Fast DRAM increasingly dominates infrastructure spend in large scale computing environments and this trend will likely worsen without an architectural shift. The cost of deployed memory can be reduced by replacing part of the conventional DRAM with lower cost albeit slower memory media, thus creating a tiered memory system where both tiers are directly addressable and cached. But, this poses numerous challenges in a highly multi-tenant warehouse-scale computing setting. The diversity and scale of its applications motivates an application-transparent solution in the general case, adaptable to specific workload demands. This paper presents TMTS(Transparent Memory Tiering System), an application-transparent memory tiering management system that implements an adaptive, hardware-guided architecture to dynamically optimize access to the various directly-addressed memory tiers without faults. TMTS has been deployed at scale for two years serving thousands of production services, successfully meeting service level objectives (SLOs) across diverse application classes in the fleet. The solution is developed in terms of system level metrics it seeks to optimize and evaluated across the diverse workload mix to guide advanced policies embodied in a user-level agent. It sustains less than 5% overall performance degradation while replacing 25% of DRAM with a much slower medium.

Though secondary metabolism is integrated in networks with primary metabolism, still the biosynthetic pathways for most of the structurally known plant secondary compounds are unknown. Transformed roots developed following integration and expression of genes from Agrobacterium rhizogenes have proved to be a valuable model system in studying secondary metabolic pathways since they generally reflect the in planta operation of these pathways both in their route and enzymology. Transformed root culture is an isolated organ-culture system and physiologically distinct from the whole-plant system. This is important on two counts. First, transformed root cultures offer a better and simpler system in which to relate enzyme activities and product formation, uncomplicated by potential transport and source-sink relationships. Secondly, there is scant information in literature concerning phenylpropanoid and terpenoid metabolisms in roots of any kind. In this chapter, we report our findings with transformed root cultures of Daucus carota on a new biochemical pathway of hydroxybenzoic acid biosynthesis with an evidence of a phenylpropanoid chain cleavage enzyme for benzenoid formation and diversion of metabolic flows from phenylpropanoids to isoprenoids in green transformed roots upon photooxidative stress. Metabolite profiling using HPLC/ESI-MS/GC-MS, HPLC-based in vitro enzyme assays, inhibitor feeding experiments, and transcript analysis formed the basis of the above investigations.

Floral scent composed of low molecular weight volatile organic compounds. The sweet fragrance of any evening blooming flower is dominated by benzenoid and terpenoid volatile compounds. Floral scent of Jasminum sambac (Oleaceae) includes three major benzenoid esters - benzylacetate, methylbenzoate, and methylsalicylate and three major terpene compounds viz. (E)-beta-ocimene, linalool and alpha-farnesene. We analyzed concentrations and emission rates of benzenoids and terpenoids during the developmental stages ofJ. sambac flower. In addition to spatial emission from different floral parts, we studied the time course mRNA accumulations of phenylalanine ammonia-lyase (PAL) and the two representative genes of terpenoid pathway, namely 1-deoxy-u-xylulose 5 -phosphate reductoisomerase (DXR) and terpene synthase (TPS). Further, in vitro activities of several enzymes of phenylpropanoid/benzenoid pathway viz., PAL and acetyl -coenzyme A: benzylalcohol acetyltransferase (BEAT), S-adenosyl-L-methionine: benzoic acid carboxyl methyl transferase (BAMT) and S-adenosyl-L-methionine: salicylic acid carboxyl methyltransferase (SAMT) were studied. All the above enzyme activities along with the in vitro activities of DXR and TPS were found to follow a certain rhythm as observed in the emission of different benzenoid and terpenoid compounds. Linalool emission peaked after petal opening and coincided with maximal expression of JsTPS gene as evidenced from RT-PCR analyses (semi -quantitative). The maximum transcript accumulation of this gene was observed in flower petals, indicating that the petals ofJ. sambac flower play an important role as a major contributor of volatile precursors. The transcripts accumulation ofJsDXR and JsTPS in different developmental stages and in different floral part showed that emissions of terpenoid volatiles in]. sambac flower are partially regulated at transcription levels. (c)2016 Elsevier Ireland Ltd. All rights reserved.

