Introduction to Cloud Computing Nabil Abdennadher nabil.abdennadher@hesge.ch 2017/2018 1
Plan Context Definition Market Cloud service models Cloud deployments models Key drivers to adopting the Cloud Barriers to Cloud 2
Context On-premise IT Cloud computing 3
Context Cloud is treated as an Operating Expense (OPEX), not a Capital Expense (CAPEX) What is OPEX? What is CAPEX? What is Total Cost of Ownership (TCO)? 4
Context Capital Expenses (CAPEX) Expenditures creating future benefits. Incurred when a business spends money either to buy fixed assets add to the value of an existing fixed asset Asset life extends beyond the taxable year Examples acquiring fixed, and in some cases, intangible assets repairing an existing asset so as to improve its useful life upgrading an existing asset if its results in a superior fixture starting or acquiring a new business 5
Context Operational Expenses (OPEX) Ongoing cost for running a product, business, or system Day-to-day expense Examples License fees and / or rental fees Maintenance and repairs Attorney and legal fees Utilities, such as telephone, subscriptions, etc. Insurance Travel and vehicle expenses 6
Context ICT related CAPEX Computer hardware and programs Network hardware and software Installation and integration of hardware and software Maintenance, warranties and support licenses 7
Context ICT related OPEX Rented space (hosting, data center) Electricity (for related equipment, cooling, backup power) Testing costs, downtime, outage and failure expenses Backup and recovery process Technology training Audit 8
Context What is my ICT consumption model? 9
10 Context
Concept of busy hours Typical weekly traffic to Amazon's e-commerce web site in 2007 11
Concept of busy hours 12 Traffic in the month of November 2007
13 Context
Plan Context Definition Market Cloud service models Cloud deployment models Key drivers to adopting the Cloud Barriers to Cloud 14
Definition There are several existing definitions of the term cloud computing from different points of view: Analyst firms Academics Industry IT companies 15
Definitions by analyst firms A style of computing in which massively scalable ITrelated capabilities are provided as a service using internet technologies to multiple external customers (Gartner 2008) An emerging IT development, deployment and delivery model, enabling real-time delivery of products, services and solutions over the internet (IDC 2008) the idea of delivering personal and business productivity applications from centralized servers (Merrill Lynch 2008) 16
Definitions by analyst firms These definitions have a common characteristics: They define CC from the perspective of the end users They focus on how it might be experienced by end users The core feature of CC is the provision of IT infrastructure and applications as a service in scalable way 17
On demand self service. 18 National Institute of Standards and Technology (NIST) definition A consumer can unilaterally provision computing capabilities... Broad network access Capabilities are available over the network and accessed through standard mechanisms Resource pooling. The provider s computing resources are pooled to serve multiple consumers Rapid elasticity Capabilities can be elastically provisioned and released Metering capability Cloud systems automatically control and optimize resource use by leveraging a metering capability http://nvlpubs.nist.gov/nistpubs/legacy/sp/nistspecialpublication8 00-145.pdf
Is cloud computing a technology? No It s a combination of pre-existing technologies 19 These technologies have matured at different rates, and were not designed as a coherent whole They have come together to create a technical ecosystem for the cloud computing These technologies are: Cloud access devices (advances in processors) Browsers and thin clients Broadband internet connection Fast inexpensive servers Virtualisation technology API
Cloud access devices The range of access devices for the cloud has expanded Home PCs, enterprise PCs, mobile phones devices, etc. are on line Example: Growth of the ios & Android and the proliferation of applications available on App store and Google market 20
Browsers and thin clients Users can now access applications from wherever they can load browser Even enterprise (SAP, Oracle, etc.) applications use browser interface The use of browsers has democratized Intuitive context No need of training (unlike fat client programs) 21
High-speed broadband access One of the most important ingredients of cloud computing It enabled mobile devices (entry points) to access to the cloud resources It s one of the most substantial differences from old utility computing concept (35 years ago) 22
Data centres and server farms Cloud services require large computing capacity These resources are usually geographically distributed Example Google Amazon (EC2) Saleforce.com 23
Virtualization Virtualization is one of the cornerstones of cloud computing IaaS providers use OS and storage virtualization to enable customers run instances of various OS in a cloud. In addition to OS and storage virtualisation, SaaS and PaaS providers implement software and database virtualization 24
API Used to invoke services supported by the cloud X -aas developers need to become familiar with specific API to deploy and manage software modules to the X -aas platform Problem: there are as many API as Cloud Service Providers (CSP). Challenge: Standardization 25 Universal Cloud Interface (UCI) Open Cloud Computing Interface (OCCI) Cloud Infrastructure Management Interface (CIMI)
To summarize from a non technical perspective Cloud computing is not the savior of IT. It is nothing but a way to deploy your enterprise architecture in a way that has the potential to be more productive and cost effective. In essence, it is a tool, not a way of life. It is not magic, it is not even new, but if approached correctly, it could be a path towards efficiency. 26 Cloud Computing and SOA Convergence in Your Enterprise 2009
