Engineering topics

Garzoglio, Gabriele

Publications and source records attributed to Garzoglio, Gabriele.

11 records

DOE OSTI2017

HEPCloud, a New Paradigm for HEP Facilities: CMS Amazon Web Services Investigation

Historically, high energy physics computing has been performed on large purpose-built computing systems. These began as single-site compute facilities, but have evolved into the distributed computing grids used today. Recently, there has been an exponential increase in the capacity and capability of commercial clouds. Cloud resources are highly virtualized and intended to be able to be flexibly deployed for a variety of computing tasks. There is a growing interest among the cloud providers to demonstrate the capability to perform large-scale scientific computing. In this paper, we discuss results from the CMS experiment using the Fermilab HEPCloud facility, which utilized both local Fermilab resources and virtual machines in the Amazon Web Services Elastic Compute Cloud. We discuss the planning, technical challenges, and lessons learned involved in performing physics workflows on a large-scale set of virtualized resources. Additionally, we will discuss the economics and operational efficiencies when executing workflows both in the cloud and on dedicated resources.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2016

A Reference Model for Virtual Machine Launching Overhead

Cloud bursting is one of the key research topics in the cloud computing communities. A well designed cloud bursting module enables private clouds to automatically launch virtual machines (VMs) to public clouds when more resources are needed. One of the main challenges in developing a cloud bursting module is to decide when and where to launch a VM so that all resources are most effectively and efficiently utilized and the system performance is optimized. However, based on system operational data obtained from FermiCloud, a private cloud developed by the Fermi National Accelerator Laboratory for scientific workflows, the VM launching overhead is not a constant. It varies with physical resource utilization, such as CPU and I/O device utilizations, at the time when a VM is launched. Hence, to make judicious decisions as to when and where a VM should be launched, a VM launching overhead reference model is needed. In this paper, we first develop a VM launching overhead reference model based on operational data we have obtained on FermiCloud. Second, we apply the developed reference model on FermiCloud and compare calculated VM launching overhead values based on the model with measured overhead values on FermiCloud. Our empirical results on FermiCloud indicate that the developed reference model is accurate. We believe, with the guidance of the developed reference model, efficient resource allocation algorithms can be developed for cloud bursting process to minimize the operational cost and resource waste.

Wu, Hao↗

DOE OSTI2016

Virtual Facility at Fermilab: Infrastructure and Services Expand to Public Clouds

In preparation for its new Virtual Facility Project, Fermilab has launched a program of work to determine the requirements for running a computation facility on-site, in public clouds, or a combination of both. This program builds on the work we have done to successfully run experimental workflows of 1000-VM scale both on an on-site private cloud and on Amazon AWS. To do this at scale we deployed dynamically launched and discovered caching services on the cloud. We are now testing the deployment of more complicated services on Amazon AWS using native load balancing and auto scaling features they provide. The Virtual Facility Project will design and develop a facility including infrastructure and services that can live on the site of Fermilab, off-site, or a combination of both. We expect to need this capacity to meet the peak computing requirements in the future. The Virtual Facility is intended to provision resources on the public cloud on behalf of the facility as a whole instead of having each experiment or Virtual Organization do it on their own. We will describe the policy aspects of a distributed Virtual Facility, the requirements, and plans to make a detailed comparison of the relative cost of the public and private clouds. Furthermore, this talk will present the details of the technical mechanisms we have developed to date, and the plans currently taking shape for a Virtual Facility at Fermilab.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2015

