DOCSLIB.ORG

Baremetal with Apache Cloudstack Apachecon Europe 2016

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text

Load more

Baremetal with Apache CloudStack ApacheCon Europe 2016 Jaydeep Marfatia Cloud, IOT and Analytics Me Director of Product Management Cloud Products Accelerite Background Project lead for open source project XenMan/ConVirt Co-Founder Convirture Corp. Architect for Oracle 10g Enterprise Manager 2 © 2016 Accelerite. All Rights Reserved. Apache CloudStack • Highly available, highly scalable Infrastructure as a Service (IaaS) cloud computing platform • Easy to deploy, turnkey solution that includes the entire "stack" of features most organizations want with an IaaS cloud • Used by many enterprises for their private cloud, as well as for running large public clouds ! • CloudStack currently supports the most popular hypervisors: VMware, KVM, Citrix XenServer, Hyper-V and more 3 © 2016 Accelerite. All Rights Reserved. Accelerite • Enterprise Infrastructure software company • Acquired CloudPlatform from Citrix early this year • Commercial product based on CloudStack • 50+ engineers and adding more 4 © 2016 Accelerite. All Rights Reserved. Recent contributions by Accelerite • Template upload from browser • DHCP/DNS Offload • LDAP integration • CoreOS/Docker on CloudStack • VPN enhancements • VMWare Networking improvements • Baremetal enhancements • 200+ Bug fixes in the product 5 © 2016 Accelerite. All Rights Reserved. CloudStack User Interface User Console 7 © 2016 Accelerite. All Rights Reserved. Admin Console 8 © 2016 Accelerite. All Rights Reserved. Self-service VM Provisioning Zone Template Compute Disk Affinity Network Launch 9 © 2016 Accelerite. All Rights Reserved. Monitoring Cloud Infrastructure 10 © 2016 Accelerite. All Rights Reserved. © 2016 Accelerite. All rights reserved. Baremetal Team Team • Harikrishna Patnala – Apache Committer, working on CloudStack 4+ years Expertise : Baremetal and Virtual Router • Jayapal Uradi – Apache Committer, 4+ years Expertise : CloudStack Networking • Suresh Sadhu – Quality Assurance of CloudStack, 5+ years Expertise : All rounder 12 © 2016 Accelerite. All Rights Reserved. Motivation for Baremetal Why Baremetal? • Meet high performance computing requirements • Hadoop, Spark, HPC • Take advantage of specialized hardware • GPUs, Blade Micro-servers • Databases • Virtual Desktop Infrastructure management • Desired for Docker Containers 14 © 2016 Accelerite. All Rights Reserved. Requirements Data Center 16 © 2016 Accelerite. All Rights Reserved. Data Center Switch Servers Rack Secondary Storage 17 © 2016 Accelerite. All Rights Reserved. Workload Partitioning Switch Hadoop Rack Kubernetes Secondary Storage 18 © 2016 Accelerite. All Rights Reserved. Multi-Tenancy Switch Tenant 1 Rack Tenant 2 Secondary Storage 19 © 2016 Accelerite. All Rights Reserved. Multi-Tenancy and Workload partitioning Switch Tenant 1 Rack Tenant 1 - DB Tenant 2 Secondary Storage 20 © 2016 Accelerite. All Rights Reserved. Data Center – VMs and Baremetal Rack Tenant 1 Network Tenant 2 Network 21 © 2016 Accelerite. All Rights Reserved. Baremetal in CloudStack • Ability to deploy Baremetal servers as easy as spinning up a VM • Multi-tenant Baremetal orchestration with network isolation • Flexibility of using mixed environments of VM and Baremetal • DB tier on Baremetal and other tiers on VMs for a multi-tier application • Extensible architecture for different flavors of hardware and network elements 22 © 2016 Accelerite. All Rights Reserved. What all do we need ? PXE server • Basic Networking • Flat network • External PXE/DHCP/DNS Servers • Advanced Networking • Isolation using VLANs • Additionally VR acts as PXE • All VM networking services mapped to Baremetal 24 © 2016 Accelerite. All Rights Reserved. Advanced Networking Network isolation with VLAN Hypervisor 2 Hypervisor 1 Hypervisor 2 Guest1 Guest1 Guest1 Guest1 VR VM1 VM2 VM3 Public VLAN99 VLAN2000 Switch Management Storage Public VLAN99 Console, Storage VMs © 2016 Accelerite. All Rights Reserved. Hypervisor 2 Advanced Networking Network isolation with VLAN Hypervisor 2 Hypervisor 1 Hypervisor 2 Guest1 Guest2 Guest1 Guest2 Guest2 Guest1 Guest1 Guest2 VR VR VM1 VM1 VM2 VM2 VM3 VM3 Public VLAN99 VLAN2000 VLAN2001 Switch Management Storage Public VLAN99 Console, Storage VMs © 2016 Accelerite. All Rights Reserved. Hypervisor 2 Advanced Networking VLAN based isolation Networking services: • DHCP • DNS • Source NAT • Static NAT • Port Forwarding • Load Balancing • Firewall • VPN • PXE © 2016 Accelerite. All Rights Reserved. Templates • PXE Server hosts the Installer Kernels and PXE Kickstart, NFS Kickstart files Server/VR Installer Kernels T F • Template Registration is T done by giving the URL P from where OS Packages Baremetal OS Package can be retrieved HTTP Host Repository 28 © 2016 Accelerite. All Rights Reserved. Example Template URL • ks=http_link_to_kickstartfile;kernel=nfs_path_to_pxe_kernel;initrd=nfs_pa