On-line feedback-based automatic resource configuration for distributed applications

A key problem in executing performance critical applications on distributed computing environments (e.g. the Grid) is the selection of resources. Research related to “automatic resource selection” aims to allocate resources on behalf of users to optimize the execution performance. However, most of current approaches are based on the static principle (i.e. resource selection is performed prior to execution) and need detailed application-specific information. In the paper, we introduce a novel on-line automatic resource selection approach. This approach is based on a simple control theory: the application continuously reports the Execution Satisfaction Degree (ESD) to the middleware Application Agent (AA), which relies on the reported ESD values to learn the execution behavior and tune the computing environment by adding/replacing/deleting resources during the execution in order to satisfy users’ performance requirements. We introduce two different policies applied to this approach to enable the AA to learn and tune the computing environment: the Utility Classification policy and the Desired Processing Power Estimation (DPPE) policy. Each policy is validated by an iterative application and a non-iterative application to demonstrate that both policies are effective to support most kinds of applications.     

A centralized resource for bacterial–fungal interactions research

Research on bacterial-fungal interactions (BFIs) has revealed that fungi and bacteria frequently interact with one another within diverse ecosystems and microbiomes. Assessing the current state of knowledge within the field of BFI research, particularly with respect to what interactions between bacteria and fungi have been previously described, is very challenging and time consuming. This is largely due to a lack of any centralized resource, with reports of BFIs being spread across publications in numerous journals using non-standardized text to describe the relationships. To address this issue, we have developed the BFI Research Portal, a publicly accessible database of previously reported interactions between bacterial and fungal taxa to serve as a centralized resource for the field. Users can query bacterial or fungal taxa to see what members from the other kingdom have been observed as interaction partners. Search results are accompanied by interactive and intuitive visual outputs, and the database is a dynamic resource that will be updated as new BFIs are reported.     

Handling conflicts of context-aware reminding system in sensorised home

In this paper, we propose a context-aware reminding system to assist dementia elders for their daily activities. We particularly focus on the issue of conflict detecting and handling in the real-world reminding applications. We adopt a planning approach to describe the constraints of reminders, and transform the constraints into temporal constraint graph for reminder conflict detecting. In real-time reminder scheduling, we define three disruptive activities, and present a context-aware reminding strategy to handle the conflicts between pre-planned and disruptive activities. Finally, the system prototype is introduced.     

A Periodic Droop Model for Two Species Competition in A Chemostat

Both uniform persistence and the existence of periodic coexistence state are established for a periodically forced Droop model on two phytoplankton species competition in a chemostat under some appropriate conditions. Numerical simulations using biological data are presented as well to illustrate the main result. Research supported in part by the NSERC of Canada and the MITACS of Canada.     

1000 islands: an integrated approach to resource management for virtualized data centers

Recent advances in hardware and software virtualization offer unprecedented management capabilities for the mapping of virtual resources to physical resources. It is highly desirable to further create a “service hosting abstraction” that allows application owners to focus on service level objectives (SLOs) for their applications. This calls for a resource management solution that achieves the SLOs for many applications in response to changing data center conditions and hides the complexity from both application owners and data center operators. In this paper, we describe an automated capacity and workload management system that integrates multiple resource controllers at three different scopes and time scales. Simulation and experimental results confirm that such an integrated solution ensures efficient and effective use of data center resources while reducing service level violations for high priority applications.

A Mechanism for Ultra-Slow Oscillations in the Cortical Default Network

When the brain is in its noncognitive “idling” state, functional MRI measurements reveal the activation of default cortical networks whose activity is suppressed during cognitive processing. This default or background mode is characterized by ultra-slow BOLD oscillations (∼0.05 Hz), signaling extremely slow cycling in cortical metabolic demand across distinct cortical regions. Here we describe a model of the cortex which predicts that slow cycling of cortical activity can arise naturally as a result of nonlinear interactions between temporal (Hopf) and spatial (Turing) instabilities. The Hopf instability is triggered by delays in the inhibitory postsynaptic response, while the Turing instability is precipitated by increases in the strength of the gap-junction coupling between interneurons. We comment on possible implications for slow dendritic computation and information processing.     

Workload characterization in a high-energy data grid and impact on resource management

The analysis of data usage in a large set of real traces from a high-energy physics collaboration revealed the existence of an emergent grouping of files that we coined “filecules”. This paper presents the benefits of using this file grouping for prestaging data and compares it with previously proposed file grouping techniques along a range of performance metrics. Our experiments with real workloads demonstrate that filecule grouping is a reliable and useful abstraction for data management in science Grids; that preserving time locality for data prestaging is highly recommended; that job reordering with respect to data availability has significant impact on throughput; and finally, that a relatively short history of traces is a good predictor for filecule grouping. Our experimental results provide lessons for workload modeling and suggest design guidelines for data management in data-intensive resource-sharing environments.

