Published on March 11, 2014
A Technical Overview of Cloud Computing Pethuru Raj PhD Infrastructure Architect IBM Global Cloud Center of Excellence (CoE) IBM India, Bangalore Email: firstname.lastname@example.org
A Captivating Cloud Case Problem • Make all NewYork Times archived articles from 1851 to 1922 available free of charge over the internet.That means, 11 Million articles as images to be scanned and converted to PDF format • IT Requirements – 100 servers & 5.5TB storage Cloud Solution • Use cloud infrastructures for reading and converting the article images to PDF • Upload data and ranTIFFPDF conversion routine • All 11 million articles processed in < 24 hours • Use 100 Cloud instances • The total cost is ~$240
A Cursory Look of a Cloud Data Center
Defining the Cloud in a Capsule Form On the Infrastructural perspective, a cloud is a dynamic pool of consolidated, centralised, increasingly federated, virtualised or non- virtualized, automated, and shared IT infrastructures (Compute, storage and network components) Automation is at multiple levels (resource provisioning, service management, job scheduling, load-balancing, security, governance, etc.) “Cloud – IT Infrastructures Optimized”
The Cloud Evolution Web Sites in the Web (Web 1.0) (Read Only) Search Engines in the Web (Search only) Social Sites in the Web (Web 2.0) (Read andWrite Only) Enterprise-scale Web Applications in the Web (E-commerce Applications such as Amazon.com, Flipkart.com, etc., E-auction applications such as e- bay, etc. E-mail applications (Gmail,Yahoo, etc.), Online Banking, …) COTS & Home-Grown, Bespoke Business Applications in theWeb ◦ These applications are not given freely as above. Therefore the cost model is to go for subscription and consumption-based charge. That means, any application needs to be firstly refactored and refined to be given as a multitenant, online, off-premise, secure, multi-device, ubiquitous, intuitive, elastic, & QoS-compliant service.
The Cloud Services Software as a Service - All kinds of software solutions are being provided as service from Clouds over the Internet to worldwide users. For example, business software as a service from Clouds (Salesforce.com, Ramco, NetSuite, Oracle, SAP, etc.) Platform Software as a Service (Design, Development, Deployment, Delivery, Integration, Management, Orchestration,, etc.) . For example, application development, testing, delivery, and management are happening in clouds. Infrastructure as a Service (Compute, Storage, Memory, and Networking). Clouds are the highly optimized, service-oriented, on-demand, elastic, and web-scale IT infrastructure Federation as a Service - With cloud brokers, geographically distributed and disparate clouds are being orchestrated to craft and deliver business-aware and people-centric services in time.This will lead to the Intercloud.
The Key Drivers for Cloud Computing
The IT Constraints The IT Infrastructure Utilization The Alignment between Business and IT The IT Agility, and Autonomy & Affordability The Quality Attributes (Scalability, Performance,Availability, Flexibility, Consumability, etc.) of IT Infrastructures The IT Complexity (due to the growing multiplicity and heterogeneity of technologies, programming languages, protocols, data formats, etc.)
