2 Table of Contents I 1 What Is It? Related Technologies Grid Computing Virtualization Utility Computing Autonomic Computing Is It New? Definition 2 Business Business Model Elasticity Commercial Solutions Amazon EC2 Google AppEngine Microsoft Windows Azure
3 Table of Contents II 3 Architecture Service Categories 4 Research Challenges Automated Service Provisioning Automated Service Provisioning Energy Management Server Consolidation Virtual Machine Migration Traffic analysis Data Security Software Frameworks Novel Cloud Architecture 5 Bibliography
4 What Is? Some technologies are often compared to and, in fact, they share some characteristics.
5 Related Technologies Grid Computing Both make distributed resources available to an application. A Cloud uses virtualization to share resources, provision resources dynamically.
6 Related Technologies Virtualization Abstracts details of hardware Provides virtualized resources Foundation of Virtualized resources are (re)assigned to applications on-demand
7 Related Technologies Utility Computing Charges customers based on usage, like electricity. Cloud providers can maximize resource utilization minimize operating costs using utility-based pricing on-demand resource provisioning
8 Related Technologies Autonomic Computing IBM, 2001 Self-managing systems Internal and external observations No human intervention Aims to reduce complexity Automatic resource provisioning Server consolidation Aims to lower cost
9 Is It New? Is It New? is not a new technology, but a new operations model that combines existing technologies.
10 Definition Definition Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. [NIST, 2009]
11 Business Illusion of infinite resources on demand No up-front commitment Ability to pay on a short-term basis Lower risks (hardware failure) maintainance costs hardware expenses (fabless companies)
12 Business Model Business Model Figure: business model
14 Commercial Solutions Amazon EC2 Virtual Machines on top of Xen Full control of software stack Images can be used to launch other VMs Conditions can trigger VM addition or removal Multiple locations: US, Europe, Asia
15 Commercial Solutions Google AppEngine Platform for traditional web applications Supports python and java Non-relational database Scaling is automatic and transparent
16 Commercial Solutions Microsoft Windows Azure Three components 1 Windows based environment (applications, data) 2 SQL Azure based on SQL Server 3.NET Services for distributed infraestructure Platform runs applications in the cloud or locally Supports.NET, C#, VB, C++, etc User must specify application needs to scale
17 Service Categories Service Categories Infraestructure as a Service (IaaS) On-demand resources, usually VMs Amazon EC2 Platform as a Service (PaaS) Operating system support Software development frameworks Google App Engine Software as a Service (SaaS) On-demand applications over the Internet Salesforce (CRM)
18 Service Categories Layers Figure: Architecture of
19 Automated Service Provisioning Automated Service Provisioning Rejecting users Resources Capacity Demand Idle resources Time (days) Figure: Fixed capacity problems. Most of the time opportunities are lost (red) or resources are wasted (yellow).
20 Automated Service Provisioning Automated Service Provisioning Resource (de-)allocation to rapid demand flunctuations. 1 Predict number of instances to handle demand Queuing theory Control theory Statistical Machine Learning 2 Predict future demand 3 Automatic resource allocation
21 Energy Management Energy Management Problems: Government regulations, environmental standards 53% of the cost is powering and cooling Researches: Energy-efficient hardware Energy-aware job-scheduling Server consolidation Energy-efficient network protocols and infraestructures
22 Energy Management Figure: HP Performance-Optimized Dataceter (POD): 40 feet and a capacity of 3,520 nodes or 12,000 LFF hard drives.
23 Energy Management Figure: HP POD inside.
24 Energy Management Figure: On the left, Microsoft s new $500 million Chicago data center. On the right, NASA cloud computing application hosted in a container and below it, IBM Portable Modular Data Center (PMDC).
25 Server Consolidation Server Consolidation Maximize servers in energy-saving state VM dependencies (communication requirements) Resource congestions Variant of vector bin-packing problem (NP-hard)
26 Virtual Machine Migration Virtual Machine Migration Xen and VMWare live VM migration (<1 sec) Detecting hotspots versus sudden workload changes
27 Traffic analysis Traffic analysis Important for management and planning decisions Much higher density of links Most existing methods have problems here: Compute only a few hundreds end hosts Assume flow patterns (MapReduce jobs)
28 Data Security Data Security Confidentiality for secure data access and transfer Cryptographic protocols Auditability Remote attestation (system state encrypted with TPM) With VM migration, it is not sufficient Virtualization platform must be trusted (SVMM) Hardware must be trusted using hardware TPM Efficient protocols are being designed
29 Software Frameworks Software Frameworks Large-scale data-intensive applications usually leverage MapReduce frameworks. Scalable Fault-tolerant Challenges: Performance and resource usage depends on application Better performance and cost can be achieved by: Selecting configuration parameter values Mitigating the bottleneck resources Adaptive scheduling in dynamic conditions Performance modeling of Hadoop jobs Energy-aware: A node should sleep while waiting new jobs HDFS must be energy-aware, also
30 Novel Cloud Architecture Novel Cloud Architectures Small data centers can be more advantageous Less power (cooling) Cheaper Better geographically distributed (interactive gaming) Using voluntary resources for cloud applications Much cheaper More suitable for non-profit applications Must deal with heterogeneous resources Machines can be turned on or off at any time How to incentivate resource donations?
