General-Purpose Computing on Graphic Processing Unit (GPGPU): metodi e tecnologie per lo sviluppo di applicazioni

Microprocessori basati su singolo processore (CPU), hanno visto una rapida crescita di performances ed un abbattimento dei costi per circa venti anni. Questi microprocessori hanno portato una potenza di calcolo nell’ordine del GFLOPS (Giga Floating Point Operation per Second) sui PC Desktop e centinaia di GFLOPS su clusters di server. Questa ascesa ha portato nuove funzionalità nei programmi, migliori interfacce utente e tanti altri vantaggi. Tuttavia questa crescita ha subito un brusco rallentamento nel 2003 a causa di consumi energetici sempre più elevati e problemi di dissipazione termica, che hanno impedito incrementi di frequenza di clock. I limiti fisici del silicio erano sempre più vicini. Per ovviare al problema i produttori di CPU (Central Processing Unit) hanno iniziato a progettare microprocessori multicore, scelta che ha avuto un impatto notevole sulla comunità degli sviluppatori, abituati a considerare il software come una serie di comandi sequenziali. Quindi i programmi che avevano sempre giovato di miglioramenti di prestazioni ad ogni nuova generazione di CPU, non hanno avuto incrementi di performance, in quanto essendo eseguiti su un solo core, non beneficiavano dell’intera potenza della CPU. Per sfruttare appieno la potenza delle nuove CPU la programmazione concorrente, precedentemente utilizzata solo su sistemi costosi o supercomputers, è diventata una pratica sempre più utilizzata dagli sviluppatori. Allo stesso tempo, l’industria videoludica ha conquistato una fetta di mercato notevole: solo nel 2013 verranno spesi quasi 100 miliardi di dollari fra hardware e software dedicati al gaming. Le software houses impegnate nello sviluppo di videogames, per rendere i loro titoli più accattivanti, puntano su motori grafici sempre più potenti e spesso scarsamente ottimizzati, rendendoli estremamente esosi in termini di performance. Per questo motivo i produttori di GPU (Graphic Processing Unit), specialmente nell’ultimo decennio, hanno dato vita ad una vera e propria rincorsa alle performances che li ha portati ad ottenere dei prodotti con capacità di calcolo vertiginose. Ma al contrario delle CPU che agli inizi del 2000 intrapresero la strada del multicore per continuare a favorire programmi sequenziali, le GPU sono diventate manycore, ovvero con centinaia e centinaia di piccoli cores che eseguono calcoli in parallelo. Questa immensa capacità di calcolo può essere utilizzata in altri campi applicativi? La risposta è si e l’obiettivo di questa tesi è proprio quello di constatare allo stato attuale, in che modo e con quale efficienza pùo un software generico, avvalersi dell’utilizzo della GPU invece della CPU.

Messa a punto e validazione di metodiche analitiche per la determinazione di micotossine in matrici biologiche mediante UPLC-MS/MS

La presenza di micotossine nelle materie prime desta grande preoccupazione a causa delle importanti implicazioni nella sicurezza di alimenti e mangimi. Lo scopo di questo lavoro è stato quello di mettere a punto e validare una metodica analitica rapida e semplice, in cromatografia liquida ad ultra prestazione accoppiata a spettrometria di massa-tandem (UPLC-MS/MS), per la determinazione simultanea di differenti micotossine: aflatossine (B1, B2, G1, G2), ocratossina A, fumonisine (B1, B2), deossinivalenolo e zearalenone in matrici biologiche. Il metodo sviluppato per l’analisi di campioni di mangime secco per cani ha mostrato prestazioni adeguate ed è stato applicato a 49 campioni reperibili in commercio, al fine di valutare la sua efficacia e di ottenere alcuni dati preliminari sulla contaminazione da micotossine in alimenti per cani disponibili sul mercato italiano. Lo studio ha evidenziato una percentuale alta di campioni positivi, contenenti principalmente fumonisine, deossinivalenolo e ocratossina A; tutti i tenori si sono dimostrati inferiori al limite di legge previsto (Racc. CE 576/2006). Una seconda metodica è stata messa a punto e validata per l’identificazione e la quantificazione micotossine in campioni di formaggio; per questa matrice è stata inserita anche l’aflatossina M1, specifica dei prodotti lattiero - caseari. Le differenti proprietà chimico-fisiche degli analiti e la complessità della matrice hanno implicato alcune difficoltà nello sviluppo della metodica. Tuttavia, il metodo validato si è mostrato rapido, semplice ed affidabile ed è stato applicato a diversi tipi di formaggi per verificarne la versatilità. I risultati preliminari hanno mostrato l’assenza di contaminazione da parte delle micotossine in oggetto. Entrambi i metodi si sono dimostrati utili per il monitoraggio di contaminanti in matrici complesse ad oggi ancora poco studiate.