th_to_pxe_initrd • ks=http://10.147.28.7/export/home/hari/rhel65/ks1.ks;kernel=10.147.2 8.7:/export/home/hari/rhel65/RHEL65/images/pxeboot/vmlinuz;initrd=1 0.147.28.7:/export/home/hari/rhel65/RHEL65/images/pxeboot/initrd.im g 29 © 2016 Accelerite. All Rights Reserved. Service Offering • Service offering allows administrator to determine size of the VM in terms of CPU and memory • e.g. small vm 2 cpu, 512 MB ram • For baremetal, CPU and memory information is specified while adding the host along with a tag and IPMI details • e.g. host-44, 8 cpu, 32 GB ram –tag ssd_bm, ipmi info : ip, user, password • Tagging scheme is used to map service offerings and baremetal host • e.g. SSD baremetal medium – 8 cpu, 32 GB ram – tag ssd_bm 30 © 2016 Accelerite. All Rights Reserved. Switch Configuration • Switch credentials • Switch topology information • host – switch port mapping Switch { "racks": [ Baremetal { Nodes "l2Switch": { "ip": "10.147.28.11", "username": "admin", "password": "password", "type": "Force10” }, "hosts": MS Node [ VR on Hypervisor { "mac": "d4:ae:52:7a:a5:9c", "port": "tengigabitethernet:0/30" }, { "mac": "d4:ae:52:7a:a5:9d", "port": "tengigabitethernet:0/31" Template } ] Storage } ] 31} © 2016 Accelerite. All Rights Reserved. Provisioning Flow Baremetal Provisioning Flow – Advanced Zone 1. User Requests Baremetal Instance Physical VR (w/ PXE & 2. Management Server configures PXE Switch DHCP on VR and physical switch with DHCP/DNS) VLANS 3. Baremetal host started with IPMI command to PXE boot 4. Host gets DHCP Management 5. Host pulls PXE image ManagementServer Management BaremetalServer 6. Image is pulled and installed Server Hosts 7. Host pulls user data from VR 33 © 2016 Accelerite. All Rights Reserved. Windows Baremetal Provisioning (Approach) • Windows Deployment server needs to be programmed • Need to invoke powershell commands from Linux • Agent to be present in WDS Server • Using RPC/or Event Queue mechanism, send Powershell commands to WDS • AD / DHCP is used for authentication and PXE booting 34 © 2016 Accelerite. All Rights Reserved. Windows Baremetal Provisioning Flow 1. User Requests Baremetal Instance DHCP/DNS Switch Active 2. Management Server configures WDS Directory WDS (w/ PXE) Server with PXE, switch configured 3. Baremetal host started with IPMI command to PXE boot 4. Host gets DHCP Management 5. Host pulls PXE image ManagementServer Management BaremetalServer 6. Image is pulled and installed Server Hosts 7. Host pulls user data from VR 35 © 2016 Accelerite. All Rights Reserved. Baremetal Use Case for Container Deployment 1. Install CoreOS on Baremetal hosts 2. Setup networking – Basic or Isolated (program the switch) 3. Create k8s cluster using cloud- config sent via userdata service on VR Node1 CoreOS 4. Use Kubernetes dashboard within Images CloudStack UI to deploy and manage containers Kubernetes Node2 5. All CloudStack tenanting and Cluster isolation features apply cloud-config Master Mgmt Virtual Dashboard Server Router 36 © 2016 Accelerite. All Rights Reserved. Extensibility Network Switch Plugin <<interface>> BaremetalSwitchBackend • getSwitchBackendType() • prepareVlan() • removePortFromVlan() Force10BaremetalSwitchBackend public static final String TYPE = "Force10" • getSwitchBackendType() { return TYPE } • prepareVlan() • removePortFromVlan() 38 © 2016 Accelerite. All Rights Reserved. Network Switch Plugin • Currently, a Dell Force10 switch is implemented • To Extend to other types, implement following two methods as a plugin Arista plugin Implementation Details: 1.Implement AristaBaremetalSwitchBackend.java and implement the interface BaremetalSwitchBackend. For example interface refer to https://github.com/apache/CloudStack/blob/master/plugins/hypervisors/baremetal/src/com/cloud/baremetal/networ kservice/Force10BaremetalSwitchBackend.java. 2. Need to implement the following three functions with appropriate actions for Arista switches in the above file. (definition of interface is in https://github.com/apache/CloudStack/blob/master/plugins/hypervisors/baremetal/src/com/cloud/baremetal/networ kservice/BaremetalSwitchBackend.java): public interface BaremetalSwitchBackend { String getSwitchBackendType(); void prepareVlan(BaremetalVlanStruct struct); void removePortFromVlan(BaremetalVlanStruct struct); } 3. Add AristaBaremetalSwitchBackend bean configuration into https://github.com/apache/CloudStack/blob/master/plugins/hypervisors/baremetal/resources/META- INF/CloudStack/core/spring-baremetal-core-context.xml. 39 © 2016 Accelerite. All Rights Reserved. Looking forward Baremetal Projects : Contributions welcome • WDS integration (WIP) • Baremetal in VPC Network • Add switch integration from different vendors • NIC Bonding • RAID • UEFI Boot • Redfish integration • Auto-discovery of hosts and NIC details 41 © 2016 Accelerite. All Rights Reserved. Q&A.
Recommended publications
  • Flexible and Integrated Resource Management for Iaas Cloud Environments Based on Programmability

