Autonomous learning of load and traffic patterns to improve cluster utilization

Adaptive clustering aims at improving cluster utilization for varying load and traffic patterns. Locality-based least-connection with replication (LBLCR) scheduling that comes with Linux is designed to help improve cluster utilization through adaptive clustering. A key issue with LBLCR, however, is that cluster performance depends much on a single threshold value that is used to determine adaptation. Once set, the threshold remains fixed, regardless of the load and traffic patterns. If a cluster of PCs is to adapt to different traffic patterns for high utilization, a good threshold has to be selected and used dynamically. We present in this paper an adaptive clustering framework that autonomously learns and adapts to different load and traffic patterns at runtime with no administrator intervention. The cluster is configured once and for all. As the patterns change, the cluster automatically expands/contracts to meet the changing demands. At the same time, the patterns are proactively learned so that when similar patterns emerge in the future, the cluster knows what to do to improve utilization. We have implemented this autonomous learning method and compared it with LBLCR using published Web traces. Experimental results indicate that our autonomous learning method produces high performance scalability and adaptability for different patterns. On the other hand LBLCR-based clustering suffers from performance scalability and adaptability for different traffic patterns since it is not designed to obtain good threshold values and use them at runtime.     

Prediction of taxon occurrence: a test on taxon‐specific change point values of stream benthic invertebrates

相似文献   

Species monophyly

In biological systematics, as well as in the philosophy of biology, species and higher taxa are individuated through their unique evolutionary origin. This is taken by some authors to mean that monophyly is a (relational) property not only of higher taxa, but also of species. A species is said to originate through speciation, and to go extinct when it splits into two daughter species (or through terminal extinction). Its unique evolutionary origin is said to bestow identity on a species through time and change, and to render species names rigid designators. Species names are thus believed to function just like names of supraspecific taxa. However, large parts of the Web of Life are composed of species that do not have a unique evolutionary origin from a single population, lineage or stem-species. Further, monophyly is an ambiguous concept if it is defined simply in terms of 'unique evolutionary origin'. Disambiguating the concept by defining a monophyletic taxon as 'a taxon that includes the ancestor and all, and only, its descendant' renders monophyly inapplicable to species. At the heart of the problem lies a fundamental distinction between species and monophyletic taxa, where species form mutually exclusive reticulated systems, while higher taxa form inclusive hierarchical systems. Examples are given both at the species level and below to illustrate the problems that result from the application of the monophyly criterion to species. The conclusion is that the concepts of exclusivity and monophyly should be treated as non-overlapping: exclusivity marks out a species synchronistically, i.e. in the present time. Monophyly marks out clades (groups of species) diachronistically, i.e. within an historical dimension.     

Finding the maximum a posteriori probability (MAP) in a Bayesian taxonomic key is NP-hard

As an alternative to dichotomous keys, tabular keys are used for taxonomic identification. With the use of computers, keys based on the Bayes formula can also be made available more widely. For the development of a key, the maximum a posterior probability (MAP) for a taxon is important because it allows to evaluate the quality of a key. If it is low, the taxon is hard to distinguish from other taxa. In this paper, we show that finding MAP in a Bayesian key is NP-hard. Estimates for MAP or other measures have to be used for the estimation of the quality of a Bayesian key.     

Can Global Weed Assemblages Be Used to Predict Future Weeds?

Predicting which plant taxa are more likely to become weeds in a region presents significant challenges to both researchers and government agencies. Often it is done in a qualitative or semi-quantitative way. In this study, we explored the potential of using the quantitative self-organising map (SOM) approach to analyse global weed assemblages and estimate likelihoods of plant taxa becoming weeds before and after they have been moved to a new region. The SOM approach examines plant taxa associations by analysing where a taxon is recorded as a weed and what other taxa are recorded as weeds in those regions. The dataset analysed was extracted from a pre-existing, extensive worldwide database of plant taxa recorded as weeds or other related status and, following reformatting, included 187 regions and 6690 plant taxa. To assess the value of the SOM approach we selected Australia as a case study. We found that the key and most important limitation in using such analytical approach lies with the dataset used. The classification of a taxon as a weed in the literature is not often based on actual data that document the economic, environmental and/or social impact of the taxon, but mostly based on human perceptions that the taxon is troublesome or simply not wanted in a particular situation. The adoption of consistent and objective criteria that incorporate a standardized approach for impact assessment of plant taxa will be necessary to develop a new global database suitable to make predictions regarding weediness using methods like SOM. It may however, be more realistic to opt for a classification system that focuses on the invasive characteristics of plant taxa without any inference to impacts, which to be defined would require some level of research to avoid bias from human perceptions and value systems.     