Measurement of the inclusive jet cross section in proton-proton collisions at root out s = 7 TeV with the ATLAS detector

The dominant process in hard proton-proton collisions is the production of hadronic jets.rnThese sprays of particles are produced by colored partons, which are struck out of their confinement within the proton.rnPrevious measurements of inclusive jet cross sections have provided valuable information for the determination of parton density functions and allow for stringent tests of perturbative QCD at the highest accessible energies.rnrnThis thesis will present a measurement of inclusive jet cross sections in proton-proton collisions using the ATLAS detector at the LHC at a center-of-mass energy of 7 TeV.rnJets are identified using the anti-kt algorithm and jet radii of R=0.6 and R=0.4.rnThey are calibrated using a dedicated pT and eta dependent jet calibration scheme.rnThe cross sections are measured for 40 GeV < pT <= 1 TeV and |y| < 2.8 in four bins of absolute rapidity, using data recorded in 2010 corresponding to an integrated luminosity of 3 pb^-1.rnThe data is fully corrected for detector effects and compared to theoretical predictions calculated at next-to-leading order including non-perturbative effects.rnThe theoretical predictions are found to agree with data within the experimental and theoretic uncertainties.rnrnThe ratio of cross sections for R=0.4 and R=0.6 is measured, exploiting the significant correlations of the systematic uncertainties, and is compared to recently developed theoretical predictions.rnThe underlying event can be characterized by the amount of transverse momentum per unit rapidity and azimuth, called rhoue.rnUsing analytical approaches to the calculation of non-perturbative corrections to jets, rhoue at the LHC is estimated using the ratio measurement.rnA feasibility study of a combined measurement of rhoue and the average strong coupling in the non-perturbative regime alpha_0 is presented and proposals for future jet measurements at the LHC are made.

A stereological comparison of villous and microvillous surfaces in small intestines of frugivorous and entomophagous bats: species, inter-individual and craniocaudal differences