    Flexible and Integrated Resource Management for Iaas Cloud Environments Based on Programmability

    UNIVERSIDADE FEDERAL DO RIO GRANDE DO SUL INSTITUTO DE INFORMÁTICA PROGRAMA DE PÓS-GRADUAÇÃO EM COMPUTAÇÃO JULIANO ARAUJO WICKBOLDT Flexible and Integrated Resource Management for IaaS Cloud Environments based on Programmability Thesis presented in partial fulfillment of the requirements for the degree of Doctor of Computer Science Advisor: Prof. Dr. Lisandro Z. Granville Porto Alegre December 2015 CIP — CATALOGING-IN-PUBLICATION Wickboldt, Juliano Araujo Flexible and Integrated Resource Management for IaaS Cloud Environments based on Programmability / Juliano Araujo Wick- boldt. – Porto Alegre: PPGC da UFRGS, 2015. 125 f.: il. Thesis (Ph.D.) – Universidade Federal do Rio Grande do Sul. Programa de Pós-Graduação em Computação, Porto Alegre, BR– RS, 2015. Advisor: Lisandro Z. Granville. 1. Cloud Computing. 2. Cloud Networking. 3. Resource Man- agement. I. Granville, Lisandro Z.. II. Título. UNIVERSIDADE FEDERAL DO RIO GRANDE DO SUL Reitor: Prof. Carlos Alexandre Netto Vice-Reitor: Prof. Rui Vicente Oppermann Pró-Reitor de Pós-Graduação: Prof. Vladimir Pinheiro do Nascimento Diretor do Instituto de Informática: Prof. Luis da Cunha Lamb Coordenador do PPGC: Prof. Luigi Carro Bibliotecária-chefe do Instituto de Informática: Beatriz Regina Bastos Haro “Life is like riding a bicycle. To keep your balance you must keep moving.” —ALBERT EINSTEIN ACKNOWLEDGMENTS First of all, I would like to thank my parents and brother for the unconditional support and example of determination and perseverance they have always been for me. I am aware that time has been short and joyful moments sporadic, but if today I am taking one more step ahead this is due to the fact that you always believed in my potential and encourage me to move on.
  • Deliverable No. 5.3 Techniques to Build the Cloud Infrastructure Available to the Community