MAIN CONCLUSION:A metabolic shift in green hairy root cultures of carrot from phenylpropanoid/benzenoid biosynthesis toward volatile isoprenoids was observed when compared with the metabolite profile of normal hairy root cultures. Hairy roots cultures of Daucus carota turned green under continuous illumination, while the content of the major phenolic compound p-hydroxybenzoic acid (p-HBA) was reduced to half as compared to normal hairy roots cultured in darkness. p-Hydroxybenzaldehyde dehydrogenase (HBD) activity was suppressed in the green hairy roots. However, comparative volatile analysis of 14-day-old green hairy roots revealed higher monoterpene and sesquiterpene contents than found in normal hairy roots. Methyl salicylate content was higher in normal hairy roots than in green ones. Application of clomazone, an inhibitor of 1-deoxy-D-xylulose 5-phosphate synthase (DXS), reduced the amount of total monoterpenes and sesquiterpenes in green hairy roots compared to normal hairy roots. However, methyl salicylate content was enhanced in both green and normal hairy roots treated with clomazone as compared to their respective controls. Because methyl-erythritol 4-phosphate (MEP) and phenylpropanoid pathways, respectively, contribute to the formation of monoterpenes and phenolic acids biosynthesis, the activities of enzymes regulating those pathways were measured in terms of their in vitro activities, in both green and normal hairy root cultures. These key enzymes were 1-deoxy-D-xylulose 5-phosphate reductoisomerase (DXR), an early regulatory enzyme of the MEP pathway, pyruvate kinase (PK), an enzyme of primary metabolism related to the MEP pathway, shikimate dehydrogenase (SKDH) which is involved in biosynthesis of aromatic amino acids, and phenylalanine ammonia-lyase (PAL) that catalyzes the first step of phenylpropanoid biosynthesis. Activities of DXR and PK were higher in green hairy roots as compared to normal ones, whereas the opposite trend was observed for SKDH and PAL activities. Gene expression analysis of DXR and PAL showed trends similar to those for the respective enzyme activities. Based on these observations, we suggest a possible redirection of metabolites from the primary metabolism toward isoprenoid biosynthesis, limiting the phenolic biosynthetic pathway in green hairy roots grown under continuous light.

Gaussian graphical models are popular for modeling high-dimensional multivariate data with sparse conditional dependencies. A mixture of Gaussian graphical models extends this model to the more realistic scenario where observations come from a heterogenous population composed of a small number of homogeneous sub-groups. In this paper we present a novel stochastic search algorithm for finding the posterior mode of high-dimensional Dirichlet process mixtures of decomposable Gaussian graphical models. Further, we investigate how to harness the massive thread-parallelization capabilities of graphical processing units to accelerate computation. The computational advantages of our algorithms are demonstrated with various simulated data examples in which we compare our stochastic search with a Markov chain Monte Carlo algorithm in moderate dimensional data examples. These experiments show that our stochastic search largely outperforms the Markov chain Monte Carlo algorithm in terms of computing-times and in terms of the quality of the posterior mode discovered. Finally, we analyze a gene expression dataset in which Markov chain Monte Carlo algorithms are too slow to be practically useful.

Cotransformed hairy roots containing a gene that encodes a fungal elicitor protein, β-cryptogein, were established in Withania somnifera, a medicinal plant widely used in Indian systems of medicine. To find out whether β-cryptogein protein endogenously elicits the pathway of withasteroid biosynthesis, withaferin A and withanolide A contents along with transcript accumulation of farnesyl pyrophosphate (FPP) synthase, 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMGR), and sterol glycosyltransferase (SGT) were analyzed in both cryptogein-cotransformed and normal hairy roots of W. somnifera. It was observed that the withaferin A and withanolide A contents were drastically higher in normal hairy roots than cryptogein-cotransformed ones. Similar trends were also observed on the levels of transcript accumulation. Subsequently, the enzyme activity of phenylalanine ammonia lyase (PAL), one of the key enzymes of phenylpropanoid pathway, was measured in both cryptogein-cotransformed and normal hairy roots of W. somnifera along with the levels of PAL transcript accumulation. Upliftment of PAL activity was observed in cryptogein-cotransformed hairy roots as compared to the normal ones, and the PAL expression also reflected a similar trend, i.e., enhanced expression in the cryptogein-cotransformed lines. Upliftment of wall-bound ferulic acid accumulation was also observed in the cryptogein-cotransformed lines, as compared to normal hairy root lines. Thus, the outcome of the above studies suggests a metabolic shift from withanolide accumulation to phenylpropanoid biosynthesis in cryptogein-cotransformed hairy roots of W. somnifera.