The extents of functional surfaces (villi, microvilli) have been estimated at different longitudinal sites, and in the entire small intestine, for three species of bats belonging to two feeding groups: insect- and fruit-eaters. In all species, surface areas and other structural quantities tended to be greatest at more cranial sites and to decline caudally. The entomophagous bat (Miniopterus inflatus) had a mean body mass (coefficient of variation) of 8.9 g (5%) and a mean intestinal length of 20 cm (6%). The surface area of the basic intestinal tube (primary mucosa) was 9.1 cm2 (10%) but this was amplified to 48 cm2 (13%) by villi and to 0.13 m2 (20%) by microvilli. The total number of microvilli per intestine was 4 x 10(11) (20%). The average microvillus had a diameter of 8 nm (10%), a length of 1.1 microns (22%) and a membrane surface area of 0.32 micron 2 (31%). In two species of fruit bats (Epomophorus wahlbergi and Lisonycteris angolensis), body masses were greater and intestines longer, the values being 76.0 g (18%) and 76.9 g (4%), and 73 cm (16%) and 72 cm (7%), respectively. Surface areas were also greater, amounting to 76 cm2 (26%) and 45 cm2 (8%) for the primary mucosa, 547 cm2 (29%) and 314 cm2 (16%) for villi and 2.7 m2 (23%) and 1.5 m2 (18%) for microvilli. An increase in the number of microvilli, 33 x 10(11) (19%) and 15 x 10(11) (24%) per intestine, contributed to the more extensive surface area but there were concomitant changes in the dimensions of microvilli. Mean diameters were 94 nm (8%) and 111 nm (4%), and mean lengths were 2.8 microns (12%) and 2.9 microns (10%), respectively. Thus, an increase in the surface area of the average microvillus to 0.83 micron 2 (12%) and 1.02 microns 2 (11%) also contributed to the greater total surface area of microvilli. The lifestyle-related differences in total microvillous surface areas persisted when structural quantities were normalised for the differences in body masses. The values for total microvillous surface area were 148 cm2g-1 (20%) in the entomophagous bat, 355 cm2g-1 (20%) in E. wahlbergi and 192 cm2g-1 (17%) in L. angolensis. This was true despite the fact that the insecteater possessed a greater length of intestine per unit of body mass: 22 mm g-1 (8%) versus 9-10 mm g-1 (9-10%) for the fruit-eaters.

Stalagmite water content as a proxy for drip water supply in tropical and subtropical areas

In this pilot study water was extracted from samples of two Holocene stalagmites from Socotra Island, Yemen, and one Eemian stalagmite from southern continental Yemen. The amount of water extracted per unit mass of stalagmite rock, termed "water yield" hereafter, serves as a measure of its total water content. Based on direct correlation plots of water yields and δ18Ocalcite and on regime shift analyses, we demonstrate that for the studied stalagmites the water yield records vary systematically with the corresponding oxygen isotopic compositions of the calcite (δ18Ocalcite). Within each stalagmite lower δ18Ocalcite values are accompanied by lower water yields and vice versa. The δ18Ocalcite records of the studied stalagmites have previously been interpreted to predominantly reflect the amount of rainfall in the area; thus, water yields can be linked to drip water supply. Higher, and therefore more continuous drip water supply caused by higher rainfall rates, supports homogeneous deposition of calcite with low porosity and therefore a small fraction of water-filled inclusions, resulting in low water yields of the respective samples. A reduction of drip water supply fosters irregular growth of calcite with higher porosity, leading to an increase of the fraction of water-filled inclusions and thus higher water yields. The results are consistent with the literature on stalagmite growth and supported by optical inspection of thin sections of our samples. We propose that for a stalagmite from a dry tropical or subtropical area, its water yield record represents a novel paleo-climate proxy recording changes in drip water supply, which can in turn be interpreted in terms of associated rainfall rates.

n-fatty acids of shale partings in the Franciscan bedded cherts

Normal saturated fatty acid (n-fatty acid) in marine sediments from coastal and pelagic environments were analyzed. The coastal sediments contain both short-chained n-fatty acids with carbon numbers from 12 to 18 and long-chained acids from 22 to 32, whereas the pelagic sediments contain predominantly short-chained acids. The relative abundance of short-chained to long-chained n-fatty acids, expressed by the molar ratio C16/C26, can be an indicator to assess the depositional environment of sedimentary rocks. The ratio of long-chained n-fatty acids (C22-C32) to the total n-fatty acids also has the potential to discriminate sedimentary environments. The indicators based on the n-fatty acids were applied to the Franciscan bedded cherts. The result shows that the bedded cherts had deposited in continuous environments from the pelagic to the coastal. This is in harmony with the same inference based on major, trace and rare earth elements and normal paraffins.