    Deliverable No. 5.3 Techniques to Build the Cloud Infrastructure Available to the Community

    Deliverable No. 5.3 Techniques to build the cloud infrastructure available to the community Grant Agreement No.: 600841 Deliverable No.: D5.3 Deliverable Name: Techniques to build the cloud infrastructure available to the community Contractual Submission Date: 31/03/2015 Actual Submission Date: 31/03/2015 Dissemination Level PU Public X PP Restricted to other programme participants (including the Commission Services) RE Restricted to a group specified by the consortium (including the Commission Services) CO Confidential, only for members of the consortium (including the Commission Services) Grant Agreement no. 600841 D5.3 – Techniques to build the cloud infrastructure available to the community COVER AND CONTROL PAGE OF DOCUMENT Project Acronym: CHIC Project Full Name: Computational Horizons In Cancer (CHIC): Developing Meta- and Hyper-Multiscale Models and Repositories for In Silico Oncology Deliverable No.: D5.3 Document name: Techniques to build the cloud infrastructure available to the community Nature (R, P, D, O)1 R Dissemination Level (PU, PP, PU RE, CO)2 Version: 1.0 Actual Submission Date: 31/03/2015 Editor: Manolis Tsiknakis Institution: FORTH E-Mail: [email protected] ABSTRACT: This deliverable reports on the technologies, techniques and configuration needed to install, configure, maintain and run a private cloud infrastructure for productive usage. KEYWORD LIST: Cloud infrastructure, OpenStack, Eucalyptus, CloudStack, VMware vSphere, virtualization, computation, storage, security, architecture. The research leading to these results has received funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement no 600841. The author is solely responsible for its content, it does not represent the opinion of the European Community and the Community is not responsible for any use that might be made of data appearing therein.
  • Tracking Known Security Vulnerabilities in Third-Party Components

    Tracking Known Security Vulnerabilities in Third-Party Components

    Tracking known security vulnerabilities in third-party components Master’s Thesis Mircea Cadariu Tracking known security vulnerabilities in third-party components THESIS submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE in COMPUTER SCIENCE by Mircea Cadariu born in Brasov, Romania Software Engineering Research Group Software Improvement Group Department of Software Technology Rembrandt Tower, 15th floor Faculty EEMCS, Delft University of Technology Amstelplein 1 - 1096HA Delft, the Netherlands Amsterdam, the Netherlands www.ewi.tudelft.nl www.sig.eu c 2014 Mircea Cadariu. All rights reserved. Tracking known security vulnerabilities in third-party components Author: Mircea Cadariu Student id: 4252373 Email: [email protected] Abstract Known security vulnerabilities are introduced in software systems as a result of de- pending on third-party components. These documented software weaknesses are hiding in plain sight and represent the lowest hanging fruit for attackers. Despite the risk they introduce for software systems, it has been shown that developers consistently download vulnerable components from public repositories. We show that these downloads indeed find their way in many industrial and open-source software systems. In order to improve the status quo, we introduce the Vulnerability Alert Service, a tool-based process to track known vulnerabilities in software projects throughout the development process. Its usefulness has been empirically validated in the context of the external software product quality monitoring service offered by the Software Improvement Group, a software consultancy company based in Amsterdam, the Netherlands. Thesis Committee: Chair: Prof. Dr. A. van Deursen, Faculty EEMCS, TU Delft University supervisor: Prof. Dr. A.
  • Architecting for the Cloud: Lessons Learned from 100 Cloudstack Deployments