We discuss a new class of spatially varying, simultaneous autoregressive (SVSAR) models motivated by interests in flexible, non-stationary spatial modelling scalable to higher dimensions. SVSAR models are hierarchical Markov random fields extending traditional SAR models. We develop Bayesian analysis using Markov chain Monte Carlo methods of SVSAR models, with extensions to spatio-temporal contexts to address problems of data assimilation in computer models. A motivating application in atmospheric science concerns global CO emissions where prediction from computer models is assessed and refined based on high-resolution global satellite imagery data. Application to synthetic and real CO data sets demonstrates the potential of SVSAR models in flexibly representing inhomogeneous spatial processes on lattices, and their ability to improve estimation and prediction of spatial fields. The SVSAR approach is computationally attractive in even very large problems; computational efficiencies are enabled by exploiting sparsity of high-dimensional precision matrices.

The light-dependent generation of active oxygen species, which can disrupt normal metabolic process of plant, is termed as photo-oxidative stress. Plants are equipped with enzymatic and non-enzymatic antioxidative defence system to reduce the effect of such stress. Hairy root culture of Daucus carota when cultivated under continuous illumination (250 μmol m(-2)s(-1)) turned green. To know the reason behind that and photo-oxidative stress response in green hairy roots, activities of several antioxidant enzymes were measured. When compared with normal hairy roots, green hairy roots showed an enhanced superoxide dismutase (SOD) activity. Treatment with a SOD inhibitor diethyldithiocarbamate led to suppression of SOD activity in a concentration-dependent manner in green hairy roots. Interestingly, SOD-suppressed root showed three-fold enhanced caffeic acid glucoside accumulation in the soluble fraction as compared to untreated ones. While ascorbate peroxidase activity showed marginal increase in green hairy roots, a decrease in the activities of guaiacol peroxidase and catalase were observed. SDS-PAGE of crude protein profile from green hairy roots showed a distinct band, which was absent in normal hairy roots. MALDI-TOF-MS/MS analysis of the extracted protein confirmed it as the large subunit of RuBisCO. RT-PCR based expression analysis of betaine aldehyde dehydrogenase showed enhanced transcript levels in green hairy roots as compared to normal hairy roots, whereas reverse trends were observed with the transcripts accumulation for phenylalanine ammonia-lyase and chalcone synthase. These findings corroborate with the in vitro BADH activities in hairy roots, and thus indicate an important role of this stress enzyme in combating photo-oxidative stress in green hairy roots upon continuous light exposure.

Methyl-jasmonate (MJ)-treated hairy roots of Daucus carota L. were used to study the influence of alternative oxidase (AOX) in phenylpropanoid metabolism. Phenolic acid accumulation, as well as total flavonoids and lignin content of the MJ-treated hairy roots were decreased by treatment with salicylhydroxamic acid (SHAM), a known inhibitor of AOX. The inhibitory effect of SHAM was concentration dependent. Treatment with propyl gallate (PG), another inhibitor of AOX, also had a similar inhibitory effect on accumulation of phenolic acid, total flavonoids and lignin. The transcript levels of two DcAOX genes (DcAOX2a and DcAOX1a) were monitored at selected post-elicitation time points. A notable rise in the transcript levels of both DcAOX genes was observed preceding the MJ-induced enhanced accumulation of phenolics, flavonoids and lignin. An appreciable increase in phenylalanine ammonia-lyase (PAL) transcript level was also observed prior to enhanced phenolics accumulation. Both DcAOX genes showed differential transcript accumulation patterns after the onset of elicitation. The transcript levels of DcAOX1a and DcAOX2a attained peak at 6hours post elicitation (hpe) and 12hpe, respectively. An increase in the transcript levels of both DcAOX genes preceding the accumulation of phenylpropanoid-derivatives and lignin showed a positive correlation between AOX activity and phenylpropanoid biosynthesis. The results provide important new insight about the influence of AOX in phenylpropanoid biosynthesis.

