Home » Uncategorized

UberCloud Voice April 2017

Testing Docker Container Performance for HPC Applications

Docker vs KVMThis article presents actual benchmark research by Alexey Ermakov and Alexey Vasyukov. The main goal for this article is to compare performance penalties when using KVM virtualization and Docker containers for creating isolated environments for HPC applications. The article provides both data obtained using synthetic tests (High Performance Linpack) and real life applications (OpenFOAM). The article highlights the influence on performance of major infrastructure configuration options – CPU type presented to VM, networking connection type used. As one of the results the authors strongly recommended to run benchmarks on exact application you’re going to run when considering virtualization or containerization as an option for HPC. Another important result is that a real «non-synthetic» application like OpenFOAM is more than 2 times faster in Docker containers than the same application in KVM. Read the article HERE.

HPC Workflows Using Containers

Shifter videoIn this video from Switzerland HPC Conference, Dr. Lucas Benedicic & Dr. Felipe A. Cruz from CSCS-ETHZ discuss a workflow for building and testing Docker containers and their deployment on an HPC system using Shifter. Docker is widely used by developers as a powerful tool for standardizing the packaging of applications across multiple environments, which greatly eases the porting efforts. On the other hand, Shifter provides a container runtime that has been specifically built to fit the needs of HPC. We will briefly introduce these tools while discussing the advantages of using these technologies to fulfill the needs of specific workflows for HPC, e.g., security, high-performance, portability and parallel scalability. We will support these points by presenting benchmark results that also highlight the use of specialized hardware like Cray Aries interconnect and NVIDIA Tesla P100 GPU accelerators. Watch the video on insideHPC HERE.

Hyperion (IDC) Paints a Bullish Picture of HPC Future and of Cloud

This HPCwire article from John Russell provides an excellent status of the actual HPC market and its trends. Hyperion Research – formerly IDC’s HPC group – painted a fascinating and complicated portrait of the HPC community’s health and prospects at the recent HPC User Forum held in Albuquerque, NM. HPC sales are up and growing ($22 billion, all HPC segments, 2016). The new kid on the block – all things ‘big’ data driven – is becoming an adolescent and behaving accordingly. And HPC Return on Investment (ROI), at least as measured by Hyperion, is $551 per $1 invested (revenue growth) and $52 per $1 of profit invested. In earlier studies HPC user willingness to deploy in the cloud has often seemed tepid. Costs, security, adequate performance were all concerns, especially so in public cloud. Hyperion suggested attitudes seem to be changing and reported a jump in the number of HPC sites using public clouds – 64 percent now up from 13 percent in 2011, but cautioned that the size and number of jobs were still limited to a small proportion. Conversely, private and hybrid cloud use was growing fast and held more near-term promise. Read all the details HERE.

Top Trends in Industrial Design: From Green Engineering to the Cloud and Beyond

SimScale for April 2017“The era of cloud-based product design is upon us,” says Gartner in their Cool Vendors in Product Design and Lifecycle Management 2016 report. What’s driving this massive transformation? The desire for flexibility and accessibility, increased collaboration, ease of use and serious cost reductions. In a recent article by Machine Design on trends in industrial design engineering, two are being shaped by the rise of cloud technology: CAD software: About 30% of the CAD enterprise software market will be software as a service (SaaS) by the end of 2018 says the IDC;  and  Computer-aided engineering (CAE): Until recently, CAE required tens of thousands of dollars in hardware and software investments. With the rise of the cloud, this technology is becoming more affordable and, as a result, more accessible. Why does this matter to design engineers? Read the answers from SimScale’s Netty Zaharia who spotlights the five industrial-design engineering trends to watch out for in 2017, in the Machine Design article about Top Trends in Industrial Design: From Green Engineering to the Cloud and Beyond.

Cloud Solution Architectures

Azure big compute architectureMicrosoft published over 20 different cloud solution architectures to help customers design and implement secure, highly-available, performant and resilient solutions on Azure. Although these architectures are especially designed for the Azure cloud they are a great collection inspiring and valid at high-level also for other clouds. Among them are three cloud solution architectures which are focusing on High Performance Computing (HPC) applications which can scale to thousands of compute cores, extend on-premises big compute, or run as a 100% cloud native solution. These HPC solutions run in Azure with no hardware infrastructure to maintain. i.e. Big compute solutions as a serviceHPC cluster deployed in the cloud, and On-premises HPC implementation bursting to Azure.

Webinar Recording available: HPC Cloud in Oil & Gas – High Performance Computing made simple and affordable

Team 99 small

Join us at this HPC Cloud Webinar on April 12 and learn about status and developments of HPC and Cloud in Oil & Gas. Find out how easy it is today to access and use HPC Cloud to improve and speedup your research. Based on 195 engineering cloud experiments we present our experience, lessons learned & recommendations on how to best use cloud, scaling to 100s of cores, allowing for complex geometries processed in less time. We demonstrate these advantages with a case study from our Team 99 about “North Sea asset life extension assessing impacts on helicopter operations”. To engineers, this cloud resource looks like your home desktop computer, no need to learn anything new, and still your jobs and data can be accessed and used in the same way, at your fingertips. The webinar recording is available for download HERE

© Copyright 2020 UberCloud™ - UberCloud is a trademark of TheUberCloud, INC. | Privacy Policy
Logo Header Menu

Credit Card Help

UberCloud Marketplace accepts international payments for products. All prices are in United States Dollars (USD $) unless specifically stated otherwise. We accept the following major card brands.
The CVC is usually the three digit number located on the back of the credit card. On some cards, the CVC is displayed on the front of the card and can be 4 digits long.
We take your privacy and security seriously. See our privacy policy, which defines the measures we take to secure your private information.
We use a well known payment processor based in San Francisco, California called Stripe. Your credit card information is processed by Stripe securely and we do not retain your credit card information on UberCloud's servers.