arXiv subjects

Marco A. S. Netto

Publications and source records attributed to Marco A. S. Netto.

18 recordsLinked to original sources

arXiv2024

Simplifying HPC resource selection: A tool for optimizing execution time and cost on Azure

Azure Cloud offers a wide range of resources for running HPC workloads, requiring users to configure their deployment by selecting VM types, number of VMs, and processes per VM. Suboptimal decisions may lead to longer execution times or additional costs for the user. We are developing an open-source tool to assist users in making these decisions by considering application input parameters, as they influence resource consumption. The tool automates the time-consuming process of setting up the cloud environment, executing the benchmarking runs, handling output, and providing users with resource selection recommendations as high level insights on run times and costs across different VM types and number of VMs. In this work, we present initial results and insights on reducing the number of cloud executions needed to provide such guidance, leveraging data analytics and optimization techniques with two well-known HPC applications: OpenFOAM and LAMMPS.

cs.DC↗

arXiv2024

HPCAdvisor: A Tool for Assisting Users in Selecting HPC Resources in the Cloud

Cloud platforms are increasingly being used to run HPC workloads. Major cloud providers offer a wide variety of virtual machine (VM) types, enabling users to find the optimal balance between performance and cost. However, this extensive selection of VM types can also present challenges, as users must decide not only which VM types to use but also how many nodes are required for a given workload. Although benchmarking data is available for well-known applications from major cloud providers, the choice of resources is also influenced by the specifics of the user's application input. This paper presents the vision and current implementation of HPCAdvisor, a tool designed to assist users in defining their HPC clusters in the cloud. It considers the application's input and utilizes a major cloud provider as a use case for its back-end component.

cs.DC↗

arXiv2021

Workflow Provenance in the Lifecycle of Scientific Machine Learning

Machine Learning (ML) has already fundamentally changed several businesses. More recently, it has also been profoundly impacting the computational science and engineering domains, like geoscience, climate science, and health science. In these domains, users need to perform comprehensive data analyses combining scientific data and ML models to provide for critical requirements, such as reproducibility, model explainability, and experiment data understanding. However, scientific ML is multidisciplinary, heterogeneous, and affected by the physical constraints of the domain, making such analyses even more challenging. In this work, we leverage workflow provenance techniques to build a holistic view to support the lifecycle of scientific ML. We contribute with (i) characterization of the lifecycle and taxonomy for data analyses; (ii) design principles to build this view, with a W3C PROV compliant data representation and a reference system architecture; and (iii) lessons learned after an evaluation in an Oil & Gas case using an HPC cluster with 393 nodes and 946 GPUs. The experiments show that the principles enable queries that integrate domain semantics with ML models while keeping low overhead (<1%), high scalability, and an order of magnitude of query acceleration under certain workloads against without our representation.

cs.DB↗

arXiv2021

Context-aware Execution Migration Tool for Data Science Jupyter Notebooks on Hybrid Clouds

Interactive computing notebooks, such as Jupyter notebooks, have become a popular tool for developing and improving data-driven models. Such notebooks tend to be executed either in the user's own machine or in a cloud environment, having drawbacks and benefits in both approaches. This paper presents a solution developed as a Jupyter extension that automatically selects which cells, as well as in which scenarios, such cells should be migrated to a more suitable platform for execution. We describe how we reduce the execution state of the notebook to decrease migration time and we explore the knowledge of user interactivity patterns with the notebook to determine which blocks of cells should be migrated. Using notebooks from Earth science (remote sensing), image recognition, and hand written digit identification (machine learning), our experiments show notebook state reductions of up to 55x and migration decisions leading to performance gains of up to 3.25x when the user interactivity with the notebook is taken into consideration.

cs.DC↗

arXiv2019

Provenance Data in the Machine Learning Lifecycle in Computational Science and Engineering