Applied studies in multiple areas involving spatial and dynamic systems increasingly challenge our modelling and computational abilities as data volumes increase, and as spatial and temporal scales move to increasingly high-resolutions with parallel increase in complexity of dependency patterns. Motivated by two challenging problems of this sort, study of cellular dynamics in bacterial communication and global Carbon monoxide emissions prediction based on high-resolution global satellite imagery, this dissertation focuses on building sparse models and computational methods for data-dense dynamic, spatial and spatio-dynamic systems. The first part of the thesis develops a novel particle filtering algorithm for very long state-space models with sparse observations arising in studies of dynamic cellular networks. The need for increasing sample size with increasing dimension is met with parallel developments in informed resample-move strategies and distributed implementation. Fundamental innovations in the particle filtering literature are identified and used for designing an efficient particle filter. The second part of the thesis focuses on sparse spatial modelling of high-resolution lattice data. Gaussian Markov random field models, defined through spatial autoregressions, are adopted for their computational properties. Their potential is evidenced in an applied example in atmospheric chemistry where the focus is on inversion of satellite data combined with computer model predictions to infer ground-level CO emissions from multiple candidate sources on a global scale. Further, extending the framework of simultaneous autoregressive models, a novel hierarchical autoregressive model is developed for non-homogeneous spatial random-fields. The final part of the thesis develops a novel space-time model for data on a rectangular lattice. The dynamic spatial factor model framework is extended with matrix normal spatial factor loadings. A new class of Gaussian Markov random field models for random matrices, defined with low-dimensional row and column conditional independence graphs, is used to model sparse spatial factor loadings. Further dimensionality reduction is achieved through the dynamic factor model framework, which makes this class of models extremely attractive for systematically evolving non-homogeneous, high-resolution space-time data on rectangular lattices. Flexible choices for prior distributions and posterior computations are presented and illustrated with a synthetic data example.

We present and discuss the use of Bayesian modeling and computational methods for atmospheric chemistry inverse analyses that incorporate evaluation of spatial structure in model-data residuals. Motivated by problems of refining bottom-up estimates of source/sink fluxes of trace gas and aerosols based on satellite retrievals of atmospheric chemical concentrations, we address the need for formal modeling of spatial residual error structure in global scale inversion models. We do this using analytically and computationally tractable conditional autoregressive (CAR) spatial models as components of a global inversion framework. We develop Markov chain Monte Carlo methods to explore and fit these spatial structures in an overall statistical framework that simultaneously estimates source fluxes. Additional aspects of the study extend the statistical framework to utilize priors on source fluxes in a physically realistic manner, and to formally address and deal with missing data in satellite retrievals. We demonstrate the analysis in the context of inferring carbon monoxide (CO) sources constrained by satellite retrievals of column CO from the Measurement of Pollution in the Troposphere (MOPITT) instrument on the TERRA satellite, paying special attention to evaluating performance of the inverse approach using various statistical diagnostic metrics. This is developed using synthetic data generated to resemble MOPITT data to define a proof-of-concept and model assessment, and then in analysis of real MOPITT data. These studies demonstrate the ability of these simple spatial models to substantially improve over standard non-spatial models in terms of statistical fit, ability to recover sources in synthetic examples, and predictive match with real data.

Hairy root cultures of Daucus carota respond to methyl-jasmonate treatment with enhanced accumulation of p-hydroxybenzoic acid. The final C1-side chain of this compound is shaped by p-hydroxybenzaldehyde dehydrogenase (HBD) that catalyzes the formation of p-hydroxybenzoic acid from p-hydroxybenzaldehyde in the presence of NAD+. HBD was biochemically characterized from cell-free hairy root extracts of D. carota. The preferred substrate for HBD was p-hydroxybenzaldehyde. The apparent K m values were 54.8 and 74.4 μM for p-hydroxybenzaldehyde and NAD+, respectively. Divalent metal cations did not significantly affect enzyme activity.

Sequential Monte Carlo analysis of time series provides a direct approach to evaluating approximate model marginal likelihoods for model comparison. We exemplify this in studies of dynamic bacterial communication in systems biology, where a long sequence of state vectors follow a complicated nonlinear dynamic model with several defining biochemical parameters. MCMC methods do not mix well in these contexts, and do not lead easily to reliable estimates of model marginal likelihood. We develop an auxiliary particle filtering algorithm that simultaneously updates latent states and fixed parameters. Our algorithm takes advantage of distributed computing to carry forward a huge number of particles to ensure accuracy of the estimates. Marginal likelihood computation is developed and illustrated in evaluation of relevance of selected model components.