Search

Ensemble data assimilation techniques form an indispensable part of numerical weather prediction. As the ensemble size grows and model resolution increases, the amount of required storage becomes a major issue. Data compression schemes may come to the rescue not only for operational weather prediction, but also for weather history archives. In this paper, we present the design and implementation of an easy-to-use framework for evaluating the impact of lossy data compression in large scale ensemble data assimilation. The framework leverages robust statistical qualifiers to determine which compression parameters can be safely applied to the climate variables. Furthermore, our proposal can be used to apply the best parameters during operation, while monitoring data integrity. We perform an exemplary study on the Lorenz96 model to identify viable compression parameters and achieve a 1/3 saving in storage space and an effective speedup of 6% per assimilation cycle, while monitoring the state integrity., Comment: 12 pages, 6 figures

Blockprint, a tool for assessing client diversity on the Ethereum beacon chain, is essential for analyzing decentralization. This paper details experiments conducted at MigaLabs to enhance Blockprint's accuracy, evaluating various configurations for the K-Nearest Neighbors (KNN) classifier and exploring the Multi-Layer Perceptron (MLP) classifier as a proposed alternative. Findings suggest that the MLP classifier generally achieves superior accuracy with a smaller training dataset. The study revealed that clients running in different modes, especially those subscribed to all subnets, impact attestation inclusion differently, leading to proposed methods for mitigating the decline in model accuracy. Consequently, the recommendation is to employ an MLP model trained with a combined dataset of slots from both default and subscribed-to-all-subnets client configurations., Comment: 2 pages

This paper presents an in-depth exploration of Data Availability Sampling (DAS) and sharding mechanisms within decentralized systems through simulation-based analysis. DAS, a pivotal concept in blockchain technology and decentralized networks, is thoroughly examined to unravel its intricacies and assess its impact on system performance. Through the development of a simulator tailored explicitly for DAS, we embark on a comprehensive investigation into the parameters that influence system behavior and efficiency. A series of experiments are conducted within the simulated environment to validate theoretical formulations and dissect the interplay of DAS parameters. This includes an exploration of approaches such as custody by row, variations in validators per node, and malicious nodes. The outcomes of these experiments furnish insights into the efficacy of DAS protocols and pave the way for the formulation of optimization strategies geared towards enhancing decentralized network performance. Moreover, the findings serve as guidelines for future research endeavors, offering a nuanced understanding of the complexities inherent in decentralized systems. This study not only contributes to the theoretical understanding of DAS but also offers practical implications for the design, implementation, and optimization of decentralized systems., Comment: 4 pages, 1 figure

There have been several studies into measuring the level of decentralization in Ethereum through applying various indices to indicate the relative dominance of entities in different domains in the ecosystem. However, these indices do not capture any correlation between those different entities, that could potentially make them the subject of external coercion, or covert collusion. We propose an index that measures the relative dominance of entities based on the application of correlation factors. We posit that this approach produces a more nuanced and accurate index of decentralization., Comment: 11 pages, 7 figures, 3 tables

Scalability in blockchain remains a significant challenge, especially when prioritizing decentralization and security. The Ethereum community has proposed comprehensive data-sharding techniques to overcome storage, computational, and network processing limitations. In this context, the propagation and availability of large blocks become the subject of research to achieve scalable data-sharding. This paper provides insights after exploring the usage of a Kademlia-based DHT to enable Data Availability Sampling (DAS) in Ethereum. It presents a DAS-DHT simulator to study this problem and validates the results of the simulator with experiments in a real DHT network, IPFS. Our results help us understand what parts of DAS can be achieved based on existing Kademlia DHT solutions and which ones cannot. We discuss the limitations of DHT solutions and discuss other alternatives.

Scalability is a common issue among the most used permissionless blockchains, and several approaches have been proposed to solve this issue. Tackling scalability while preserving the security and decentralization of the network is a significant challenge. To deliver effective scaling solutions, Ethereum achieved a major protocol improvement, including a change in the consensus mechanism towards Proof of Stake. This improvement aimed a vast reduction of the hardware requirements to run a node, leading to significant sustainability benefits with a lower network energy consumption. This work analyzes the resource usage behavior of different clients running as Ethereum consensus nodes, comparing their performance under different configurations and analyzing their differences. Our results show higher requirements than claimed initially and how different clients react to network perturbations. Furthermore, we discuss the differences between the consensus clients, including their strong points and limitations.