9 For Consolidated,Virtualized,Adaptive, & Shared Infrastructures 85% idleIn distributed computing environments, up to 85% of computing capacity sits idle. Explosion of information driving 54% growth in storage shipments every year. 1.5x 70¢ per $1 70% on average is spent on maintaining current IT infrastructures versus adding new capabilities. The Concerns & Challenges of Enterprise IT
The Pictorial Representation of the IT Challenges IT Constrictions –Traditional IT Setup
The Business IT Goals More with Less – Less Wastage, Slippage and Pilferage Adaptive & Instant-On Enterprise IT On-Demand, Converged, Real-time & Dynamic IT Infrastructures Affordable Yet High-Performance Computing Green, Lean, and Elastic IT Distributed Deployment & Centralised IT Monitoring and Management Transitioning to Virtual IT
The Way Forward Transitioning to Next-Generation IT Infrastructures that are Cloud-ready Software-defined Policy-based & Orchestration-enabled Programmable, Secure & Sharable Accessible & Autonomic Federated yet Converged Distributed Deployment yet Centrally Managed Through a host of rationalization, simplification, automation, and optimization techniques
Why the Cloud Paradigm is very Popular? Cloud is a grant conglomeration and convergence of proven, mission- critical and enterprise-scale technologies Path-breaking Impacts on Business & IT IT as a Service Generic & Green Technology Breeds Innovations on the Business front IT Optimization to be Lean IT and Business Agility Shared and Service Era Quality of Service (QoS) Attributes
“Smartly leveraging a dynamic pool of commodity servers to perform and provide the varying computing needs of a multitude of distributed organizations and users as a service over the open and public Internet” Cloud is just an advanced, optimized, and programmable IT environment providing the illusion of infinite compute and storage power. Applications, platforms and infrastructures become publicly available, discoverable, interoperable, reusable, & composable network services Defining Cloud Computing
Mainframe -> Personal -> Client / Server -> Cloud -> Intercloud (Centralization to Distribution to Centralization to Federation) The Journey towards the Cloud Era
The Prominent CloudTypes Public Cloud or External Cloud* - Resources are dynamically provisioned on a fine-grained, self- service basis over the Internet Hybrid Cloud – A combination of Public and Private cloud Private Cloud, Internal Cloud, Enterprise Cloud – Emulation of the Public Cloud on a private network
The Cloud Delivery Models With the technology-enabled convergence of all kinds of cyber and physical systems is maturing and stabilizing, the vision of “anything as a service (AaaS)” (Both IT as well as physical services) is to see the light soon. Cloud is the most crucial and core Infrastructure Component in that vision.
The Cloud Ecosystem
The Implications of Cloud Computing Business Cases – Newer Deployment, Delivery, Consumption, Management, Monitoring, Subscription & Pricing Models Technical Cases – The materialization of Consolidated, Centralized, Converged, Federated,Virtualized, Automated and Shared IT Infrastructures Use Cases – Self-Servicing, Simplicity, Consumability, Ubiquity, Utility, etc. 19
© 2010 Wipro Ltd - Confidential20 The public, open, cheap Internet is the principal communication infrastructure for Cloud. All kinds of IT services (infrastructures and platforms) are being availed from Cloud centers. Business services are accordingly manipulated,deployed, delivered, orchestrated, maintained and managed in cloud platforms.The Cloud characteristics are ◦ Consolidated,centralised,federated, and converged Commoditized Servers – Increasingly common, casual and cheap servers are being utilised in Cloud centers. ◦ Optimized & Shared – Cloud resources and assets are subjected to rationalization, optimization,simplification and shared across different organizations and multiple users over the network ◦ Virtualized – Increasingly servers, storages, databases, services, applications,and networks are virtualised ◦ Automated – Most of the manual tasks in data centers and server farms are being automated to significantly reduce manual intervention,interpretation and instruction ◦ Software-defined and Programmability – Like software, all kinds of hardware and network components are being programmed to be flexible, configurable, customizable, replaceable, sustainable, etc. ◦ Workload Optimized Infrastructures – Cloud centers comprises specific appliances, clusters, bare metal servers, etc. ◦ Application-aware Clouds – Dedicated clouds are being established for specific needs such as sensor, device, knowledge, data, and mobile clouds. The Cloud Summary
© 2010 Wipro Ltd - Confidential21 The Cloud Realization Technologies
The Cloud Realization Technologies ◦ Consolidation, Centralization & Federation, Convergence &Virtualization ◦ Service oriented Architecture (SOA) & Software as a Service (SaaS) ◦ Cluster,Autonomic, On-demand, Grid and Utility Computing ◦ Ultra-high Bandwidth Ambient Communication ◦ Rationalization, Optimization, & Simplification ◦ Multi-tenancy and Sharing ◦ Automation (Resource provisioning & Management,Workload Management & Job Scheduling, Load Balancing, Integration, Self-Servicing, etc.)