Machine Learning (ML) has become essential in several industries. In Computational Science and Engineering (CSE), the complexity of the ML lifecycle comes from the large variety of data, scientists' expertise, tools, and workflows. If data are not tracked properly during the lifecycle, it becomes unfeasible to recreate a ML model from scratch or to explain to stakeholders how it was created. The main limitation of provenance tracking solutions is that they cannot cope with provenance capture and integration of domain and ML data processed in the multiple workflows in the lifecycle while keeping the provenance capture overhead low. To handle this problem, in this paper we contribute with a detailed characterization of provenance data in the ML lifecycle in CSE; a new provenance data representation, called PROV-ML, built on top of W3C PROV and ML Schema; and extensions to a system that tracks provenance from multiple workflows to address the characteristics of ML and CSE, and to allow for provenance queries with a standard vocabulary. We show a practical use in a real case in the Oil and Gas industry, along with its evaluation using 48 GPUs in parallel.

cs.DC↗

arXiv2018

A Scalable Machine Learning System for Pre-Season Agriculture Yield Forecast

Yield forecast is essential to agriculture stakeholders and can be obtained with the use of machine learning models and data coming from multiple sources. Most solutions for yield forecast rely on NDVI (Normalized Difference Vegetation Index) data, which is time-consuming to be acquired and processed. To bring scalability for yield forecast, in the present paper we describe a system that incorporates satellite-derived precipitation and soil properties datasets, seasonal climate forecasting data from physical models and other sources to produce a pre-season prediction of soybean/maize yield---with no need of NDVI data. This system provides significantly useful results by the exempting the need for high-resolution remote-sensing data and allowing farmers to prepare for adverse climate influence on the crop cycle. In our studies, we forecast the soybean and maize yields for Brazil and USA, which corresponded to 44% of the world's grain production in 2016. Results show the error metrics for soybean and maize yield forecasts are comparable to similar systems that only provide yield forecast information in the first weeks to months of the crop cycle.

cs.CY↗

arXiv2018

A Manifesto for Future Generation Cloud Computing: Research Directions for the Next Decade

The Cloud computing paradigm has revolutionised the computer science horizon during the past decade and has enabled the emergence of computing as the fifth utility. It has captured significant attention of academia, industries, and government bodies. Now, it has emerged as the backbone of modern economy by offering subscription-based services anytime, anywhere following a pay-as-you-go model. This has instigated (1) shorter establishment times for start-ups, (2) creation of scalable global enterprise applications, (3) better cost-to-value associativity for scientific and high performance computing applications, and (4) different invocation/execution models for pervasive and ubiquitous applications. The recent technological developments and paradigms such as serverless computing, software-defined networking, Internet of Things, and processing at network edge are creating new opportunities for Cloud computing. However, they are also posing several new challenges and creating the need for new approaches and research strategies, as well as the re-evaluation of the models that were developed to address issues such as scalability, elasticity, reliability, security, sustainability, and application models. The proposed manifesto addresses them by identifying the major open challenges in Cloud computing, emerging trends, and impact areas. It then offers research directions for the next decade, thus helping in the realisation of Future Generation Cloud Computing.

cs.DC↗

arXiv2018

DeepDownscale: a Deep Learning Strategy for High-Resolution Weather Forecast

Running high-resolution physical models is computationally expensive and essential for many disciplines. Agriculture, transportation, and energy are sectors that depend on high-resolution weather models, which typically consume many hours of large High Performance Computing (HPC) systems to deliver timely results. Many users cannot afford to run the desired resolution and are forced to use low resolution output. One simple solution is to interpolate results for visualization. It is also possible to combine an ensemble of low resolution models to obtain a better prediction. However, these approaches fail to capture the redundant information and patterns in the low-resolution input that could help improve the quality of prediction. In this paper, we propose and evaluate a strategy based on a deep neural network to learn a high-resolution representation from low-resolution predictions using weather forecast as a practical use case. We take a supervised learning approach, since obtaining labeled data can be done automatically. Our results show significant improvement when compared with standard practices and the strategy is still lightweight enough to run on modest computer systems.

cs.LG↗

arXiv2018

JobPruner: A Machine Learning Assistant for Exploring Parameter Spaces in HPC Applications

High Performance Computing (HPC) applications are essential for scientists and engineers to create and understand models and their properties. These professionals depend on the execution of large sets of computational jobs that explore combinations of parameter values. Avoiding the execution of unnecessary jobs brings not only speed to these experiments, but also reductions in infrastructure usage---particularly important due to the shift of these applications to HPC cloud platforms. Our hypothesis is that data generated by these experiments can help users in identifying such jobs. To address this hypothesis we need to understand the similarity levels among multiple experiments necessary for job elimination decisions and the steps required to automate this process. In this paper we present a study and a machine learning-based tool called JobPruner to support parameter exploration in HPC experiments. The tool was evaluated with three real-world use cases from different domains including seismic analysis and agronomy. We observed the tool reduced 93% of jobs in a single experiment, while improving quality in most scenarios. In addition, reduction in job executions was possible even considering past experiments with low correlations.