Modern public blockchains like Ethereum rely on p2p networks to run distributed and censorship-resistant applications. With its wide adoption, it operates as a highly critical public ledger. On its transition to become more scalable and sustainable, shifting to PoS without sacrificing the security and resilience of PoW, Ethereum offers a range of consensus clients to participate in the network. In this paper, we present a methodology to measure the performance of the consensus clients based on the latency to receive messages from the p2p network. The paper includes a study that identifies the incentives and limitations that the network experiences, presenting insights about the latency impact derived from running the software in different locations., Comment: THE FIFTH INTERNATIONAL CONFERENCE ON BLOCKCHAIN COMPUTING AND APPLICATIONS (BCCA 2023)

Like most modern blockchain networks, Ethereum has relied on economic incentives to promote honest participation in the chain's consensus. The distributed character of the platform, together with the "randomness" or "luck" factor that both proof of work (PoW) and proof of stake (PoS) provide when electing the next block proposer, pushed the industry to model and improve the reward system of the system. With several improvements to predict PoW block proposal rewards and to maximize the extractable rewards of the same ones, the ultimate Ethereum's transition to PoS applied in the Paris Hard-Fork, more generally known as "The Merge", has meant a significant modification on the reward system in the platform. In this paper, we aim to break down both theoretically and empirically the new reward system in this post-merge era. We present a highly detailed description of the different rewards and their share among validators' rewards. Ultimately, we offer a study that uses the presented reward model to analyze the performance of the network during this transition.

Particle filters are a group of algorithms to solve inverse problems through statistical Bayesian methods when the model does not comply with the linear and Gaussian hypothesis. Particle filters are used in domains like data assimilation, probabilistic programming, neural networkoptimization, localization and navigation. Particle filters estimate the probabilitydistribution of model states by running a large number of model instances, the so called particles. The ability to handle a very large number of particles is critical for high dimensional models.This paper proposes a novel paradigm to run very large ensembles of parallel model instances on supercomputers. The approach combines an elastic and fault tolerant runner/server model minimizing data movementswhile enabling dynamic load balancing. Particle weights are computed locally on each runner andtransmitted when available to a server that normalizes them, resamples new particles based on their weight, and redistributes dynamically the work torunners to react to load imbalance. Our approach relies on a an asynchronously manageddistributed particle cache permitting particles to move from one runner to another inthe background while particle propagation goes on. This also enables the number ofrunners to vary during the execution either in reaction to failures and restarts, orto adapt to changing resource availability dictated by external decision processes.The approach is experimented with the Weather Research and Forecasting (WRF) model, toassess its performance for probabilistic weather forecasting. Up to 2555particles on 20442 compute cores are used to assimilate cloud cover observations into short--range weather forecasts over Europe.

Deep Neural Network (DNN) frameworks use distributed training to enable faster time to convergence and alleviate memory capacity limitations when training large models and/or using high dimension inputs. With the steady increase in datasets and model sizes, model/hybrid parallelism is deemed to have an important role in the future of distributed training of DNNs. We analyze the compute, communication, and memory requirements of Convolutional Neural Networks (CNNs) to understand the trade-offs between different parallelism approaches on performance and scalability. We leverage our model-driven analysis to be the basis for an oracle utility which can help in detecting the limitations and bottlenecks of different parallelism approaches at scale. We evaluate the oracle on six parallelization strategies, with four CNN models and multiple datasets (2D and 3D), on up to 1024 GPUs. The results demonstrate that the oracle has an average accuracy of about 86.74% when compared to empirical results, and as high as 97.57% for data parallelism., Comment: The International ACM Symposium on High-Performance Parallel and Distributed Computing 2021 (HPDC'21)

Achieving the equilibrium between scalability, sustainability, and security while keeping decentralization has prevailed as the target solution for decentralized blockchain applications over the last years. Several approaches have been proposed by multiple blockchain teams to achieve it, Ethereum being among them. Ethereum is on the path of a major protocol improvement called Ethereum 2.0 (Eth2), implementing Sharding and introducing the Proof-of-Stake (PoS). As the change of consensus mechanism is a delicate matter, this improvement will be achieved through different phases, the first of which is the implementation of the Beacon Chain. As Ethereum1, Eth2 relies on a decentralized peer-to-peer (p2p) network for the message distribution. Up to date, we estimate that there are around 5.000 nodes in the Eth2 main net geographically distributed. However, the topology of this one still prevails unknown. In this paper, we present the results obtained from the analysis we performed on the Eth2 p2p network. Describing the topology of the network, as possible hazards that this one implies.