A middleware approach for pipelining communications in clusters

The Pipelining Communications Middleware (PCM) approach provides a flexible, simple, high-performance mechanism to connect parallel programs running on high performance computers or clusters. This approach enables parallel programs to communicate and coordinate with each other to address larger problems than a single program can solve. The motivation behind the PCM approach grew out of using files as an intermediate transfer stage between processing by different programs. Our approach supersedes this practice by using streaming data set transfers as an “online” communication channel between simultaneously active parallel programs. Thus, the PCM approach addresses the issue of sending data from a parallel program to another parallel program without exposing details such as number of nodes allocated to the program, specific node identifiers, etc. This paper outlines and analyzes our proposed computation and communication model to provide efficient and convenient communications between parallel programs running on high performance computing systems or clusters. We also discuss the PCM challenges as well as current PCM implementations. Our approach achieves scalability, transparency, coordination, synchronization and flow control, and efficient programming. We experimented with data parallel applications to evaluate the performance of the PCM approach. Our experiment results show that the PCM approach achieves nearly ideal throughput that scales linearly with the underlying network medium speed. PCM performs well with small and large data transfers. Furthermore, our experiments show that network infrastructure plays the most significant role in the PCM performance.

A general mathematical performance model for wormhole-switched irregular networks

Irregular topologies are desirable network structures for building scalable cluster systems and very recently they have also been employed in SoC (system-on-chip) design. Many analytical models have been proposed in the literature to evaluate the performance of networks with different topologies such as hypercube, torus, mesh, hypermesh, Cartesian product networks, star graph, and k-ary n-cube; however, to the best of our knowledge, no mathematical model has been presented for irregular networks. Therefore, as an effort to fill this gap, this paper presents a comprehensive mathematical model for fully adaptive routing in wormhole-switched irregular networks. Moreover, since our approach holds no assumption for the network topology, the proposed analytical model covers all the aforementioned models (i.e. it covers both regular and irregular topologies). Furthermore, the model makes no preliminary assumption about the deadlock-free routing algorithm applied to the network. Finally, besides the generality of the model regarding the topology and routing algorithm, our analysis shows that the analytical model exhibits high accuracy which enables it to be used for almost all topologies with all traffic loads.     

Constructing a personalized prognostic risk model for colorectal cancer using machine learning and multi-omics approach based on epithelial–mesenchymal transition-related genes

The progression and the metastatic potential of colorectal cancer (CRC) are intricately linked to the epithelial–mesenchymal transition (EMT) process. The present study harnesses the power of machine learning combined with multi-omics data to develop a risk stratification model anchored on EMT-associated genes. The aim is to facilitate personalized prognostic assessments in CRC. We utilized publicly accessible gene expression datasets to pinpoint EMT-associated genes, employing a CoxBoost algorithm to sift through these genes for prognostic significance. The resultant model, predicated on gene expression levels, underwent rigorous independent validation across various datasets. Our model demonstrated a robust capacity to segregate CRC patients into distinct high- and low-risk categories, each correlating with markedly different survival probabilities. Notably, the risk score emerged as an independent prognostic indicator for CRC. High-risk patients were characterized by an immunosuppressive tumor milieu and a heightened responsiveness to certain chemotherapeutic agents, underlining the model's potential in steering tailored oncological therapies. Moreover, our research unearthed a putative repressive interaction between the long non-coding RNA PVT1 and the EMT-associated genes TIMP1 and MMP1, offering new insights into the molecular intricacies of CRC. In essence, our research introduces a sophisticated risk model, leveraging machine learning and multi-omics insights, which accurately prognosticates outcomes for CRC patients, paving the way for more individualized and effective oncological treatment paradigms.     

Do we know enough? A scientific and ethical analysis of the basis for genetic-based personalized nutrition

This article discusses the prospects and limitations of the scientific basis for offering personalized nutrition advice based upon individual genetic information. Two divergent scientific positions are presented, with an ethical comment. The crucial question is whether the current knowledge base is sufficiently strong for taking an ethically responsible decision to offer personalized nutrition advice based upon gene–diet–health interaction. According to the first position, the evidence base for translating the outcomes of nutrigenomics research into personalized nutritional advice is as yet immature. There is also limited evidence that genotype-based dietary advice will motivate appropriate behavior changes. Filling the gaps in our knowledge will require larger and better randomized controlled trials. According to the second position, personalized nutrition must be evaluated in relation to generally accepted standard dietary advice—partly derived from epidemiological observations and usually not proven by clinical trials. With personalized nutrition, we cannot demand stronger evidence. In several specific cases of gene–diet interaction, it may be more beneficial for individuals with specific genotypes to follow personalized advice rather than general dietary recommendations. The ethical comment, finally, considers the ethical aspects of deciding how to proceed in the face of such uncertainty. Two approaches for an ethically responsible way forward are proposed. Arguing from a precautionary approach, it is suggested that personalized dietary advice should be offered only when there is strong scientific evidence for health effects, followed by stepwise evaluation of unforeseen behavioral and psychological effects. Arguing from theoretical and applied ethics as well as psychology, it is also suggested that personalized advice should avoid paternalism and instead focus on supporting the autonomous choice of each person.     