    Architecting for the Cloud: Lessons Learned from 100 Cloudstack Deployments

    Architecting for the cloud: lessons learned from 100 CloudStack deployments Sheng Liang CTO, Cloud Platforms, Citrix CloudStack History 2008 2009 2010 2011 2012 Sept 2008: Nov 2009: May 2010: July 2011: April 2012: VMOps CloudStack Cloud.com Citrix Apache Founded 1.0 GA Launch & Acquires CloudStack CloudStack Cloud.com 2.0 GA The inventor of IaaS cloud – Amazon EC2 Amazon eCommerce Platform EC2 API Amazon Proprietary Orchestration Software Open Source Xen Hypervisor Commodity Networking Storage Servers CloudStack is inspired by Amazon EC2 Amazon CloudPortaleCommerce Platform CloudEC2 APIAPIs Amazon ProprietaryCloudStack Orchestration Software ESX Hyper-VOpen SourceXenServer Xen Hypervisor KVM OVM Commodity Networking Storage Servers There will be 1000s of clouds SP Data center mgmt Desktop Owner | Operator Owner and automation Cloud IT Horizontal Vertical General Purpose Special Purpose Learning from 100s of CloudStack deployments Service Providers Web 2.0 Enterprise What is the biggest difference between traditional-style data center automation and Amazon-style cloud? How to handle failures • Server failure comes from: ᵒ 70% - hard disk ᵒ 6% - RAID controller ᵒ 5% - memory ᵒ 18% - other factors 8% • Application can still fail for Annual Failure Rate of servers other reasons: ᵒ Network failure ᵒ Software bugs Kashi Venkatesh Vishwanath and ᵒ Human admin error Nachiappan Nagappan, Characterizing Cloud Computing Hardware Reliability, SoCC’10 11 Internet Core Routers … Access Routers Aggregation Switches Load Balancers … Top of Rack Switches Servers •Bugs in failover mechanism •Incorrect configuration 40 % •Protocol issues such Effectiveness of network as TCP back-off, redundancy in reducing failures timeouts, and spanning tree reconfiguration Phillipa Gill, Navendu Jain & Nachiappan Nagappan, Understanding Network Failures in Data Centers: Measurement, Analysis and Implications , SIGCOMM 2011 13 A.
  • Inequalities in Open Source Software Development: Analysis of Contributor’S Commits in Apache Software Foundation Projects

    Inequalities in Open Source Software Development: Analysis of Contributor’S Commits in Apache Software Foundation Projects

    RESEARCH ARTICLE Inequalities in Open Source Software Development: Analysis of Contributor’s Commits in Apache Software Foundation Projects Tadeusz Chełkowski1☯, Peter Gloor2☯*, Dariusz Jemielniak3☯ 1 Kozminski University, Warsaw, Poland, 2 Massachusetts Institute of Technology, Center for Cognitive Intelligence, Cambridge, Massachusetts, United States of America, 3 Kozminski University, New Research on Digital Societies (NeRDS) group, Warsaw, Poland ☯ These authors contributed equally to this work. * [email protected] a11111 Abstract While researchers are becoming increasingly interested in studying OSS phenomenon, there is still a small number of studies analyzing larger samples of projects investigating the structure of activities among OSS developers. The significant amount of information that OPEN ACCESS has been gathered in the publicly available open-source software repositories and mailing- list archives offers an opportunity to analyze projects structures and participant involve- Citation: Chełkowski T, Gloor P, Jemielniak D (2016) Inequalities in Open Source Software Development: ment. In this article, using on commits data from 263 Apache projects repositories (nearly Analysis of Contributor’s Commits in Apache all), we show that although OSS development is often described as collaborative, but it in Software Foundation Projects. PLoS ONE 11(4): fact predominantly relies on radically solitary input and individual, non-collaborative contri- e0152976. doi:10.1371/journal.pone.0152976 butions. We also show, in the first published study of this magnitude, that the engagement Editor: Christophe Antoniewski, CNRS UMR7622 & of contributors is based on a power-law distribution. University Paris 6 Pierre-et-Marie-Curie, FRANCE Received: December 15, 2015 Accepted: March 22, 2016 Published: April 20, 2016 Copyright: © 2016 Chełkowski et al.
  • Cristina Opriceana, Hajime Tazaki (IIJ Research Lab.) Linux Netdev 2.2, Seoul, Korea 08 Nov