Scalability is a common issue among the most used permissionless blockchains, and several approaches have been proposed accordingly. As Ethereum is set to be a solid foundation for a decentralized Internet web, the need for tackling scalability issues while preserving the security of the network is an important challenge. In order to successfully deliver effective scaling solutions, Ethereum is on the path of a major protocol improvement called Ethereum 2.0 (Eth2), which implements sharding. As the change of consensus mechanism is an extremely delicate matter, this improvement will be achieved through different phases, the first of which is the implementation of the Beacon Chain. For this, a specification has been developed and multiple groups have implemented clients to run the new protocol. In this work, we analyse the resource usage behaviour of different clients running as Eth2 nodes, comparing their performance and analysing differences. Our results show multiple network perturbations and how different clients react to it.

Deep learning (DL) applications are increasingly being deployed on HPC systems, to leverage the massive parallelism and computing power of those systems for DL model training. While significant effort has been put to facilitate distributed training by DL frameworks, fault tolerance has been largely ignored. In this work, we evaluate checkpoint-restart, a common fault tolerance technique in HPC workloads. We perform experiments with three state-of-the-art DL frameworks common in HPC Chainer, PyTorch, and TensorFlow). We evaluate the computational cost of checkpointing, file formats and file sizes, the impact of scale, and deterministic checkpointing. Our evaluation shows some critical differences in checkpoint mechanisms and exposes several bottlenecks in existing checkpointing implementations. We provide discussion points that can aid users in selecting a fault-tolerant framework to use in HPC. We also provide takeaway points that framework developers can use to facilitate better checkpointing of DL workloads in HPC.

Operating systems have historically had to manage only a single type of memory device. The imminent availability of heterogeneous memory devices based on emerging memory technologies confronts the classic single memory model and opens a new spectrum of possibilities for memory management. Transparent data movement between different memory devices based on access patterns of applications is a desired feature to make optimal use of such devices and to hide the complexity of memory management to the end-user. However, capturing memory access patterns of an application at runtime comes at a cost, which is particularly challenging for large scale parallel applications that may be sensitive to system noise. In this work, we focus on the access pattern profiling phase prior to the actual memory relocation. We study the feasibility of using Intel's Processor Event-Based Sampling (PEBS) feature to record memory accesses by sampling at runtime and study the overhead at scale. We have implemented a custom PEBS driver in the IHK/McKernel lightweight multi-kernel operating system, one of whose advantages is minimal system interference due to the lightweight kernel's simple design compared to other OS kernels such as Linux. We present the PEBS overhead of a set of scientific applications and show the access patterns identified in noise-sensitive HPC applications. Our results show that clear access patterns can be captured with a 10% overhead in the worst-case and 1% in the best case when running on up to 128k CPU cores (2,048 Intel Xeon Phi Knights Landing nodes). We conclude that online memory access profiling using PEBS at large scale is promising for memory management in heterogeneous memory environments., Comment: 8 pages, 16 figures, conference

This work is based on the seminar titled ``Resiliency in Numerical Algorithm Design for Extreme Scale Simulations'' held March 1-6, 2020 at Schloss Dagstuhl, that was attended by all the authors. Naive versions of conventional resilience techniques will not scale to the exascale regime: with a main memory footprint of tens of Petabytes, synchronously writing checkpoint data all the way to background storage at frequent intervals will create intolerable overheads in runtime and energy consumption. Forecasts show that the mean time between failures could be lower than the time to recover from such a checkpoint, so that large calculations at scale might not make any progress if robust alternatives are not investigated. More advanced resilience techniques must be devised. The key may lie in exploiting both advanced system features as well as specific application knowledge. Research will face two essential questions: (1) what are the reliability requirements for a particular computation and (2) how do we best design the algorithms and software to meet these requirements? One avenue would be to refine and improve on system- or application-level checkpointing and rollback strategies in the case an error is detected. Developers might use fault notification interfaces and flexible runtime systems to respond to node failures in an application-dependent fashion. Novel numerical algorithms or more stochastic computational approaches may be required to meet accuracy requirements in the face of undetectable soft errors. The goal of this Dagstuhl Seminar was to bring together a diverse group of scientists with expertise in exascale computing to discuss novel ways to make applications resilient against detected and undetected faults. In particular, participants explored the role that algorithms and applications play in the holistic approach needed to tackle this challenge., Comment: 45 pages, 3 figures, submitted to The International Journal of High Performance Computing Applications