31 Bibliography Zhang, Q et al (2010) Cloud computing: state-of-the-art and research challenges Journal of Internet Services and Applications 1(1):7-18 Armbrust M et al (2009) Above the Clouds: A Berkeley View of UC Berkeley Technical Report Mell, Peter and Grance, Tim (2009) The NIST Definition of (v15) Data Center Knowledge category/technology/containers/
Special Publication 800-146 DRAFT Cloud Computing Synopsis and Recommendations Recommendations of the National Institute of Standards and Technology Lee Badger Tim Grance Robert Patt-Corner Jeff Voas NIST
En vue de l'obtention du DOCTORAT DE L'UNIVERSITÉ DE TOULOUSE Délivré par : Institut National Polytechnique de Toulouse (INP Toulouse) Discipline ou spécialité : Réseaux, Télécommunications, Systèmes et
Institute of Parallel and Distributed Systems University of Stuttgart Universitätsstraße 38 D 70569 Stuttgart Diplomarbeit Nr. 3242 Data security in multi-tenant environments in the cloud Tim Waizenegger
Paper Towards an Agent-Based Augmented Cloud Roman Dębski, Aleksander Byrski, and Marek Kisiel-Dorohinicki Department of Computer Science, AGH University of Science and Technology, Kraków, Poland Abstract
Plug Into The Cloud with Oracle Database 12c ORACLE WHITE PAPER DECEMBER 2014 Disclaimer The following is intended to outline our general product direction. It is intended for information purposes only,
How AWS Pricing Works May 2015 (Please consult http://aws.amazon.com/whitepapers/ for the latest version of this paper) Page 1 of 15 Table of Contents Table of Contents... 2 Abstract... 3 Introduction...
A new Breed of Managed Hosting for the Cloud Computing Age A Neovise Vendor White Paper, Prepared for SoftLayer Executive Summary Traditional managed hosting providers often suffer from issues that cause
Microsoft Private Cloud Fast Track Microsoft Private Cloud Fast Track is a reference architecture designed to help build private clouds by combining Microsoft software with Nutanix technology to decrease
No One (Cluster) Size Fits All: Automatic Cluster Sizing for Data-intensive Analytics Herodotos Herodotou Duke University firstname.lastname@example.org Fei Dong Duke University email@example.com Shivnath Babu Duke
Microsoft System Center 2012 R2 Why Microsoft? For Virtualizing & Managing SharePoint July 2014 v1.0 2014 Microsoft Corporation. All rights reserved. This document is provided as-is. Information and views
Special Publication 800-125 Guide to Security for Full Virtualization Technologies Recommendations of the National Institute of Standards and Technology Karen Scarfone Murugiah Souppaya Paul Hoffman NIST
Just in Time Clouds: Enabling Highly-Elastic Public Clouds over Low Scale Amortized Resources Rostand Costa 1,2, Francisco Brasileiro 1 1 Federal University of Campina Grande Systems and Computing Department
Special Publication 800-145 The NIST Definition of Cloud Computing Recommendations of the National Institute of Standards and Technology Peter Mell Timothy Grance NIST Special Publication 800-145 The NIST
Securing Traditional and Cloud-Based Datacenters With Next-generation Firewalls February 2015 Table of Contents Executive Summary 3 Changing datacenter characteristics 4 Cloud computing depends on virtualization
Software-Defined Networking: The New Norm for Networks ONF White Paper April 13, 2012 Table of Contents 2 Executive Summary 3 The Need for a New Network Architecture 4 Limitations of Current Networking
UNIVERSITAT POLITÈCNICA DE CATALUNYA AUGURES, Profit-Aware Web Infrastructure Management by Nicolas Poggi M. Advisor: David Carrera A thesis submitted in partial fulfillment for the degree of Doctor of
Cloud Service Level Agreement Standardisation Guidelines Brussels 24/06/2014 1 Table of Contents Preamble... 4 1. Principles for the development of Service Level Agreement Standards for Cloud Computing...
THÈSE En vue de l obtention du DOCTORAT DE L UNIVERSITÉ DE TOULOUSE Délivré par : l Université Toulouse 3 Paul Sabatier (UT3 Paul Sabatier) Présentée et soutenue le 03/07/2014 par : Cheikhou THIAM Anti
ascent Thought leadership from Atos white paper Data Analytics as a Service: unleashing the power of Cloud and Big Data Your business technologists. Powering progress Big Data and Cloud, two of the trends
WHITEPAPER CLOUD Possible Use of Cloud Technologies in Public Administration Version 1.0.0 2012 Euritas THE BEST WAY TO PREDICT THE FUTURE IS TO CREATE IT. [Willy Brandt] 2 PUBLISHER'S IMPRINT Publisher:
WHITEPAPER Microsoft SQL Server Databases Thrive in the Cloud Virtualizing Data-Intensive Applications for Page 2 Overview As more and more organizations embrace cloud computing to save money, increase
An Oracle White Paper October, 2013 Delivering Database as a Service (DBaaS) using Oracle Enterprise Manager 12c Executive Overview...2 Evolution of Database as a Service...2 Managing the Database Lifecycle...4
Ernst & Young LLP One Commerce Square Suite 700 2005 Market Street Philadelphia, PA 19103 Tel: +1 215 448 5000 Fax: +1 215 448 5500 ey.com Report of Independent Auditor To the Management of Verizon Communications
Elastic Cloud Computing in the Open Cirrus Testbed implemented via Eucalyptus International Symposium on Grid Computing 2009 (Taipei) Christian Baun The cooperation of and Universität Karlsruhe (TH) Agenda
white paper Public or Private Cloud: The Choice is Yours Current Cloudy Situation Facing Businesses There is no debate that most businesses are adopting cloud services at a rapid pace. In fact, a recent
Managed Hosting: Best Practices to Support Education Strategy in the Career College Sector Online learning is playing a critical role in the delivery of Teaching and Learning and the overall experience
Optimizing with Citrix NetScaler White Paper Three keys to building the best front-end network for virtual desktop delivery www.citrix.com Executive summary Motivated by the compelling benefits virtual