    Cristina Opriceana, Hajime Tazaki (IIJ Research Lab.) Linux Netdev 2.2, Seoul, Korea 08 Nov

    Network stack personality in Android phone Cristina Opriceana, Hajime Tazaki (IIJ Research Lab.) Linux netdev 2.2, Seoul, Korea 08 Nov. 2017 1 Librarified Linux taLks (LLL) Userspace network stack (NUSE) in general (netdev0.1) kernel CI with libos and ns-3 (netdev1.1) Network performance improvement of LKL (netdev1.2, by Jerry Chu) How bad/good with LKL and hrtimer (BBR) (netdev2.1) Updating Android network stack (netdev2.2) 2 Android a platform of billions devices billions installed Linux kernel Questions When our upstreamed code available ? What if I come up with a great protocol ? https://developer.android.com/about/dashboards/index.html 3 Android (cont'd) When our upstreamed code available ? wait until base kernel is upgraded backport specific function What if I come up with a great protocol ? craft your own kernel and put into your image Long delivery to all billions devices 4 Approaches to alleviate the issue Virtualization (KVM on Android) Overhead isn't negligible to embedded devices Project Treble (since Android O) More modular platform implementation Fushia Rewrite OS from scratch QUIC (transport over UDP) Rewrite transport protocols on UDP https://source android com/devices/architecture/treble https://source.android.com/devices/architecture/treble An alternate approach network stack personality use own network stack implemented in userspace no need to replace host kernels but (try to) preserve the application compatibility NUSE (network stack in userspace) No delay of network stack update Application can choose a network stack if needed 56 Userspace implementations Toys, Misguided People Selfish Motivation Trying to present that a Toy is practically useful 7 Linux Kernel Library intro (again) Out-of-tree architecture (h/w-independent) Run Linux code on various ways with a reusable library h/w dependent layer on Linux/Windows /FreeBSD uspace, unikernel, on UEFI, network simulator (ns-3) Android 8 LKL: current status Sent RFC (Nov.
  • 60 Recipes for Apache Cloudstack

    60 Recipes for Apache Cloudstack

    60 Recipes for Apache CloudStack Sébastien Goasguen 60 Recipes for Apache CloudStack by Sébastien Goasguen Copyright © 2014 Sébastien Goasguen. All rights reserved. Printed in the United States of America. Published by O’Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472. O’Reilly books may be purchased for educational, business, or sales promotional use. Online editions are also available for most titles (http://safaribooksonline.com). For more information, contact our corporate/ institutional sales department: 800-998-9938 or [email protected]. Editor: Brian Anderson Indexer: Ellen Troutman Zaig Production Editor: Matthew Hacker Cover Designer: Karen Montgomery Copyeditor: Jasmine Kwityn Interior Designer: David Futato Proofreader: Linley Dolby Illustrator: Rebecca Demarest September 2014: First Edition Revision History for the First Edition: 2014-08-22: First release See http://oreilly.com/catalog/errata.csp?isbn=9781491910139 for release details. Nutshell Handbook, the Nutshell Handbook logo, and the O’Reilly logo are registered trademarks of O’Reilly Media, Inc. 60 Recipes for Apache CloudStack, the image of a Virginia Northern flying squirrel, and related trade dress are trademarks of O’Reilly Media, Inc. Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks. Where those designations appear in this book, and O’Reilly Media, Inc. was aware of a trademark claim, the designations have been printed in caps or initial caps. While every precaution has been taken in the preparation of this book, the publisher and authors assume no responsibility for errors or omissions, or for damages resulting from the use of the information contained herein.
  • Installing Management Node Remotely