Mapping of quantitative trait loci for mycoplasma and tetanus antibodies and interferon-gamma in a porcine F2 Duroc × Pietrain resource population

The aim of the present study was to detect quantitative trait loci (QTL) for innate and adaptive immunity in pigs. For this purpose, a Duroc × Pietrain F₂ resource population (DUPI) with 319 offspring was used to map QTL for the immune traits blood antibodies and interferon-gamma using 122 microsatellites covering all autosomes. Antibodies response to Mycoplasma hyopneumoniae and tetanus toxoid vaccine and the interferon-gamma (IFNG) serum concentration were measured at three different time points and were used as phenotypes. The differences of antibodies and interferon concentration between different time points were also used for the linkage mapping. Line-cross and imprinting QTL analysis, including two-QTL, were performed using QTL Express. A total of 30 QTL (12, 6, and 12 for mycoplasma, tetanus antibody, and IFNG, respectively) were identified at the 5% chromosome-wide-level significant, of which 28 were detected by line-cross and 2 by imprinting model. In addition, two QTL were identified on chromosome 5 using the two-QTL approach where both loci were in repulsion phase. Most QTL were detected on pig chromosomes 2, 5, 11, and 18. Antibodies were increased over time and immune traits were found to be affected by sex, litter size, parity, and month of birth. The results demonstrated that antibody and IFNG concentration are influenced by multiple chromosomal areas. The flanking markers of the QTL identified for IFNG on SSC5 did incorporate the position of the porcine IFNG gene. The detected QTL will allow further research in these QTL regions for candidate genes and their utilization in selection to improve the immune response and disease resistance in pig.     

Error-based or target-based? A unified framework for learning in recurrent spiking networks

The field of recurrent neural networks is over-populated by a variety of proposed learning rules and protocols. The scope of this work is to define a generalized framework, to move a step forward towards the unification of this fragmented scenario. In the field of supervised learning, two opposite approaches stand out, error-based and target-based. This duality gave rise to a scientific debate on which learning framework is the most likely to be implemented in biological networks of neurons. Moreover, the existence of spikes raises the question of whether the coding of information is rate-based or spike-based. To face these questions, we proposed a learning model with two main parameters, the rank of the feedback learning matrix R and the tolerance to spike timing τ_⋆. We demonstrate that a low (high) rank R accounts for an error-based (target-based) learning rule, while high (low) tolerance to spike timing promotes rate-based (spike-based) coding. We show that in a store and recall task, high-ranks allow for lower MSE values, while low-ranks enable a faster convergence. Our framework naturally lends itself to Behavioral Cloning and allows for efficiently solving relevant closed-loop tasks, investigating what parameters (R,τ⋆) are optimal to solve a specific task. We found that a high R is essential for tasks that require retaining memory for a long time (Button and Food). On the other hand, this is not relevant for a motor task (the 2D Bipedal Walker). In this case, we find that precise spike-based coding enables optimal performances. Finally, we show that our theoretical formulation allows for defining protocols to estimate the rank of the feedback error in biological networks. We release a PyTorch implementation of our model supporting GPU parallelization.     

A regression-based analytic model for capacity planning of multi-tier applications

The multi-tier implementation has become the industry standard for developing scalable client-server enterprise applications. Since these applications are performance sensitive, effective models for dynamic resource provisioning and for delivering quality of service to these applications become critical. Workloads in such environments are characterized by client sessions of interdependent requests with changing transaction mix and load over time, making model adaptivity to the observed workload changes a critical requirement for model effectiveness. In this work, we apply a regression-based approximation of the CPU demand of client transactions on a given hardware. Then, we use this approximation in an analytic model of a simple network of queues, each queue representing a tier, and show the approximation’s effectiveness for modeling diverse workloads with a changing transaction mix over time. Using two case studies, we investigate factors that impact the efficiency and accuracy of the proposed performance prediction models. Experimental results show that this regression-based approach provides a simple and powerful solution for efficient capacity planning and resource provisioning of multi-tier applications under changing workload conditions.