Genetic diversity and phylogenetic relationship in AA Oryza species as revealed by Rim2/Hipa CACTA transposon display

CACTA is a class 2 transposon, that is very abundantly present in plant genomes. Using Rim2/Hipa CACTA transposon display (hereafter Rim2/Hipa-TD), we analyzed several A-genome diploid Oryza species that have a high distribution of the CACTA motifs. High levels of polymorphism were detected within and between the Oryza species. The African taxa, O. glaberrima and O. barthii, both showed lower levels of polymorphism than the Asian taxa, O. sativa, O. rufipogon, and O. nivara. However, O. longistaminata, another African taxon, showed levels of polymorphism that were similar to the Asian taxa. The Latin American taxon, O. glumaepatula, and the Australian taxon, O. meridionalis, exhibited intermediate levels of polymorphism between those of the Asian and African taxa. The lowest level of polymorphism was observed in O. glaberrima (32.1%) and the highest level of polymorphism was observed in O. rufipogon (95.7%). The phylogenetic tree revealed three major groups at the genetic similarity level of 0.409. The first group consisted of three Asian taxa, O. sativa, O. rufipogon and O. nivara. The second group consisted of three African taxa, O. glaberrima, O. barthii, O. longistaminata, and an American taxon, O. glumaepatula. The third group contained an Australian taxon, O. meridionalis. The clustering patterns of these species matched well with their geographical origins. Rim2/Hipa-TD appears to be a useful marker system for studying the genetic diversity and species relationships among the AA diploid Oryza species.     

Climatic calibration of 80 aeropollinic taxa along a European transect

The aim of calibration is to measure the relationships between climate and taxa, in order to use the taxa as indicators for the estimation of bioclimates. The indicator capacity of a taxon (ICT) measures the probability of two events, together or separately, either the ordering of abundance, within the range of a factor, or the confining of presences inside a limited part of the range. ICT is obtained from numbering inequalities or matches, on such orderings, without any arithmetical operation upon the initial data. ICT measures the indicator capacity of a taxon for a factor, even if the taxon is intermittent or if its gradient is irregular.ICT increases with the threshold of the sum of temperatures, for some taxa, such as Arbutus, Celtis, Cistus, Coriaria, Cupressaceae, Helianthemum, Olea, Onobrychis, Pistacia, Silene, Thymelaeaceae. These taxa indicate a superior threshold above the classical one of 0 °C. Conversely, ICT decreases when the threshold increases, for other taxa, such as Abies, Aesculus, Fraxinus, Juglans, Mercurialis, Populus, Resedaceae, Salix, Thalictrum, Tilia. A taxon may be an indicator for the whole range of a temperature, if its abundance increases rather regularly, such as Chenopodiaceae, Erica, Olea, Plantago, Tamarix, Umbelliferae, Vitis, or if its abundance decreases, as for Betula, Filicidae, Fraxinus, Juglans, Tilia. Conversely, some other taxa are indicators only in a part of the range, such as Calluna, Galium and Platanus.The indicator capacity is used to estimate the thermic climate according to the flora. For each taxon, ICT depends on A and F, which are the ranks of abundance and thermal factor. For a given rank A, observed in a spectrum, ICT depends on F; ICT is the probability of the spectrum to be in the rank F. The maximum value of ICT, for all taxa in the spectrum, indicates the probable rank of that spectrum. Along the transect, the estimated temperatures happen to be equal to the measured ones for all the tested spectra (except one) and for all the factors, even if a small part of the flora is tested.     