One of the major challenges in using extreme scale systems efficiently is to mitigate the impact of faults. Application-level checkpoint/restart (CR) methods provide the best trade-off between productivity, robustness, and performance. There are many solutions implementing CR at the application level. They all provide advanced I/O capabilities to minimize the overhead introduced by CR. Nevertheless, there is still room for improvement in terms of programmability and flexibility, because end-users must manually serialize and deserialize application state using low-level APIs, modify the flow of the application to consider restarts, or rewrite CR code whenever the backend library changes. In this work, we propose a set of compiler directives and clauses that allow users to specify CR operations in a simple way. Our approach supports the common CR features provided by all the CR libraries. However, it can also be extended to support advanced features that are only available in some CR libraries, such as differential checkpointing, the use of HDF5 format, and the possibility of using fault-tolerance-dedicated threads. The result of our evaluation revealed a high increase in programmability. On average, we reduced the number of lines of code by 71%, 94%, and 64% for FTI, SCR, and VeloC, respectively, and no additional overhead was perceived using our solution compared to using the backend libraries directly. Finally, portability is enhanced because our programming model allows the use of any backend library without changing any code.

To improve power efficiency, researchers are experimenting with dynamically adjusting the supply voltage of systems below the nominal operating points. However, production systems are typically not allowed to function on voltage settings that is below the reliable limit. Consequently, existing software fault tolerance studies are based on fault models, which inject faults on random fault locations using fault injection techniques. In this work we study whether random fault injection is accurate to simulate the behavior of undervolted SRAMs. Our study extends the Gem5 simulator to support fault injection on the caches of the simulated system. The fault injection framework uses fault maps, which describe the faulty bits of SRAMs, as inputs. To compare random fault injection and hardware guided fault injection, we use two types of fault maps. The first type of maps are created through undervolting real SRAMs and observing the location of the erroneous bits, whereas the second type of maps are created by corrupting random bits of the SRAMs. During our study we corrupt the L1-Dcache of the simulated system and we monitor the behavior of the two types of fault maps on the resiliency of six benchmarks. The difference among the resiliency of a benchmark when tested with the different fault maps can be up to 24%.

Many social and economic systems can be represented as attributed networks encoding the relations between entities who are themselves described by different node attributes. Finding anomalies in these systems is crucial for detecting abuses such as credit card frauds, web spams or network intrusions. Intuitively, anomalous nodes are defined as nodes whose attributes differ starkly from the attributes of a certain set of nodes of reference, called the context of the anomaly. While some methods have proposed to spot anomalies locally, globally or within a community context, the problem remain challenging due to the multi-scale composition of real networks and the heterogeneity of node metadata. Here, we propose a principled way to uncover outlier nodes simultaneously with the context with respect to which they are anomalous, at all relevant scales of the network. We characterize anomalous nodes in terms of the concentration retained for each node after smoothing specific signals localized on the vertices of the graph. Besides, we introduce a graph signal processing formulation of the Markov stability framework used in community detection, in order to find the context of anomalies. The performance of our method is assessed on synthetic and real-world attributed networks and shows superior results concerning state of the art algorithms. Finally, we show the scalability of our approach in large networks employing Chebychev polynomial approximations., Comment: Accepted paper at the 34th AAAI Conference on Artificial Intelligence (AAAI-20), New York, US. 8 pages, 5 figures

High-performance computing (HPC) requires resilience techniques such as checkpointing in order to tolerate failures in supercomputers. As the number of nodes and memory in supercomputers keeps on increasing, the size of checkpoint data also increases dramatically, sometimes causing an I/O bottleneck. Differential checkpointing (dCP) aims to minimize the checkpointing overhead by only writing data differences. This is typically implemented at the memory page level, sometimes complemented with hashing algorithms. However, such a technique is unable to cope with dynamic-size datasets. In this work, we present a novel dCP implementation with a new file format that allows fragmentation of protected datasets in order to support dynamic sizes. We identify dirty data blocks using hash algorithms. In order to evaluate the dCP performance, we ported the HPC applications xPic, LULESH 2.0 and Heat2D and analyze them regarding their potential of reducing I/O with dCP and how this data reduction influences the checkpoint performance. In our experiments, we achieve reductions of up to 62% of the checkpoint time., Comment: This project has received funding from the European Unions Seventh Framework Programme (FP7/2007-2013) and the Horizon 2020 (H2020) funding framework under grant agreement no. H2020-FETHPC-754304 (DEEP-EST); and the LEGaTO Project (legato- project.eu), grant agreement No 780681