Virtualization Replicating the proven “Separation of Concerns” and “Divide & Conquer” Techniques in Software Engineering for Hardware Engineering is the essence of virtualization. Virtualization facilitates programming and managing hardware rationally Virtualization is the core technique for creating a dynamic pool ofVirtual Machines (VMs) by decomposing Physical machines On the reverse side, composing thoseVMs on need basis systematically to do better and bigger things Hypervisors (VMMs) is the tool for provisioning, de-provisioning & monitoring VMs VMs are easily configurable, replaceable, scalable, migratable, etc. VMs are the new commodity servers and all the relevant intelligence gets transferred to hypervisors, the operating system of operating systems. The snapshot of eachVM is persisted in a storage
Physical Infrastructure Fibre Channel storage Fibre ChannelEthernet NFS storage iSCSI storage Network applications operating system physical host
Virtual Infrastructure hypervisor ESXi host Fibre Channel Fibre Channel storage Ethernet NFS storage iSCSI storage network virtual machines
Bare Metal (Type I)Virtualization
Hosted (Type II)Virtualization
TheVM Snapshots Encapsulation VM 1 VM 2 VM 3 Datastore (NFS)
The Spread ofVirtualization Through a layer of abstraction, virtualization lays a stimulating foundation for decimating all kinds of dependencies.That is, the tight coupling between software and hardware gets eliminated to ensure true portability. Any software runs on any hardware. A typicalVM comprises vCPU, Memory, Storage, vSwitch, etc. Virtualization penetrates into every resource ◦ ServerVirtualization ◦ StorageVirtualization ◦ NetworkVirtualization ◦ ApplicationVirtualization ◦ DataVirtualization ◦ ServiceVirtualization ◦ DesktopVirtualization
Flexible Deployment Rapid Deployment Server Consolidation Business Flexibility Energy Efficiency High Availability Management Automation Enhanced QoS The Benefits ofVirtualization
Virtual Machine Virtual Machine Hardware hardware 3D 1 IDE controller 4 devices up to 3 parallel ports up to 4 serial/com ports HD audio 1 USB controller 20 Devices 1 floppy controller 2 Devices 1-10 NICs 15 devices per adapter up to 1TB of RAM up to 4 SCSI adapters up to 64 vCPUs
Why UseVirtual Machines? Easy to relocate: ◦ Encapsulated into files ◦ Independent of physical hardware Easy to manage: ◦ Isolated from other virtual machines ◦ Insulated from hardware changes Provides the ability to support legacy applications Allows servers to be consolidated Virtual machinePhysical machine Difficult to relocate: Moves require downtime Specific to physical Hardware Difficult to manage: Requires physical maintenance Hardware failures cause downtime Hardware has limitations: Hardware changes limit application support Servers are physically individual
The Power of Service Orientation (SO) Expressing and Exposing Everything as a Service Not only software applications but also hardware and network modules are being presented as services Services are having the following distinctions publicly discoverable, network accessible, easily consumable, Highly reusable, and composable, Semantically and syntactically Interoperable, replaceable, sustainable, serviceable, etc.