The phylogenetic position of Brachiopoda—a comparison of morphological and molecular data

Analyses of rRNA and rDNA among Metazoa result in a hypothesis of a sistergroup relationship of Brachiopoda and certain spiralian taxa, whereas analyses of morphological data imply that Brachiopoda show affinities to Deuterostomia within the Radialia. Regarding Brachiopoda as a derived spiralian taxon must be followed by a reinterpretation of the evolution of distinct brachiopod morphological characters—like cleavage pattern, coelom or larva. The experimental insertion of a monophyletic taxon consisting of Brachiopoda and Phoronida into a widely accepted phylogenetic tree of Spiralia leads to the hypothesis that at least trimeric organization, mesosomal tentacular apparatus and heterogeneously assembled metanephridia are products of convergent evolution in Brachiopoda plus Phoronida and Deuterostomia. The hypothesis of a radialian nature of Brachiopoda and Phoronida, as implied by morphological data, remains as the most parsimonious possibility to explain the evolution of seven regarded characters (cleavage pattern, larva, tentacular apparatus, coelom, metameric segmentation, metanephridia and chaetae) in Brachiopoda. Due to the conflicting results of both methods a hitherto undetected systematical problem is discussed possibly hindering data comparability. If the course of evolution can principally be inferred from the information preserved in recent and fossil animals, the results should be congruent in the analyses of both, molecular and morphological data.     

The probability of correctly resolving a split as an experimental design criterion in phylogenetics

We illustrate how recently developed large sequence-length approximations to probabilities of correct phylogenetic reconstruction for maximum likelihood estimation can be used to evaluate experimental design strategies. The specific criterion of interest is the probability of correctly resolving an a priori defined split of interest in a phylogenetic tree. Design strategies considered include increased taxon sampling and increasing sequence length. Our analyses of specific examples strongly suggest that it is better to sample taxa that connect as close as possible to the split of interest. Assuming this can be done, these examples suggest it is better to sample additional taxa than to add a comparable number of sites for the existing taxa. If the rates of evolution in the added taxa are slow, it is better to choose taxa connecting to a long edge, but if rates are comparable to a sister lineage, it is not necessarily the best strategy to sample taxa connected to a long edge. We also examined deleting taxa while increasing the number of sites. Although deleting a small number of taxa distant from the split of interest can be beneficial, deleting too many or making poor choices as to what should be deleted can lead to smaller probabilities of correct reconstruction than for the original sequence data.     

Mitochondrial genomes in the Hymenoptera and their utility as phylogenetic markers

Abstract.  In this study, we assessed the ability of mitochondrial genome sequences to recover a test phylogeny of five hymenopteran taxa from which phylogenetic relationships are well accepted. Our analyses indicated that the test phylogeny was well recovered in all nucleotide Bayesian analyses when all the available holometabolan (i.e. outgroup) taxa were included, but only in Bayesian analyses excluding third codon positions when only the hymenopteran representatives and a single outgroup were included. This result suggests that taxon sampling of the outgroup might be as important as taxon sampling of the ingroup when recovering hymenopteran phylogenetic relationships using whole mitochondrial genomes. Parsimony analyses were more sensitive to both taxon sampling and the analytical model than Bayesian analyses, and analyses using the protein dataset did not recover the test phylogeny. In general, mitochondrial genomes did not resolve the position of the Hymenoptera within the Holometabola with confidence, suggesting that an increased taxon sampling, both within the Holometabola and among outgroups, is necessary.     

Temporal changes of biodiversity in urban running waters – Results of a twelve-year monitoring study

Freshwater biodiversity underlies severe threats, mainly suffering from habitat degradation by anthropogenic land use, in particular by urbanisation. However, recent long-term studies indicate recovery of stream macroinvertebrate diversity due to improved water quality at least in North America and Europe. We monitored macroinverbrates at 56 urban stream sites over a 12-year period (2009–2020) in Braunschweig, a German urban district. We utilised these data to investigate spatio-temporal changes in taxon richness and assemblage structure as well as factors potentially affecting the resulting patterns. Overall taxon richness was increasing over the study period, comprising both all taxa and taxa being indicators for healthy stream conditions. 53.6% of the sites had significant positive trends becoming most eminent since 2014, despite decelerating since 2018, the beginning of an extra-ordinary dry period. Only 10.7% of the study sites had negative trends. Assemblage structure was shaped by environmental factors like stream width and water quality. Over-average taxon richness including positive trends and higher numbers of indicator taxa of healthy stream conditions was found in streams with higher flow velocity, good saprobic conditions and more natural streambed structure. In contrast, low taxon richness and predominance of tolerant taxa were found in streams with more degraded conditions. Most of the environmental conditions having positive effects on taxon richness were improved by various programs set up by the environmental authorities. We therefore conclude, if urban stressors like organic pollution and structural degradation can be mitigated by revitalisation and water quality improvement, urban streams can have good potential for increasing biodiversity and improving ecological functioning.     