Progress on the FabrIc for Frontier Experiments project at Fermilab

The FabrIc for Frontier Experiments (FIFE) project is an ambitious, major-impact initiative within the Fermilab Scientific Computing Division designed to lead the computing model for Fermilab experiments. FIFE is a collaborative effort between experimenters and computing professionals to design and develop integrated computing models for experiments of varying needs and infrastructure. The major focus of the FIFE project is the development, deployment, and integration of Open Science Grid solutions for high throughput computing, data management, database access and collaboration within experiment. To accomplish this goal, FIFE has developed workflows that utilize Open Science Grid sites along with dedicated and commercial cloud resources. The FIFE project has made significant progress integrating into experiment computing operations several services including new job submission services, software and reference data distribution through CVMFS repositories, flexible data transfer client, and access to opportunistic resources on the Open Science Grid. Hence, the progress with current experiments and plans for expansion with additional projects will be discussed. FIFE has taken a leading role in the definition of the computing model for Fermilab experiments, aided in the design of computing for experiments beyond Fermilab, and will continue to define the future direction of high throughput computing for future physics experiments worldwide

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Data Intensive Scientific Workflows on a Federated Cloud: CRADA Final Report

The Fermilab Scientific Computing Division and the KISTI Global Science Experimental Data Hub Center have built a prototypical large-scale infrastructure to handle scientific workflows of stakeholders to run on multiple cloud resources. The demonstrations have been in the areas of (a) Data-Intensive Scientific Workflows on Federated Clouds, (b) Interoperability and Federation of Cloud Resources, and (c) Virtual Infrastructure Automation to enable On-Demand Services.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2015

Understanding the Performance and Potential of Cloud Computing for Scientific Applications

In this paper, commercial clouds bring a great opportunity to the scientific computing area. Scientific applications usually require significant resources, however not all scientists have access to sufficient high-end computing systems, may of which can be found in the Top500 list. Cloud Computing has gained the attention of scientists as a competitive resource to run HPC applications at a potentially lower cost. But as a different infrastructure, it is unclear whether clouds are capable of running scientific applications with a reasonable performance per money spent. This work studies the performance of public clouds and places this performance in context to price. We evaluate the raw performance of different services of AWS cloud in terms of the basic resources, such as compute, memory, network and I/O. We also evaluate the performance of the scientific applications running in the cloud. This paper aims to assess the ability of the cloud to perform well, as well as to evaluate the cost of the cloud running scientific applications. We developed a full set of metrics and conducted a comprehensive performance evlauation over the Amazon cloud. We evaluated EC2, S3, EBS and DynamoDB among the many Amazon AWS services. We evaluated the memory sub-system performance with CacheBench, the network performance with iperf, processor and network performance with the HPL benchmark application, and shared storage with NFS and PVFS in addition to S3. We also evaluated a real scientific computing application through the Swift parallel scripting system at scale. Armed with both detailed benchmarks to gauge expected performance and a detailed monetary cost analysis, we expect this paper will be a recipe cookbook for scientists to help them decide where to deploy and run their scientific applications between public clouds, private clouds, or hybrid clouds.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Modeling the Virtual Machine Launching Overhead under Fermicloud

FermiCloud is a private cloud developed by the Fermi National Accelerator Laboratory for scientific workflows. The Cloud Bursting module of the FermiCloud enables the FermiCloud, when more computational resources are needed, to automatically launch virtual machines to available resources such as public clouds. One of the main challenges in developing the cloud bursting module is to decide when and where to launch a VM so that all resources are most effectively and efficiently utilized and the system performance is optimized. However, based on FermiCloud’s system operational data, the VM launching overhead is not a constant. It varies with physical resource (CPU, memory, I/O device) utilization at the time when a VM is launched. Hence, to make judicious decisions as to when and where a VM should be launched, a VM launch overhead reference model is needed. The paper is to develop a VM launch overhead reference model based on operational data we have obtained on FermiCloud and uses the reference model to guide the cloud bursting process.