    Installing Management Node Remotely

    Installing Management Node Remotely This chapter contains the following topics: • Overview to Installation of Management Node Remotely, on page 1 • Overview to Cisco VIM Baremetal Manager REST API, on page 5 • Installing Cisco VIM Baremetal Manager Management Node On a UCS C-series Server, on page 6 • Preparing the Cisco VIM Baremetal Manager Management Node from Cisco VIM Software Hub Server, on page 7 Overview to Installation of Management Node Remotely Cisco VIM fully automates the installation operation of the cloud. In releases prior to Cisco VIM 3.4.1, the management node installation was always manual, as the bootstrap of the cloud happens from there. Using this feature, the management node, referred to as Cisco VIM Baremetal Manager is automatically installed over a layer 3 network to accelerate the Cisco VIM installation process. Note In this chapter, the term Cisco VIM Baremetal Manager and Remote Install of Management Node (RIMN) are used interchangeably. Remote Install of Management Node Remote Install of Management Node (RIMN) software is deployed on the RIMN deployment node from where one or more management nodes are installed. Cisco VIM Baremetal Manager or RIMN supports remote installation of servers across WAN or LAN with either IPv4 or IPv6 connectivity. Cisco VIM Baremetal Manager can be installed on the Cisco VIM Baremetal Manager deployment node by using air-gapped installation. After you install the RIMN software on its management node, you must define an input file for bare-metal config (in YAML format) and use Cisco VIM Baremetal Manager CLI or Rest API to deploy the user-specified ISO into the target platform (as depicted in the figure below): Installing Management Node Remotely 1 Installing Management Node Remotely Hardware Requirements for RIMN RIMN solution is built based on the interaction of several components as depicted below: • Rest-API and CLI: Pushes the received input data into Etcd datastore.
  • View the Slides

    View the Slides

    RedLeaf: Isolation and Communication in a Safe Operating System Vikram Narayanan1, Tianjiao Huang1, David Detweiler1, Dan Appel1, Zhaofeng Li1, Gerd Zellweger2, Anton Burtsev1 OSDI ’20 1University of California, Irvine 2VMware Research History of Isolation Cedar Ka�eOS Multics Pilot Scomp SPIN J-Kernel Mondrian VINO Singularity 1973 1980 1983 1995 1996 1999 2002 2005 Year • Isolation of kernel subsystems • Final report of Multics (1976) • Scomp (1983) • Systems remained monolithic • Isolation was expensive 1 Isolation mechanisms • Hardware Isolation • Segmentation (46 cycles)1 • Page table isolation (797 cycles)2 • VMFUNC (396 cycles)3 • Memory protection keys (20-26 cycles)4 • Language based isolation • Compare drivers written (DPDK-style) in a safe high-level language (C, Rust, Go, C#, etc.)5 • Managed runtime and Garbage collection (20-50% overhead on a device-driver workload) 1L4 Microkernel: Jochen Liedtke 2https://sel4.systems/About/Performance/ 3Lightweight Kernel Isolation with Virtualization and VM Functions, VEE 2020 4Hodor: Intra-process isolation for high-throughput data plane libraries 5The Case for Writing Network Drivers in High-Level Programming Languages, ANCS 2019 2 • Linear types • Enforces type and memory safety • Statically checked at compile time • Safety without runtime garbage collection overhead Rust Traditional Safe languages vs Rust Java, C# etc. A 3 • Linear types • Enforces type and memory safety • Statically checked at compile time • Safety without runtime garbage collection overhead Rust Traditional Safe languages vs Rust Java, C# etc. A Vector 3 • Linear types • Enforces type and memory safety • Statically checked at compile time • Safety without runtime garbage collection overhead Rust Traditional Safe languages vs Rust Java, C# etc.
  • A Generic Development and Deployment Framework for Cloud Computing and Distributed Applications

    A Generic Development and Deployment Framework for Cloud Computing and Distributed Applications