Across‐taxa incongruence in patterns of collecting bias

If biological collections tend to be taken near accessible areas, and the number of such areas is limited, then we should expect a similar spatial distribution of collecting effort across taxa. Alternatively, if researchers working on a given taxon pick collection localities based on idiosyncratic criteria, then there should be no spatial similarity in collecting effort. This study compares the spatial distribution of collecting effort for plants and birds in Amazonia. Collection localities were transformed into a Thiessen network where polygon size works as a surrogate for collecting effort. A correlation between botanical and ornithological datasets, with an adjustment for spatial autocorrelation, showed little congruence in the spatial distribution of collecting effort between the two taxa. This incongruence of the distribution of collection effort among taxa suggests that the identification of priority areas for research, and correction for Wallacean and Linnean shortfalls based on taxon‐specific studies, should not be generalized.     

Electrophoretic delineation of species boundaries within the short-necked freshwater turtles of Australia (Testudines: Ghelidae)

A total of 222 specimens from 55 populations of short-necked chelid turtle was collected from drainages in Australia and Papua New Guinea. Two populations were initially considered to belong to different diagnosable taxa if all individuals in one population could be distinguished from all individuals in the other by fixed allozyme differences. When two populations or diagnosable taxa shared allozymes at all presumptive loci, their profiles were combined into a single diagnosable taxon. Comparisons between populations and emerging diagnosable taxa were repeated until no further changes were possible. The species Elseya dentata comprised five clearly diagnosable taxa, differing by between 4 and 19 fixed allozyme differences. The currently recognized El. latisternum and El. novaeguineae were each a single diagnosable taxon, and there were three diagnosable taxa, including a sibling pair, that could not be assigned to a currendy described Elseya species. In contrast, all forms of Emydura were very closely related, with no two taxa differing by more than three fixed allozyme differences. There were three diagnosable taxa in the north (Em. victoriae, Em. subglobosa and one new form), though support for them was marginal. In the south, Em. macquarii, Em. krefflii and Em. signata formed only a single diagnosable taxon, even sharing rare alleles. If the phylogenetic species concept is adopted, there is support for recognition of 16 species of short-necked turtle in Australia, including Pseudemydura umbrina. Currendy only 10 are described. Our data also provide evidence of reproductive isolation in some cases (sympatric or parapatric), and comparative evidence (sensu Mayr) in others, than the traditional biological species concept applies also to these diagnosable taxa.     

Genetic and historical relationships among geothermally adapted Agrostis (bentgrass) of North America and Kamchatka: evidence for a previously unrecognized, thermally adapted taxon

Agrostis species have been known to evolve ecotypes rapidly in response to unusual edaphic conditions. The geographic distribution of Agrostis taxa in Lassen Volcanic National Park (California) and Yellowstone National Park (Wyoming) in the United States and the Valley of the Geysers (Kamchatka Peninsula) in Russia suggests that Agrostis scabra might have independently evolved morphologically similar ecotypes several times. We used RAPDs to show that, contrary to expectation, the thermal populations are not independently evolved, but instead constitute a single taxon that currently has four names. A UPGMA including the four thermal and nine nonthermal Agrostis taxa showed that the thermal cluster divides into geographically distinct subclusters, but that two morphologically distinct thermal taxa do not cluster independently. Even though currently confused with the thermal populations, nonthermal A. scabra is not closely related. An analysis of molecular variance (AMOVA) showed significant differentiation between the thermal populations and the nonthermal species sampled in this study. Splitting a hypothesized thermal operational taxonomic unit (OTU) into its components (geographically separated populations) does not greatly affect the partitioning of variation among OTUs. All thermal populations therefore should be assigned to the same taxon, but its taxonomic rank cannot be determined at this time.     

Compartmentalized organization of ecological niche occupation in insular invertebrate communities