cs.DC↗

arXiv2018

HPC Cloud for Scientific and Business Applications: Taxonomy, Vision, and Research Challenges

High Performance Computing (HPC) clouds are becoming an alternative to on-premise clusters for executing scientific applications and business analytics services. Most research efforts in HPC cloud aim to understand the cost-benefit of moving resource-intensive applications from on-premise environments to public cloud platforms. Industry trends show hybrid environments are the natural path to get the best of the on-premise and cloud resources---steady (and sensitive) workloads can run on on-premise resources and peak demand can leverage remote resources in a pay-as-you-go manner. Nevertheless, there are plenty of questions to be answered in HPC cloud, which range from how to extract the best performance of an unknown underlying platform to what services are essential to make its usage easier. Moreover, the discussion on the right pricing and contractual models to fit small and large users is relevant for the sustainability of HPC clouds. This paper brings a survey and taxonomy of efforts in HPC cloud and a vision on what we believe is ahead of us, including a set of research challenges that, once tackled, can help advance businesses and scientific discoveries. This becomes particularly relevant due to the fast increasing wave of new HPC applications coming from big data and artificial intelligence.

cs.DC↗

arXiv2016

Helping HPC Users Specify Job Memory Requirements via Machine Learning

Resource allocation in High Performance Computing (HPC) settings is still not easy for end-users due to the wide variety of application and environment configuration options. Users have difficulties to estimate the number of processors and amount of memory required by their jobs, select the queue and partition, and estimate when job output will be available to plan for next experiments. Apart from wasting infrastructure resources by making wrong allocation decisions, overall user response time can also be negatively impacted. Techniques that exploit batch scheduler systems to predict waiting time and runtime of user jobs have already been proposed. However, we observed that such techniques are not suitable for predicting job memory usage. In this paper we introduce a tool to help users predict their memory requirements using machine learning. We describe the integration of the tool with a batch scheduler system, discuss how batch scheduler log data can be exploited to generate memory usage predictions through machine learning, and present results of two production systems containing thousands of jobs.

cs.DC↗

arXiv2016

SLA-aware Interactive Workflow Assistant for HPC Parameter Sweeping Experiments

A common workflow in science and engineering is to (i) setup and deploy large experiments with tasks comprising an application and multiple parameter values; (ii) generate intermediate results; (iii) analyze them; and (iv) reprioritize the tasks. These steps are repeated until the desired goal is achieved, which can be the evaluation/simulation of complex systems or model calibration. Due to time and cost constraints, sweeping all possible parameter values of the user application is not always feasible. Experimental Design techniques can help users reorganize submission-execution-analysis workflows to bring a solution in a more timely manner. This paper introduces a novel tool that leverages users' feedback on analyzing intermediate results of parameter sweeping experiments to advise them about their strategies on parameter selections tied to their SLA constraints. We evaluated our tool with three applications of distinct domains and search space shapes. Our main finding is that users with submission-execution-analysis workflows can benefit from their interaction with intermediate results and adapt themselves according to their domain expertise and SLA constraints.

cs.DC↗

arXiv2016

Job Placement Advisor Based on Turnaround Predictions for HPC Hybrid Clouds

Several companies and research institutes are moving their CPU-intensive applications to hybrid High Performance Computing (HPC) cloud environments. Such a shift depends on the creation of software systems that help users decide where a job should be placed considering execution time and queue wait time to access on-premise clusters. Relying blindly on turnaround prediction techniques will affect negatively response times inside HPC cloud environments. This paper introduces a tool to make job placement decisions in HPC hybrid cloud environments taking into account the inaccuracy of execution and waiting time predictions. We used job traces from real supercomputing centers to run our experiments, and compared the performance between environments using real speedup curves. We also extended a state-of-the-art machine learning based predictor to work with data from the cluster scheduler. Our main findings are: (i) depending on workload characteristics, there is a turning point where predictions should be disregarded in favor of a more conservative decision to minimize job turnaround times and (ii) scheduler data plays a key role in improving predictions generated with machine learning using job trace data---our experiments showed around 20% prediction accuracy improvements.