Sea ice cover, and bottom water and sediment characteristics at stations in the Amundsen Gulf, Canada

Seasonal dynamics in the activity of Arctic shelf benthos have been the subject of few local studies, and the pronounced among-site variability characterizing their results makes it difficult to upscale and generalize their conclusions. In a regional study encompassing five sites at 100-595 m water depth in the southeastern Beaufort Sea, we found that total pigment concentrations in surficial sediments, used as proxies of general food supply to the benthos, rose significantly after the transition from ice-covered conditions in spring (March-June 2008) to open-water conditions in summer (June-August 2008), whereas sediment Chl a concentrations, typical markers of fresh food input, did not. Macrobenthic biomass (including agglutinated foraminifera >500 µm) varied significantly among sites (1.2-6.4 g C/m**2 in spring, 1.1-12.6 g C/m**2 in summer), whereas a general spring-to-summer increase was not detected. Benthic carbon remineralisation also ranged significantly among sites (11.9-33.2 mg C/m**2/day in spring, 11.6-44.4 mg C/m**2/day in summer) and did in addition exhibit a general significant increase from spring-to-summer. Multiple regression analysis suggests that in both spring and summer, sediment Chl a concentration is the prime determinant of benthic carbon remineralisation, but other factors have a significant secondary influence, such as foraminiferan biomass (negative in both seasons), water depth (in spring) and infaunal biomass (in summer). Our findings indicate the importance of the combined and dynamic effects of food supply and benthic community patterns on the carbon remineralisation of the polar shelf benthos in seasonally ice-covered seas.

Determination of sulfite and manganese in sediments from the K/T boundary

The solid-state-physics technique of electron spin resonance (ESR) has been employed in an exploratory study of marine limestones and impact-related deposits from Cretaceous-Tertiary (KT) boundary sites including Spain (Sopelana and Caravaca), New Jersey (Bass River), the U.S. Atlantic continental margin (Blake Nose, ODP Leg 171B/1049/A), and several locations in Belize and southern Mexico within -600 km of the Chicxulub crater. The ESR spectra of SO3(1-) (a radiation-induced point defect involving a sulfite ion substitutional for CO3(2-) which has trapped a positive charge) and Mn(2+) in calcite were singled out for analysis because they are unambiguously interpretable and relatively easy to record. ESR signal strengths of calcite-related SO3(1-) and Mn(2+) have been studied as functions of stratigraphic position in whole-rock samples across the KT boundary at Sopelana, Caravaca, and Blake Nose. At all three of these sites, anomalies in SO3(1-) and/or Mn(2+) intensities are noted at the KT boundary relative to the corresponding background levels in the rocks above and below. At Caravaca, the SO3(1-) background itself is found to be lower by a factor of 2.7 in the first 30,000 years of the Tertiary relative to its steady-state value in the last 15,000 years of the Cretaceous, indicating either an abrupt and quasi-permanent change in ocean chemistry (or temperature) or extinction of the marine biota primarily responsible for fixing sulfite in the late Cretaceous limestones. An exponential decrease in the Mn(2+) concentration per unit mass calcite, [Mn(2+)], as the KT boundary at Caravaca is approached from below (1/e characteristic length =1.4 cm) is interpreted as a result of post-impact leaching of the seafloor. Absolute ESR quantitative analyses of proximal impact deposits from Belize and southern Mexico group naturally into three distinct fields in a twodimensional [SO3(1-)]-versus-[Mn(2+)] scatter plot. These fields contain (I) limestone ejecta clasts, (II) accretionary lapilli, and (III) a variety of SO3(1-) -depleted/Mn(2+) enriched impact deposits. Data for the investigated non-impact-related Cretaceous and Tertiary marine limestones (Spain and Blake Nose) fall outside of these three fields. With reference to thes enon-impact deposits, fields I, II, and III can be respectively characterized as Mn(2+) -depleted, SO3(1-) -enhanced, and SO3(1-) -depleted. It is proposed that (1) field I represents calcites from the Yucatin Platform, and that the Mn(2+) -depleted signature can be used as an indicator of primary Chicxulub ejecta in deep marine environments and (2) field II represents calcites that include a component formed in the vapor plume, either from condensation in the presence of CO2/SO3(1-) -rich vapors, or reactions between CaO and CO2/SO3 rich vapors, and that this SO3(1-) -enhanced signature can be used as an indicator of impact vapor plume deposits. Given these two propositions, the ESR data for the Blake Nose deposits are ascribed to the presence of basal coarse calcitic Chicxulub ejecta clasts, while the finer components that are increasingly represented toward the top are interpreted to contain high- SO3(1-) calcite from the vapor plume. The apparently-undisturbed Bass River deposit may contain even higher concentrations of vapor-plume calcite. None of the three components included in field III appear to be represented at distal, deep marine KT-boundary sites; this field may include several types of impact-related deposits of diverse origins and diagenetic histories.

