Experimental Analysis of a Gossip-Based Service for Scalable, Distributed Failure Detection and Consensus

Gossip protocols and services provide a means by which failures can be detected in large, distributed systems in an asynchronous manner without the limits associated with reliable multicasting for group communications. Extending the gossip protocol such that a system reaches consensus on detected faults can be performed via a flat structure, or it can be hierarchically distributed across cooperating layers of nodes. In this paper, the performance of gossip services employing flat and hierarchical schemes is analyzed on an experimental testbed in terms of consensus time, resource utilization and scalability. Performance associated with a hierarchically arranged gossip scheme is analyzed with varying group sizes and is shown to scale well. Resource utilization of the gossip-style failure detection and consensus service is measured in terms of network bandwidth utilization and CPU utilization. Analytical models are developed for resource utilization and performance projections are made for large system sizes.     

Gossip-Style Failure Detection and Distributed Consensus for Scalable Heterogeneous Clusters

Gossip protocols provide a means by which failures can be detected in large, distributed systems in an asynchronous manner without the limits associated with reliable multicasting for group communications. However, in order to be effective with application recovery and reconfiguration, these protocols require mechanisms by which failures can be detected with system-wide consensus in a scalable fashion. This paper presents three new gossip-style protocols supported by a novel algorithm to achieve consensus in scalable, heterogeneous clusters. The round-robin protocol improves on basic randomized gossiping by distributing gossip messages in a deterministic order that optimizes bandwidth consumption. Redundant gossiping is completely eliminated in the binary round-robin protocol, and the round-robin with sequence check protocol is a useful extension that yields efficient detection times without the need for system-specific optimization. The distributed consensus algorithm works with these gossip protocols to achieve agreement among the operable nodes in the cluster on the state of the system featuring either a flat or a layered design. The various protocols are simulated and evaluated in terms of consensus time and scalability using a high-fidelity, fault-injection model for distributed systems comprised of clusters of workstations connected by high-performance networks.     

Messaging on Gigabit Ethernet: Some Experiments with GAMMA and Other Systems

The Genoa Active Message MAchine (GAMMA) is a lightweight communication system based on the Active Ports paradigm, originally designed for efficient implementation over low-cost Fast Ethernet interconnects. In this paper we report about the recently completed porting of GAMMA to the Packet Engines GNIC-II and the Netgear GA620 Gigabit Ethernet adapters, and provide a comparison among GAMMA, MPI/GAMMA, TCP/IP, and MPICH, on such commodity interconnects, using different performance metrics. With a combination of low end-to-end latency (9.5 s with GNIC-II, 32 s with GA620) and high transmission throughput (almost 97 MByte/s with GNIC-II and 125 MByte/s with GA620, the latter obtained without changing the firmware of the adapter), GAMMA demonstrates the potential for Gigabit Ethernet lightweight protocols to yield messaging performance comparable to the best Myrinet-based messaging systems. This result is of interest, given the envisaged drop in cost of Gigabit Ethernet due to the transition from fiber optic to UTP cabling and ever increasing mass market production of such standard interconnect. We also reports about a technique for message fragmentation that is commonly exploited to increase the throughput with short message. When a different, though more widely used, performance metrics is considered, such a technique results into a performance loss rather than improvement.     

Key Message Approach to Optimize Communication of Parallel Applications on Clusters

Over the past few years, cluster/distributed computing has been gaining popularity. The proliferation of the cluster/distributed computing is due to the improved performance and increased reliability of these systems. Many parallel programming languages and related parallel programming models have become widely accepted. However, one of the major shortcomings of running parallel applications on cluster/distributed computing environments is the high communication overhead incurred. To reduce the communication overhead, and thus the completion time of a parallel application, this paper describes a simple, efficient and portable Key Message (KM) approach to support parallel computing on cluster/distributed computing environments. To demonstrate the advantage of the KM approach, a prototype runtime system has been implemented and evaluated. Our preliminary experimental results show that the KM approach has better improvement on communication of a parallel application when network background load increases or the computation to communication ratio of the application decreases.     

DAIRRy-BLUP: A High-Performance Computing Approach to Genomic Prediction

In genomic prediction, common analysis methods rely on a linear mixed-model framework to estimate SNP marker effects and breeding values of animals or plants. Ridge regression–best linear unbiased prediction (RR-BLUP) is based on the assumptions that SNP marker effects are normally distributed, are uncorrelated, and have equal variances. We propose DAIRRy-BLUP, a parallel, Distributed-memory RR-BLUP implementation, based on single-trait observations (y), that uses the Average Information algorithm for restricted maximum-likelihood estimation of the variance components. The goal of DAIRRy-BLUP is to enable the analysis of large-scale data sets to provide more accurate estimates of marker effects and breeding values. A distributed-memory framework is required since the dimensionality of the problem, determined by the number of SNP markers, can become too large to be analyzed by a single computing node. Initial results show that DAIRRy-BLUP enables the analysis of very large-scale data sets (up to 1,000,000 individuals and 360,000 SNPs) and indicate that increasing the number of phenotypic and genotypic records has a more significant effect on the prediction accuracy than increasing the density of SNP arrays.     

Relationship between burst properties and sensitivity to input: A theoretical analysis

This paper examines the sensitivity of endogenous bursters to a brief input pulse. The interneurons of the lobster cardiac ganglion were selected as a case study.Using a mathematical model specifically developed for the neurons in the cardiac ganglion of the lobster (Av-Ron et al., 1993), we show a tight link between burst characteristics and certain other parameters. We show that cells with different burst properties differ in their sensitivity to an input of a brief pulse.Irrespective of these differences, all cells display a bimodal response to a brief pulse applied during the quiescent period. During the first three-quarters of the quiescent period, they respond by producing a single spike at most. During the remaining one-quarter, the brief pulse can initiate the cells' intrinsic burst. Our predictions fit experimental results obtained by Tazaki and Cooke (1979).The results obtained herein are discussed with respect to fault tolerance considerations.     