cs.DC↗

arXiv2016

Exploiting Workload Cycles for Orchestration of Virtual Machine Live Migrations in Clouds

Virtual machine live migration in cloud environments aims at reducing energy costs and increasing resource utilization. However, its potential has not been fully explored because of simultaneous migrations that may cause user application performance degradation and network congestion. Research efforts on live migration orchestration policies still mostly rely on system level metrics. This work introduces an Application-aware Live Migration Architecture (ALMA) that selects suitable moments for migrations using application characterization data. This characterization consists in recognizing resource usage cycles via Fast Fourier Transform. From our experiments, live migration times were reduced by up to 74% for benchmarks and by up to 67% for real applications, when compared to migration policies with no application workload analysis. Network data transfer during the live migration was reduced by up to 62%.

cs.DC↗

arXiv2015

An SLA-based Advisor for Placement of HPC Jobs on Hybrid Clouds

Several scientific and industry applications require High Performance Computing (HPC) resources to process and/or simulate complex models. Not long ago, companies, research institutes, and universities used to acquire and maintain on-premise computer clusters; but, recently, cloud computing has emerged as an alternative for a subset of HPC applications. This poses a challenge to end-users, who have to decide where to run their jobs: on local clusters or burst to a remote cloud service provider. While current research on HPC cloud has focused on comparing performance of on-premise clusters against cloud resources, we build on top of existing efforts and introduce an advisory service to help users make this decision considering the trade-offs of resource costs, performance, and availability on hybrid clouds. We evaluated our service using a real test-bed with a seismic processing application based on Full Waveform Inversion; a technique used by geophysicists in the oil & gas industry and earthquake prediction. We also discuss how the advisor can be used for other applications and highlight the main lessons learned constructing this service to reduce costs and turnaround times.

cs.DC↗

arXiv2015

Using Application Data for SLA-aware Auto-scaling in Cloud Environments

With the establishment of cloud computing as the environment of choice for most modern applications, auto-scaling is an economic matter of great importance. For applications like stream computing that process ever changing amounts of data, modifying the number and configuration of resources to meet performance requirements becomes essential. Current solutions on auto-scaling are mostly rule-based using infrastructure level metrics such as CPU/memory/network utilization, and system level metrics such as throughput and response time. In this paper, we introduce a study on how effective auto-scaling can be using data generated by the application itself. To make this assessment, two algorithms are proposed that use a priori knowledge of the data stream and use sentiment analysis from soccer-related tweets, triggering auto-scaling operations according to rapid changes in the public sentiment about the soccer players that happens just before big bursts of messages. Our application-based auto-scaling was able to reduce the number of SLA violations by up to 95% and reduce resource requirements by up to 33%.

cs.DC↗

arXiv2014

Big Data Computing and Clouds: Trends and Future Directions

This paper discusses approaches and environments for carrying out analytics on Clouds for Big Data applications. It revolves around four important areas of analytics and Big Data, namely (i) data management and supporting architectures; (ii) model development and scoring; (iii) visualisation and user interaction; and (iv) business models. Through a detailed survey, we identify possible gaps in technology and provide recommendations for the research community on future directions on Cloud-supported Big Data computing and analytics solutions.

cs.DC↗

arXiv2013

Patience-aware Scheduling for Cloud Services: Freeing Users from the Chains of Boredom

Scheduling of service requests in Cloud computing has traditionally focused on the reduction of pre-service wait, generally termed as waiting time. Under certain conditions such as peak load, however, it is not always possible to give reasonable response times to all users. This work explores the fact that different users may have their own levels of tolerance or patience with response delays. We introduce scheduling strategies that produce better assignment plans by prioritising requests from users who expect to receive the results earlier and by postponing servicing jobs from those who are more tolerant to response delays. Our analytical results show that the behaviour of users' patience plays a key role in the evaluation of scheduling techniques, and our computational evaluation demonstrates that, under peak load, the new algorithms typically provide better user experience than the traditional FIFO strategy.

cs.DC↗