Detailed analysis of the organic carbon content in sediment cores M0002 and M0004

During IODP Expedition 302 (Arctic Coring Expedition-ACEX), the first scientific drilling campaign in the permantly ice-covered central Arctic Ocean, a 430 m thick sequence of upper Cretaceaous to Quaternary sediments has been drilled. The lower half of this sequence is composed of organic-carbon-rich (black shale-type) sediments with total organic carbon contents of about 1-14%. Significant amounts of the organic matter preserved in these sediments is of algae-type origin and accumulated under anoxic/euxinic conditions. Here, for the first time detailed data on the source-rock potential of these black shales are presented, indicating that most of the Eocene sediments have a (fair to) good source-rock potential, prone to generate a gas/oil mixture. The source-rock potential of the Campanian and upper Paleocene sediments, on the other hand, is rather low. The presence of oil or gas already generated in situ, however, can be ruled out due to the immaturity of the ACEX sediments.

Quantitative coccolithophore analyses in sediment core MD01-2446

Quantitative coccolithophore analyses were performed in core MD01-2446, located in the mid-latitude North Atlantic, to reconstruct climatically induced sea-surface water conditions throughout Marine Isotope Stages (MIS) 14-9. The data are compared to new and available paleoenvironmental proxies from the same site as well as other nearby North Atlantic records that support the coccolithophore signature at glacial-interglacial to millennial climate scale. Total coccolithophore absolute abundance increases during interglacials but abruptly drops during the colder glacial phases and deglaciations. Coccolithophore warm-water taxa (wwt) indicate that MIS11c and MIS9e experienced warmer and more stable conditions throughout the whole photic zone compared to MIS13. MIS11 was a long-lasting warmer and stable interglacial characterized by a climate optimum during MIS11c when a more prominent influence of the subtropical front at the site is inferred. The wwt pattern also suggests distinct interstadial and stadial events lasting about 4-10 kyr. The glacial increases of Gephyrocapsa margereli-G. muellerae 3-4 µm along with higher values of Corg, additionally supported by the total alkenone abundance at Site U1313, indicate more productive surface waters, likely reflecting the migration of the polar front into the mid-latitude North Atlantic. Distinctive peaks of G. margereli-muellerae (> 4 µm), C. pelagicus pelagicus, Neogloboquadrina pachyderma left coiling, and reworked nannofossils, combined with minima in total nannofossil accumulation rate, are tracers of Heinrich-type events during MIS12 and MIS10. Additional Heinrich-type events are suggested during MIS12 and MIS14 based on biotic proxies, and we discuss possible iceberg sources at these times. Our results improve the understanding of mid-Brunhes paleoclimate and the impact on phytoplankton diversity in the mid-latitude North Atlantic region.

Geochemical parameters of sediments and water of the Black Sea, data of Cruise 8 of R/V Vityaz-2

The book presents results of comprehensive geological investigations carried out during Cruise 8 of R/V "Vityaz-2" to the western part of the Black Sea in 1984. Systematic studies in the Black Sea during about hundred years have not weakened interest in the sea. Lithological and geochemical studies of sediments in estuarine areas of the Danube and the Kyzyl-Irmak rivers, as well as in adjacent parts of the deep sea and some other areas were the main aims of the cruise. Data on morphological structures of river fans, lithologic and chemical compositions of sediments in the fans and their areal distribution, forms of occurrence of chemical elements, role of organic matter and gases in sedimentation and diagenesis are given and discussed in the book.