    Computing and Informatics, Vol. 32, 2013, 461{485 A GENERIC DEVELOPMENT AND DEPLOYMENT FRAMEWORK FOR CLOUD COMPUTING AND DISTRIBUTED APPLICATIONS Binh Minh Nguyen, Viet Tran, Ladislav Hluchy´ Institute of Informatics Slovak Academy of Sciences D´ubravsk´acesta 9 845 07 Bratislava, Slovakia e-mail: fminh.ui, viet.ui, [email protected] Communicated by Jacek Kitowski Abstract. Cloud computing has paved the way for advance of IT-based on demand services. This technology helps decrease capital expenditure and operation costs, solve scalability issue and many more user and provider constraints. However, devel- opment and deployment of distributed applications on cloud environment becomes a more and more complex task. Cloud users must spend a lot of time to prepare, in- stall and configure their applications on clouds. In addition, after development and deployment, the applications almost cannot move from one cloud to another due to the lack of interoperability between them. To address these problems, in this paper we present a novel development and deployment framework for cloud distributed applications/services. Our approach is based on abstraction and object-oriented programming technique, allowing users to easily and rapidly develop and deploy their services into cloud environment. The approach also enables service migration and interoperability among the clouds. Keywords: Cloud computing, distributed application, abstraction, object-oriented programming, interoperability Mathematics Subject Classification 2010: 68-M14 462 B. M. Nguyen, V. Tran, L. Hluch´y 1 INTRODUCTION Cloud computing is described as a business model for on-demand delivery of com- putation power, in which consumers pay providers what they used (\pay-as-you- go").
  • Guide to the Open Cloud Open Cloud Projects Profiled

    Guide to the Open Cloud Open Cloud Projects Profiled

    Guide to the Open Cloud Open cloud projects profiled A Linux Foundation publication January 2015 www.linuxfoundation.org Introduction The open source cloud computing landscape has changed significantly since we published our first cloud guide in October 2013. This revised version adds new projects See also the rise of Linux container and technology categories that have since technology with the advent of Docker gained importance, and in some cases and its emerging ecosystem. You will be radically change how companies approach hard pressed to find an enterprise Linux building and deploying an open source distribution that isn’t yet working on Docker cloud architecture. integration and touting its new container strategy. Even VMware vSphere, Google In 2013, many cloud projects were still Cloud Platform, and Microsoft Azure are working out their core enterprise features rushing to adapt their cloud platforms to the and furiously building in functionality. And open source Docker project. enterprises were still very much in the early stages of planning and testing their public, This rapid pace of innovation and resulting private or hybrid clouds–and largely at the disruption of existing platforms and vendors orchestration layer. can now serve as a solid case study for the role of open source software and Now, not only have cloud projects collaboration in advancing the cloud. consistently (and sometimes dramatically) grown their user and developer Other components of the cloud infrastructure communities, lines of code and commits have also followed suit, hoping to harness over the past year, their software is the power of collaboration. The Linux increasingly enterprise-ready.
  • Enterprise Cloud Analytics

    Enterprise Cloud Analytics

    IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 3, Ver. IV (May – Jun. 2015), PP 12-16 www.iosrjournals.org Enterprise Cloud Analytics K.Vasuki# S.M.Srihari Shankar* #PG Scholar, Department of CSE, Sri Shakthi Institute of Engineering and Technology *Assistant Professor, Department of CSE,Sri Shakthi Institute of Engineering and Technology. Abstract: Cloud computing revolutionize IT and business by offering computing as a utility over the internet. The evolution from internet to a cloud computing platform, the emerging development paradigm and technology and how these will change the way enterprise applications should be architected for cloud deployment plays an important role but these enterprise technologies are critical to cloud computing. New cloud analytics and business intelligence (BI) services can help businesses (organizations) better manage big data and cloud applications.Analysing and gathering business intelligence (BI) has never been easy, but today BI is complicated further by overwhelming amounts of data loads and the number of data entry and access points. New cloud analytics advancements may offer BI relief and even profit-increasing predictability for enterprises. These new cloud analytics applications can deliver functional capabilities that can be easily, quickly and economically deployed, producing tangible and measurable benefits far more rapidly than in the past. Many organizations that recognized, effectively analysing their business needs and providing the data they require to make the right business decisions depends on a combination of internally generated data and externally available data. Keywords:Enterprise clouds, Business Intelligence, Analytics, business models. I. Introduction The term ―Cloud computing‖ is internet-based computing in which large groups of remote servers are networked to allow the centralized data storage, and online access to computer services or resources.