronts in the SBS. Our dataset highlights areas still lacking bathymetric constraint, as well as regions for further investigation, including the likely routes of palaeo-ice streams. The new compilation is a major improvement upon previous grids and will be a key dataset for incorporating into simulations of ocean circulation, ice-sheet change and history. It will also serve forecasts of ice stability and future sea-level contributions from ice loss in West Antarctica, required for the next IPCC assessment report in 2013.

Compilation of Mg/Ca ratios of Neogloboquadrina pachyderma (sinistral) and SST estimates from the northern North Atlantic

A map of estimated calcification temperatures of the planktic foraminifer Neogloboquadrina pachyderma sinistral (TNps) for the Nordic Seas and the northern North Atlantic for the Last Glacial Maximum was produced from oxygen isotopes with support of Mg/Ca ratios. To arrive at the reconstruction, several constraints concerning the plausible salinity and ?18O-fields were employed. The reconstruction indicates inflow of temperate waters in a wedge along the eastern border of the Nordic Seas and at least seasonally ice-free waters. The reconstruction from oxygen isotopes shows similarities with Mg/Ca based paleotemperatures in the southern and southeastern sector, while unrealistically high Mg/Ca values in the central Nordic Seas prevent the application of the method in this area. The oxygen isotope based reconstruction shows some agreement with temperature reconstructions based on the modern analogue technique, but with somewhat lower temperatures and a stronger internal gradient inside the Nordic Seas. All told, our results suggest a much more ice-free and dynamic high latitude ocean than the CLIMAP reconstruction.