Organic matter preservation in sediments of the Peru and Oman continental margin

Detailed petrographical and bulk geochemical investigations of organic matter (OM) have been performed on sediments deposited below or close to upwelling areas offshore Peru (ODP-Leg 112; Sites 679, 681, 688) and Oman (ODP-Leg 117; Sites 720, 723, 724) in order to obtain a quantitative understanding of its accumulation and degradation. Microscopical as well as nanoscopical investigations reveal that the OM in sediments affected by upwelling mechanisms mainly (up to 98%) consists of unstructured (amorphous) organic aggregates without any apparent biological structures. In sediments which are not or to a lesser extent affected by upwelling (Site 720) terrestrial OM predominates. Organic carbon (TOC) contents are highly variable and range between 9.8% in sediments deposited below upwelling cells and 0.2% in sediments outside the upwelling zone. The TOC/sulphur ratios of the sediments scatter widely. The samples from the deep-water locations (Sites 688 and 720), show C/S-ratios of "normal" marine sediments, whereas at the other locations no correlation or even a negative correlation between sulphur and TOC concentration exists. In most of the upwelling-influenced sediments OM contains a significant amount of sulphur. The incorporation of sulphur into the OM followed microbial sulphate reduction and occurred in the upper meters of the sedimentary column. Below, OM is still present in vast amounts and relatively hydrogen-rich, but is nevertheless non-metabolizable and becomes the limiting factor for bacterial sulphate reduction. According to mass balance calculations 90-99% of the OM produced in the photic zone was remineralized and 1-3% was consumed by microbial sulphate reduction. The aerobic and anaerobic processes have greatly affected degradation and conservation of OM.

Particle flux measurements from moorings and surface sediment studies of sediment cores in the SE Pacific Ocean

Flux of siliceous plankton and taxonomic composition of diatom and silicoflagellate assemblages were determined from sediment trap samples collected in coastal upwelling-influenced waters off northern Chile (30°S, CH site) under "normal" or non-El Niño (1993-94) and El Niño conditions (1997-98). In addition, concentration of biogenic opal and siliceous plankton, and diatom and silicoflagellate assemblages preserved in surface sediments are provided for a wide area between 27° and 43°S off Chile. Regardless of the year, winter upwelling determines the maximum production pattern of siliceous microorganisms, with diatoms numerically dominating the biogenic opal flux. During the El Niño year the export is markedly lower: on an annual basis, total mass flux diminished by 60%, and diatom and silicoflagellate export by 75%. Major components of the diatom flora maintain much of their regular seasonal cycle of flux maxima and minima during both sampling periods. Neritic resting spores (RS) of Chaetoceros dominate the diatom flux, mirroring the influence of coastal-upwelled waters at the CH trap site. Occurrence of pelagic diatoms species Fragilariopsis doliolus, members of the Rhizosoleniaceae, Azpeitia spp. and Nitzschia interruptestriata, secondary components of the assemblage, reflects the intermingling of warmer waters of the Subtropical Gyre. Dictyocha messanensis dominates the silicoflagellate association almost year-around, but Distephanus pulchra delivers ca. 60% of its annual production in less than three weeks during the winter peak. The siliceous thanatocoenosis is largely dominated by diatoms, whose assemblage shows significant qualitative and quantitative variations from north to south. Between 27° and 35°S, the dominance of RS Chaetoceros, Thalassionema nitzschioides var. nitzschioides and Skeletonema costatum reflects strong export production associated with occurrence of coastal upwelling. Both highest biogenic opal content and diatom concentration at 35° and 41°-43°S coincide with highest pigment concentrations along the Chilean coast. Predominance of the diatom species Thalassiosira pacifica and T. poro-irregulata, and higher relative contribution of the silicoflagellate Distephanus speculum at 41°-43°S suggest the influence of more nutrient-rich waters and low sea surface temperatures, probably associated with the Antarctic Circumpolar Water.