Epigenetic regulation of genes during development： A conserved theme from flies to mammals

Eukaryotic genome is organized in form of chromatin within the nucleus. This organization is important for compaction of DNA as well as for the proper expression of the genes. During early embryonic development, genomic packaging receives variety of signals to eventually set up cell type specific expression patterns of genes. This process of regulated chromatinization leads to "cell type specific epigenomes". The expression states attained during differentiation process need to be maintained subsequently throughout the life of the organism. Epigenetie modifications are responsible for chromatin dependent regulatory mechanism and play a key role in maintenance of the expression state-a process referred to as cellular memory. Another key feature in the packaging of the genome is formation of chro- matin domains that are thought to be structural as well as functional units of the higher order chromatin organization. Boundary elements that function to define such domains set the limits of regulatory elements and that of epigenetie modifications. This connection of epige- netic modification, chromatin structure and genome organization has emerged from several studies. Hox genes are among the best studied in this context and have led to the significant understanding of the epigenetic regulation during development. Here we discuss the evolu- tionarily conserved features of epigenetic mechanisms emerged from studies on homeotic gene clusters.     

A Potential New Gene (tccA) on IncP Plasmid RK2 and Transposon Tn1721:Relationship of Its Product to the TrwC Relaxase/Helicase of IncW Plasmid R388

Computational analysis of the fully sequenced 60-kb genome of broad-host-range IncPα plasmid RK2 revealed a previously unreported potential protein-coding sequence, an 80-codon open reading frame (tccA), located in the region between the vegetative origin of replication (oriV) and thetetRgene of the tetracycline resistance determinant. The coding region is also present in the transposon Tn1721 tetregion, which is nearly identical to thetetregion of RK2. Remarkably, the predicted polypeptide product of the coding region displays 56% identity and 72% similarity with the C-terminal domain of the TrwC relaxase/helicase protein of IncW plasmid R388.     

The Source for the Difference Between Sulfhydryl and Hydroxyl Anions in Their Nucleophilic Addition Reaction to a Carbonyl Group: A DFT Approach.

Ab initio calculations show that sulfhydryl anion has a significantly lower potential than the hydroxide anion for stabilizing the products of its attack on carbonyl moieties - the tetrahedral complexes (TC). In this paper we analyze the factors that contribute to this phenomenon. Quantum mechanical MO ab initio calculations were used for studies of two reaction series, one for the attack of hydroxyl and one for the attack of sulfhydryl anion on different carbonyl compounds and their analogs. All of the anionic TCs formed by HS^- are characterized by higher charge transfer, but are significantly less stable than the relevant TC of HO^-. To explain the phenomenon we used a simple qualitative model based on Density Functional Theory (DFT). The crucial role of the occupied valence MOs is demonstrated in the process of electronegativity equalization between the donor and acceptor fragments in the final TC product. The sulfhydryl anion has significantly lower potential to stabilize TC products in comparison with the hydroxide anion because of the larger extent of electron back-donation from the electrophiles HOMO_A to the nucleophiles LUMO_D. This electron back-donation thus reduces the stability of the anionic TC in the case of HS^- and may account for the calculational results. Applications of this work to enzyme reactions help in understanding the differences in mechanisms of serine and cysteine proteases and may be used to guide the design of inhibitors for these enzymes. In perspective, the back-donation phenomenon discussed here may be applied to the study of electron transfer processes involving oxidation-reduction enzymes.     

Poly (AT) deletion/insertion polymorphism of the XPC gene contributes to urinary system cancer susceptibility: A meta-analysis

Numerous studies have investigated the association between xeroderma pigmentosum complementation group C (XPC) poly (AT) deletion/insertion (PAT −/+) polymorphism and cancer susceptibility; however, the findings are inconsistent. Therefore, we performed a meta-analysis based on 32 publications including 10,214 cases and 11,302 controls to acquire a more robust estimation of the relationship. We searched publications from MEDLINE, EMBASE and CBM which assessed the associations between XPC PAT −/+ polymorphism and cancer risk. We calculated pooled odds ratio (OR) and 95% confidence interval (CI) by using either fixed-effects or random-effects model. We found that individuals carrying the PAT +/+ genotype have significantly increased cancer risk (PAT +/+ vs. PAT −/−: OR = 1.18, 95% CI = 1.03–1.35 and recessive model: OR = 1.19, 95% CI = 1.06–1.33). Further stratification analysis showed a significantly increased risk for prostate cancer (PAT +/+ vs. PAT −/−: OR = 2.20, 95% CI = 1.39–3.48, recessive model: OR = 2.07, 95% CI = 1.33–3.23 and PAT + vs. PAT −: OR = 1.39, 95% CI = 1.12–1.71), bladder cancer (recessive model: OR = 1.33, 95% CI = 1.03–1.72), Caucasian ethnicity (recessive model: OR = 1.21, 95% CI = 1.02–1.43), population-based studies (recessive model: OR = 1.23, 95% CI = 1.05–1.43) and studies with relatively large sample size (PAT +/+ vs. PAT −/−: OR = 1.18, 95% CI = 1.04–1.35 and recessive model: OR = 1.20, 95% CI = 1.08–1.33). Despite some limitations, this meta-analysis established solid statistical evidence for the association between the XPC PAT +/+ genotype and cancer risk, especially for urinary system cancer, but this association warrants further validation in single large studies.