IT as a Service
Cloud Integration and Orchestration
The Onset of Federated Clouds
Why Cloud Federation? Cloud service providers collaborate dynamically to share their virtual infrastructure for crafting the Intercloud. Load Balancing Prevention from Vendor Lock-ins Prevention from Power Outages & Failures Capacity Management Efficient use of Surplus Resources Scaling Data to other CSPs 38
Cloud Platforms and Infrastructures
The Key Cloud Platforms Integrated Development Environments (IDEs) Deployment & Execution Containers Application-specific Servers (E-commerce, etc.) Integration Buses, Brokers, & Fabrics Orchestration Engines Service Delivery Platforms (SDPs) Management & Governance Platforms Scalable SQL and NoSQL Databases
The Cloud Infrastructures Generic Clouds Special-Purpose Clouds Private Cloud Public Cloud Hybrid Cloud Community Cloud Cloud in a Box / Cloud Appliance Data / Information Cloud Science Cloud Knowledge Cloud Service Cloud Mobile Cloud Sensor Cloud Government Cloud Device Cloud
Cloud Reference Architecture
Envisaging the Cloud Applications Clouds will be the core, converged, and cognitive IT Infrastructure behind the vision of Smarter Planet Platforms,Applications and Services 1. Cloud-enabled Applications – Analyzed for the cloud fitment, modernized, migrated, and delivered from Clouds 2. Cloud-native Applications – designed on the three basic tenets (Resiliency, Economics, and Security) deployed, tuned, delivered and managed
The Design Considerations of Cloud Native Applications Services • All functionality is published and consumed via web services Handling Failures • Every Integration point will eventually fail one time or another • Be prepared to handle all kind of failures Horizontal Scalability • Design for Scale Out Asynchronous Processing • Break down the task, process requests asynchronously • Use queues to decouple functionality • Eventual consistency model Stateless Model • Build stateless services that can be scaled out and load balanced Minimize Human Intervention • Go DevOps/NoOps
The Steps towards Cloud-Enabled Applications First do the tool-based fitment analysis on all the applications to be moved to cloud environments.This includes the check of compatible environments in the cloud to host, configure, manage and deliver those applications To understand the dependencies (application, data, etc.) To do a deeper analysis of the business, technical and use cases of moving applications to clouds The barriers such as the controllability, security, availability, performance, scalability, etc. need to be taken into consideration before the migration To check whether appropriate cloud integrators, brokers and orchestrators are made available in order to ensure seamless and spontaneous access, leverage and compose them together
Cloud Application Domains The general trend is that all kinds of ICT applications are accordingly modified, modernized to be multi-tenant,migrated to cloud servers and delivered to worldwide users concurrently via the open web.That is, software as a service (SaaS) is the progressive paradigm and “everything as a service” is the ultimate vision with the maturity and stability of the emerging and evolving cloud concepts. Primarily Social Networking (Web 2.0) Sites E-Business,E-Commerce, & Mobile Applications Enterprise Information Systems (EISs) such as ERP, CRM, KM, CM, SCM, etc. Images &Video Processing, Storage,Analysis, Management and Surveillance Big Data capture, storage, mining, processing, and analyses for descriptive, predictive and prescriptive analytics towards real-time business intelligence (BI) The Internet of Things (IoT) / Machine-to-Machine (M2M) Integration / Cyber Physical Systems (CPS) / Smarter EnvironmentsApplications and Services Different kinds of Platforms and Databases are being primed for Clouds
The Cloud Applications Architectures
A Sample Next-Gen Cloud Application Cloud-supported Next Gen Applications
Cloud-based Smart Healthcare
Big Data Analytics in Clouds
The Special-Purpose Clouds 1. Sensor Cloud 2. Device Cloud 3. Mobile Cloud 4. Government Cloud
The Sensor Cloud An infrastructure that allows truly pervasive computation using sensors as interface between physical and cyber worlds, the data-compute clusters as the cyber backbone and the Internet as the communication medium. It integrates large-scale sensor networks with sensing applications (Smarter Homes, Hospitals, Hotels, Cities, etc.) and cloud computing infrastructures. It collects and processes data from various sensor networks. Enables large-scale data sharing and collaborations among users and applications on the cloud. Delivers cloud services via sensor-rich mobile devices. Allows cross-disciplinary applications that span organizational boundaries.
Sensor – Cloud Integration: Use Cases
Sensor-to-Cloud Integration Acquisition of data feeds from numerous body area (blood sugar, heat, perspiration, etc.) and wide area (water quality, weather monitoring, etc.) sensor networks in real time. Real-time processing of heterogeneous data sources in order to make critical decisions. Automatic formation of workflows and invocation of services on the cloud one after another to carry out complex tasks. Highly swift data processing using the immense processing power of the cloud to provide quick response to the user.
The Government Cloud Governments are mandated to vigorously and rigorously pursue incorporating smartness in their service conceptualization, concretization and delivery aspects to that they can empower their constituents and citizens with a bevy of next-generation personal, social, professional services in a time-bound, responsible, transparent, and proactive manner. The need therefore is to establish and sustain service-oriented clouds. Service Delivery Platform (SDP) is a major ingredient in fulfilling the above- mentioned goals in an automated way. As citizen-centric services are undergoing a variety of changes and challenges, an unified and standards-compliant platform is the most sought-after for governments to strengthen the service delivery mechanism.