Garzoglio, Gabriele↗

DOE OSTI2014

Overhead-Aware-Best-Fit (OABF) Resource Allocation Algorithm for Minimizing VM Launching Overhead

FermiCloud is a private cloud developed in Fermi National Accelerator Laboratory to provide elastic and on-demand resources for different scientific research experiments. The design goal of the FermiCloud is to automatically allocate resources for different scientific applications so that the QoS required by these applications is met and the operational cost of the FermiCloud is minimized. Our earlier research shows that VM launching overhead has large variations. If such variations are not taken into consideration when making resource allocation decisions, it may lead to poor performance and resource waste. In this paper, we show how we may use an VM launching overhead reference model to minimize VM launching overhead. In particular, we first present a training algorithm that automatically tunes a given refer- ence model to accurately reflect FermiCloud environment. Based on the tuned reference model for virtual machine launching overhead, we develop an overhead-aware-best-fit resource allocation algorithm that decides where and when to allocate resources so that the average virtual machine launching overhead is minimized. The experimental results indicate that the developed overhead-aware-best-fit resource allocation algorithm can significantly improved the VM launching time when large number of VMs are simultaneously launched.

Wu, Hao↗

DOE OSTI2014

Exploring Infiniband Hardware Virtualization in OpenNebula towards Efficient High-Performance Computing

has been widely accepted that software virtualization has a big negative impact on high-performance computing (HPC) application performance. This work explores the potential use of Infiniband hardware virtualization in an OpenNebula cloud towards the efficient support of MPI-based workloads. We have implemented, deployed, and tested an Infiniband network on the FermiCloud private Infrastructure-as-a-Service (IaaS) cloud. To avoid software virtualization towards minimizing the virtualization overhead, we employed a technique called Single Root Input/Output Virtualization (SRIOV). Our solution spanned modifications to the Linux’s Hypervisor as well as the OpenNebula manager. We evaluated the performance of the hardware virtualization on up to 56 virtual machines connected by up to 8 DDR Infiniband network links, with micro-benchmarks (latency and bandwidth) as well as w a MPI-intensive application (the HPL Linpack benchmark).

Pais Pitta de Lacerda Ruivo, Tiago↗

DOE OSTI2014

Enabling On-Demand Scientific Workflows on a Federated Cloud

The Fermilab Grid and Cloud Computing Department and the KISTI Global Science experimental Data hub Center are working on a multi-year Collaborative Research and Development Agreement.With the knowledge developed in the first year on how to provision and manage a federation of virtual machines through Cloud management systems. In this second year, we expanded the work on provisioning and federation, increasing both scale and diversity of solutions, and we started to build on-demand services on the established fabric, introducing the paradigm of Platform as a Service to assist with the execution of scientific workflows. We have enabled scientific workflows of stakeholders to run on multiple cloud resources at the scale of 1,000 concurrent machines. The demonstrations have been in the areas of (a) Virtual Infrastructure Automation and Provisioning, (b) Interoperability and Federation of Cloud Resources, and (c) On-demand Services for ScientificWorkflows.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Big Data over a 100G network at Fermilab

As the need for Big Data in science becomes ever more relevant, networks around the world are upgrading their infrastructure to support high-speed interconnections. To support its mission, the high-energy physics community as a pioneer in Big Data has always been relying on the Fermi National Accelerator Laboratory to be at the forefront of storage and data movement. This need was reiterated in recent years with the data-taking rate of the major LHC experiments reaching tens of petabytes per year. At Fermilab, this resulted regularly in peaks of data movement on the Wide area network (WAN) in and out of the laboratory of about 30 Gbit/s and on the Local are network (LAN) between storage and computational farms of 160 Gbit/s. To address these ever increasing needs, as of this year Fermilab is connected to the Energy Sciences Network (ESnet) through a 100 Gb/s link. To understand the optimal system-and application-level configuration to interface computational systems with the new highspeed interconnect, Fermilab has deployed a Network Research & Development facility connected to the ESnet 100G Testbed. For the past two years, the High Throughput Data Program (HTDP) has been using the Testbed to identify gaps in data movement middleware [5] when transferring data at these high-speeds. The program has published evaluations of technologies typically used in High Energy Physics, such as GridFTP [4], XrootD [9], and Squid [8]. Furthermore, this work presents the new R&D facility and the continuation of the evaluation program.

97 MATHEMATICS AND COMPUTING↗