Understanding the mechanisms of species distribution within ecosystems is a fundamental question of ecological research. The current worldwide changes and loss of habitats associated with a decline in species richness render this topic a key element for developing mitigation strategies. Ecological niche theory is a widely accepted concept to describe species distribution along environmental gradients where each taxon occupies its own distinct set of environmental parameters, that is, its niche. Niche occupation has been described in empirical studies for different closely related taxa, like ant, ungulate, or skink species, just to name a few. However, how species assemblages of whole ecosystems across multiple taxa are structured and organized has not been investigated thoroughly, although considering all taxa of a community would be essential when analyzing realized niches. Here, we investigated the organization of niche occupation and species distribution for the whole ground‐associated invertebrate community of small tropical insular ecosystems. By correlating environmental conditions with species occurrences using partial canonical correspondence analysis (pCCA), we demonstrated that the ground‐associated invertebrate community does not spread evenly across the overall niche space, but instead is compartmentalized in four distinct clusters: crustacean and gastropod taxa occurred in one cluster, attributable to the beach habitat, whereas hexapods and spider taxa occurred in three distinct inland clusters, attributable to distinct inland habitats, that is, grassland, open forest, and dense forest. Within the clusters, co‐occurrence pattern analysis suggested only a few negative interactions between the different taxa. By studying ground‐associated insular invertebrate communities, we have shown that species distribution and niche occupation can be, similar to food webs, organized in a compartmentalized way. The compartmentalization of the niche space might thereby be a mechanism to increase ecosystem resilience, as disturbances cascade more slowly throughout the ecosystem.     

Biotic element analysis in biogeography

Biotic element analysis is an alternative to the areas-of-endemism approach for recognizing the presence or absence of vicariance events in a given region. If an ancestral biota was fragmented by vicariance events, biotic elements or clusters of distribution areas should emerge. We propose a statistical test for clustering of distribution areas based on a Monte Carlo simulation with a null model that considers the spatial autocorrelation in the data. The hypothesis tested is that the observed degree of clustering of ranges can be explained by the range size distribution, the varying number of taxa per cell, and the spatial autocorrelation of the occurrences of a taxon alone. A method for the delimitation of biotic elements which uses model-based Gaussian clustering is introduced. We demonstrate our methods and show the importance of grid size by means of a case study, an analysis of the distribution patterns of southern African species of the weevil genus Scobius. The example highlights the difficulties in delimiting areas of endemism if dispersal has occurred and illustrates the advantages of the biotic element approach.     

Taxon mapping exemplifies punctuated equilibrium and atavistic saltation

Two or more exemplars of the same taxon forming a nonmonophyletic group on a molecular tree may be viewed as representing surviving populations of a deep shared ancestral taxon, and if different species of the same genus, then theoretically phenotypically static remnants of punctuated equilibrium. That taxon may be mapped on a molecular cladogram and evolutionarily resolved at the taxon level inclusive of all exemplars. The technique for mapping taxa on a molecular tree, termed here caulistics, is much like mapping traits but recovers macroevolutionary information at the taxon level. All lineages arising from the mapped taxon are its direct descendants. Mapped taxa superimposed or overlapping may reveal packaged adaptive traits. When a mapped taxon is well split by another mapped taxon on a molecular tree, atavistic saltation based on triggering an epigenetically retained trait complex is a theoretical explanation. Caulistics combines traditional taxonomy and molecular phylogenetics to reveal previously unknown aspects of the macroevolutionary past.     

Propositions for a character-state-based biological taxonomy

A new procedure for defining taxa upon a single character state is developed. It is centred on the designation of two specimens, belonging to two distinct species, exhibiting the same given character state, as type material, and referred to as ‘cladotypes’. A taxon name/definition designates a monophyletic group until one of the following assumptions is falsified: (1) the character state typified by cladotypes is homologous in individuals that are designated as cladotypes, and (2) cohesion mechanisms isolated individuals exhibiting the type‐character‐state from those that do not. A taxon defined by a character state that is found to be a combination of several character states is to be redefined upon a character state shared by its cladotypes. If several character states are available, the character state that makes the taxon the least inclusive taxon including cladotypic species (i.e., species to which belong cladotypes) is to be preferred. Taxon names designate obsolete phylogenetic hypotheses if the first assumption is falsified (such names are to be kept for this purpose, i.e., they are not to be recycled in another definition). Rules governing adaptation of previously erected names are proposed. Main cases of taxa synonymy involve definitions based on different pairs of cladotypes but referring to the same type‐character‐state; and definitions based on the same character‐state initially hypothesised as acquired by convergence in cladotypic species pairs, but later demonstrated as originating from a unique ancestor. Taxa could be synonyms if a permanent splitting event did not segregate individuals exhibiting a new character state, qualified as type‐character‐state, from individuals already assigned to a previously erected taxon. This procedure accommodates potentially any species concept, but is not tied to any; it is an extension of the composite species concept. Species are treated in a different way than other taxa: they are defined as sets of individuals belonging to the same evolving (segments of) metapopulation lineages as a holotype specimen, and do not need a defining character state.