Fault-tolerance has always been an important topic when it comes to running massively parallel programs at scale. Statistically, hardware and software failures are expected to occur more often on systems gathering millions of computing units. Moreover, the larger jobs are, the more computing hours would be wasted by a crash. In this paper, we describe the work done in our MPI runtime to enable both transparent and application-level checkpointing mechanisms. Unlike the MPI 4.0 User-Level Failure Mitigation (ULFM) interface, our work targets solely Checkpoint/Restart and ignores other features such as resiliency. We show how existing checkpointing methods can be practically applied to a thread-based MPI implementation given sufficient runtime collaboration. The two main contributions are the preservation of high-speed network performance during transparent C/R and the over-subscription of checkpoint data replication thanks to a dedicated user-level scheduler support. These techniques are measured on MPI benchmarks such as IMB, Lulesh and Heatdis, and associated overhead and trade-offs are discussed., Comment: This research has been partially sponsored by the European Union s Horizon 2020 Programme under the LEGaTO Project (www.legato-project.eu), grant agreement 780681 and the Mont-Blanc 2020 project, grant agreement no. 779877

A main challenge in mining network-based data is finding effective ways to represent or encode graph structures so that it can be efficiently exploited by machine learning algorithms. Several methods have focused in network representation at node/edge or substructure level. However, many real life challenges such as time-varying, multilayer, chemical compounds and brain networks involve analysis of a family of graphs instead of single one opening additional challenges in graph comparison and representation. Traditional approaches for learning representations relies on hand-crafting specialized heuristics to extract meaningful information about the graphs, e.g statistical properties, structural features, etc. as well as engineered graph distances to quantify dissimilarity between networks. In this work we provide an unsupervised approach to learn embedding representation for a collection of graphs so that it can be used in numerous graph mining tasks. By using an unsupervised neural network approach on input graphs, we aim to capture the underlying distribution of the data in order to discriminate between different class of networks. Our method is assessed empirically on synthetic and real life datasets and evaluated in three different tasks: graph clustering, visualization and classification. Results reveal that our method outperforms well known graph distances and graph-kernels in clustering and classification tasks, being highly efficient in runtime., Comment: 17 pages, 8 figures, 4 tables

Several social, medical, engineering and biological challenges rely on discovering the functionality of networks from their structure and node metadata, when it is available. For example, in chemoinformatics one might want to detect whether a molecule is toxic based on structure and atomic types, or discover the research field of a scientific collaboration network. Existing techniques rely on counting or measuring structural patterns that are known to show large variations from network to network, such as the number of triangles, or the assortativity of node metadata. We introduce the concept of multi-hop assortativity, that captures the similarity of the nodes situated at the extremities of a randomly selected path of a given length. We show that multi-hop assortativity unifies various existing concepts and offers a versatile family of 'fingerprints' to characterize networks. These fingerprints allow in turn to recover the functionalities of a network, with the help of the machine learning toolbox. Our method is evaluated empirically on established social and chemoinformatic network benchmarks. Results reveal that our assortativity based features are competitive providing highly accurate results often outperforming state of the art methods for the network classification task., Comment: 20 pages, 7 figures. arXiv admin note: text overlap with arXiv:1705.10817

Magnetic skyrmions are promising candidates as information carriers in spintronic devices. The transport of individual skyrmions in a fast and controlled way is a key issue in this field. Here we introduce a novel platform for accelerating, guiding and compressing skyrmions along predefined paths. The guiding mechanism is based on two parallel defect-lines (rails), one attractive and the other repulsive. Numerical simulations, using parameters from state-of-the-art experiments, show that the speed of the skyrmions along the rails is increased up to a factor of ten with respect to the non-defect case whereas the distance between rails can be as small as the initial radius of the skyrmions. In this way, the flux of information that can be coded and transported with magnetic skyrmions could be significantly increased., Comment: 6 pages, 4 figures

Numerous social, medical, engineering and biological challenges can be framed as graph-based learning tasks. Here, we propose a new feature based approach to network classification. We show how dynamics on a network can be useful to reveal patterns about the organization of the components of the underlying graph where the process takes place. We define generalized assortativities on networks and use them as generalized features across multiple time scales. These features turn out to be suitable signatures for discriminating between different classes of networks. Our method is evaluated empirically on established network benchmarks. We also introduce a new dataset of human brain networks (connectomes) and use it to evaluate our method. Results reveal that our dynamics based features are competitive and often outperform state of the art accuracies., Comment: This paper is under review as a conference paper at ECML-PKDD 2017