The Government Cloud Services
A Device Cloud Architecture
A Device Cloud Architecture
A Mobile Cloud Architecture
A Mobile Cloud Architecture
The Concerns & Research Challenges of Cloud Computing
The General Concerns of Cloud computing Integration & Orchestration Interoperability & Portability Visibility & Controllability Trust Deficit Accountability & Auditability Vendor Lock-In Security & Privacy Availability
The Cloud – Research Challenges EffectiveHandlingofBigDataforReal-timeAnalytics DealingwithReal-timerequirementsparticularlystreamed multimedia PerformanceEngineering&Enhancement(PE2) Approaches SecurityAlgorithms(Message/Data,Application/Service,Network, &PhysicalSecurity),Identity,Access,&KeyManagements EnergyEfficiency&Elasticity LoadBalancing,JobScheduling,WorkloadautomationResource ProvisioningandDe-ProvisioningAlgorithms ResourceManagement&UsageBilling
The Cloud – Research Challenges Performability Analysis for Infrastructure Cloud (Stochastic Models Approach) Dynamic Capacity Planning for Infrastructure Cloud Replication Scheme for Cloud Storage Secure and Trustworthy Clouds Performance Interference Effects for QoS-Aware Clouds Scalability, consistency & economical processing of large scale data on the cloud High Performance Scientific Applications in Cloud Automatic IO Filtering for Optimizing Cloud Analytics Scheduling Cloud Capacity for Time-Varying Customer Demand Shared Resource Monitoring and Throughput Optimization in Clouds Integrating Graph Partitioning into large graph processing in the cloud Exploiting Performance Heterogeneity in Public Clouds Dynamically Scaling Applications in the Cloud Performance Estimation and Enhancement – Measurement-based Approach
The Cloud – Research Challenges Managing Parallelism for Stream Processing in the Cloud Profit-Based Experimental Analysis of IaaS Cloud Performance Clustering Techniques for High Availability Green Techniques for energy-aware Clouds Indexing Multi-dimensional Data in a Cloud System Adaptive Provisioning of Stream Processing Systems in the Cloud Cloud Micro-Elasticity via VM State Coloring Variations in Performance and Scalability when Migrating n-Tier Applications to Different Clouds Self-adaptive Cloud Capacity Planning Automated Provisioning and De-Provisioning Techniques Data integrity in the cloud - the Correctness of Cloud-based Data
A Sample List of Book Chapters
Pethuru Raj PhD email@example.com www.peterindia.net http://www.linkedin.com/in/peterindia https://www.facebook.com/sweetypeter
A Weekly Radio Show by Business Leaders for Business Leaders. Cloud Talk Radio is a weekly show in which co-hosts Robert J. Chandler, President & CEO of ...
Cloud-Talk. Innerhalb dieser Gruppe werden Technologie, Rahmenbedingungen, Planungs- und Absicherungsaspekte von Cloud-Strategien diskutiert.
The fastest way to grow your business with the leader in Technology CloudTalk focuses on brand development and site-use sophistication.
Cloud-Talk. Innerhalb dieser Gruppe werden Technologie, Rahmenbedingungen, Planungs- und Absicherungsaspekte von Cloud-Strategien diskutiert.
Glenfis AG Zürich, Cloud and Sourcing Strategy Coaching and Training,Cloud Migration and Defining a Cloud Target Operation Model,TOM
Cloud storage made easy. Just Cloud offers ultimate online storage, syncing and file sharing. JustCloud cloud storage is rated number 1 on many comparison ...
This is the talk page for discussing improvements to the Cloud article. This is not a forum for general discussion of the article's subject. Put new text ...
SoundCloud is an audio platform that lets you listen to what you love and share the sounds you create.
Ich stimme zu, dass diese Seite Cookies für Analysen, personalisierte Inhalte und Werbung verwendet.
iCloud makes sure you always have the latest versions of your most important things — documents, photos, notes, contacts, and more — on all your devices.