To summarize from a technical perspective CC is based on pay-per-use business models Main features of CC are based on virtualization and dynamic scalability on demand Cloud services are consumed either via web browser or defined API CC is Elastic and massively scalable Self-provisioning of resources 27
Plan Context Definition Market Cloud service models Cloud deployment models Key drivers to adopting the Cloud Barriers to Cloud Examples 28
Cloud related-market forecast Dearth of information about just how big the cloud business is Until 2014, AWS revenue is lumped in with a number of other peripheral Amazon business units in the other category until April 2015. Amazon, in its first-quarter results, announced for the first time AWS figures: $1.57 billion revenue revenue increased 49% from a year earlier http://www.forbes.com/sites/benkepes/2014/07/29/just-how-big-isamazons-cloud-business/#296580ef55e5 29
Gartner Magic Quadrant for IaaS 30
Cloud Computing Market 31
Cloud Computing Market 32
33 Swiss perspective
Plan Context Definition (attempt to define) Market Cloud service models Cloud deployments models Key drivers to adopting the Cloud Barriers to Cloud 34
Cloud Services Models Cloud Computing Service Models - User Mapping Google apps, Salesforce, etc. Google App Engine (GAE), Force.com. Amazon, Azure, hepiacloud. Source: www.appectual.com 35
Software-as-a-Service (SaaS) Traditional methods of purchasing software 36 capital expenditure (capex) Isolated, single tenant model SaaS operating expenditure (opex) Multitenant architecture model Benefits of SaaS Outsource hosting and management of applications to a third party Limit copying and distribution, guarantee an ongoing revenue stream without preloading software User level (end users)
Platform-as-a-Service (PaaS) PaaS : the development environment is hosted in the cloud and accessed through a browser PaaS is a variation of SaaS development environment is offered as a service Deployment is done without any specialized system administration skills Example Develop web applications using desktop development tools, such as Eclipse 37
Platform-as-a-Service (PaaS) To build web applications, developers need specialized skills 38 Backend server development (ex: Java/J2EE) Frontend client development (ex: javascript, etc.) Website administration PaaS allow general developers to setup web applications without any specialized expertise Google App. Engine is a good example PaaS offers to democratize the development of web applications. Developers level
Infrastructure-as-a-Service (IaaS) The service provider rents to its clients 39 Machines (CPU) Disks network connections... using virtualization technology. On a virtual machine the user accesses a standard operating system environment and is able to install and configure all the layers on top of it. Examples : Amazon Web Service Elastic, Compute Cloud (EC2), Microsoft Azure, Google Compute Engine IT level
40 Governance in the cloud
41
Plan Context Definition (attempt to define) Market Cloud service models Cloud deployment models Key drivers to adopting the Cloud Barriers to Cloud 42
Public cloud computing A public cloud computing is hosted, operated and managed by a third-party vendor The service is offered to multiple customers over a common infrastructure 43 Cloud Security and Privacy, Tim Mather et al, 2008
Private Cloud Computing Emulates cloud computing on private networks Based on products which automate virtualization Does not take advantage of the pay-per-use basis 44
Hybrid Cloud Computing Execute core applications and sensitive data on private clouds Non-core applications are deployed on public cloud 45 Cloud Security and Privacy, Tim Mather et al, 2008
Plan Context Definition Market Cloud service models Cloud deployment models Key drivers to adopting the Cloud Barriers to Cloud 46
Key drivers to adopting the Cloud Small initial investments and low ongoing costs Economies of scale Open standards Sustainability https://www.getfilecloud.com/blog/2014/09/exploring-the-key-driversbehind-enterprise-cloud-adoption/#.v9qb-pn961t 47
Small initial investments and low ongoing costs When setting up a new business, building an IT department is a low priority compared to R&D, marketing, securing the next round of funding No hardware, software, network devices have to be purchased 48
Open standards Most of the CC are based on open standards Open standards are essential to allow for continued growth in the cloud, they are the foundation of the cloud Examples: GNU/Linux, ssh, http, etc. BUT so far, there are no adopted standards regarding Cloud API ( access to cloud resources) 49
Sustainability Traditionally, companies periodically invest in order to keep their IT services up-to-date. The objective is to: avoid failure keep pace with business changes With Cloud computing, companies rely on their Cloud Service Providers to minimise failures 50
Plan Context Market Definition Cloud service models Cloud deployments models Key drivers to adopting the Cloud Barriers to Cloud 51
Security Privacy Barriers to Cloud (technical challenges) New cloud services are introduced in rapid pace Tools are continuously evolving Moving large data is expensive Quality of Service Internet dependence http://www.business2community.com/cloud-computing/moving-cloud-top- 5-barriers-cloud-adoption-break-0987489#ez7M2Fzr6B5EUWST.97 52
Barriers to Cloud (non technical challenges) Not Invented Here (NIH) syndrom Vendor lock-in Security-risks Service Level Agreements Legal Political 53
To conclude Cloud computing is for the information age what electrification is for industrial age in the end, the saving offered by utilities become too compelling to resist, even for the largest enterprises The big switch, W.W. Norton et al, 2008 54