1. 1 Cloud Computing CS 325 4/26/2016

2. “Computation may someday be organized as a public utility.” - John McCarthy, 1961 2

3. Cloud Computing  No longer the next big thing – the current big thing  Began in 2007 – IBM and Google “Blue Cloud”  Name cloud inspired by cloud symbol representing internet in diagrams  Amazon popularized idea of the cloud 3

4. Questions to answer  What clouds have you used today?  What is a cloud? 4

5. Introduction - Cloud Definition Cloud computing is a set of service- oriented architectures, which allow users to access a number of resources in a way that is elastic, cost-efficient, and on- demand. 5

6. Introduction - Cloud Definition The U.S. National Institute of Standards and Technology (NIST) defines cloud computing as:  A model for enabling ubiquitous, 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. 6

7. Introduction - Cloud Definition Cloud Definition  Scalable resource allocation  Tailored services  Software as a Service (SaaS)  Platform as a Service (PaaS)  Infrastructure as a Service (IaaS)  Billed like a utility  public clouds 7

8. Cloud Computing  Everyone has an opinion on how to use the cloud  Applications on the internet – email, tax prep  Storage for business, personal data  Web services for photos, maps, GPS  Rent a virtual server, load software on it, turn it on /off, clone it if sudden workload demand  Store, secure data for authorized access (really?)  Use a platform including OS, Apache, MySQL, Python, PHP 8

9. Cloud Computing Characteristics  So what are its characteristics? Described as: On-demand computing, pay as you go, software as a service, utility computing Usually costs, but cost-effective Emphasizes availability Virtualization Scalable (expand on current hardware) Elastic (dynamically add hardware as needed) Distributed and highly parallel approach Replication, replication, replication … 9

10. Characteristics of Cloud Computing  On-demand self service:  Resources can be provisioned on-demand by the users, without requiring interactions with the cloud service provider. The process of provisioning resources is automated.  Broad network access:  Resources can be accessed over the network using standard access mechanisms that provide platform- independent access. Using heterogeneous platforms such as workstations, laptops, tablets and smartphones. 10

11. Characteristics of Cloud Computing  Resource pooling:  The computing and storage resources provided by cloud service providers are pooled to serve multiple users using multi-tenancy. Cloud architectures allow multiple users to be served by the same physical hardware.  Rapid elasticity:  Cloud computing resources can be provisioned rapidly and elastically. Resources can be rapidly scaled up or down based on demand. 11

12. Characteristics of Cloud Computing  Multi-tenancy:  The multi-tenanted approach of the cloud allows multiple users to make use of the same shared resources.  In virtual multi-tenancy, computing and storage resources are shared among multiple users.  In organic multi-tenancy every component in the system architecture is shared among multiple tenants 12

13. Types of Clouds  Public Cloud  Marketed based on Resources offered, availability, security, price  Local/Private Cloud  Cloud architectures tailored to an organization’s needs.  Hybrid Cloud  Combination of public and local cloud resources. 13

14. Large Public Clouds 14

15. Popular Open Source Clouds 15

16. What Motivated Cloud Computing Initial motivation:  Web-scale problems Data, processing intensive problems Crawling, indexing, searching the Web Sensor networks Solutions:  Large data centers Hundreds of thousands of compute nodes Redundant VMs, data backup How to access:  Highly-interactive Web applications (thin client) Low power/performance device with network interface 16

17. Public clouds such as Amazon AWS and Microsoft Azure are great, but what if an organization wants to build their own local cloud? 17

18. Seven Principles of Cloud Computing Open Architecture (CCOA) 1. Integrated Ecosystem Management for Cloud 2. Virtualization for Cloud Infrastructure 3. Service-Orientation for Common Reusable Services 4. Extensible Provisioning and Subscription for Cloud 5. Configurable Enablement for Cloud Offerings 6. Unified Information Representation and Exchange Framework 7. Cloud Quality and Governance 18

19. Seven Principles of Cloud Computing Open Architecture (CCOA) 1. Integrated Ecosystem Management for Cloud  Includes all involved services and solutions vendors, partners, and end users to provide shared resources. 2. Virtualization for Cloud Infrastructure  The cloud infrastructure management module covers software image management, hardware virtualization, and legacy application packaging.  Virtualized resource management, interfacing, networking, etc. 3. Service-Orientation for Common Reusable Services  Horizontal and Vertical cloud architectures. Horizontal approach – general solution cloud (Amazon AWS) Vertical approach – specific solution cloud (Dropbox) 19

20. Seven Principles of Cloud Computing Open Architecture (CCOA) 4. Extensible Provisioning and Subscription for Cloud  Handles service providers provisioning process and service consumers’ subscription process.  Availability, Security, SLA 5. Configurable Enablement for Cloud Offerings  IaaS - Virtualized resources  PaaS - Deployment frameworks/platforms  SaaS - End user software 6. Unified Information Representation and Exchange Framework  Messages exchanged among cloud clients, partners and vendors. 20

21. Seven Principles of Cloud Computing Open Architecture (CCOA) 7. Cloud Quality and Governance  Responsible for the identification and definition of quality indicators for Cloud Computing environment  A set of descriptions used for guidance to govern the design, deployment, operation, and management of the cloud offerings. 21

22. Local Cloud Architectures - IaaS  Local Cloud?  Small to medium sized Up to 100 compute nodes  What resources would we need to do this? Compute Servers Persistent Storage Servers VM Image Server(s) Cloud Administrative Server(s) Network Infrastructure –Copper 22

23. Local Cloud Architectures - IaaS  Compute Servers  CPU, RAM, Local disk (magnetic, SSD) resources given to the user. In the form of virtual machines.  Hosts virtual machines using a hypervisor Software that creates and run virtual machines Xen, KVM, ESXi –Grid VGX  Hybrid approach to hypervisor selection is common. CPU intensive Disk I/O intensive Network intensive 23

24. Local Cloud Architectures - IaaS  Persistent Storage Servers  What they are: VMs hosted on the compute servers are stateless.  What they do: Used for long term storage of data.  Virtual Machine Image Server  Modified Persistent Storage Server. Repository of available VM images. 24

25. Local Cloud Architecture IaaS Architecture 25

26. Local Cloud Architectures – Managing Resources in IaaS Creating the Local Infrastructure as a Service Cloud 26

27. Local Cloud Architectures – Managing Resources in IaaS  Virtual machines are created to provide computing resources to users.  These VM’s can be located on many heterogeneous compute nodes.  They can be dynamically launched Variable virtual CPU cores, RAM, local Disk Space, etc.  Supporting middleware is needed for management of virtualized resources  Middleware communication between head cloud node and all cloud compute nodes using sockets. 27

28. Local Cloud Architectures – Managing Resources in IaaS  VM network attributes:  Each VM must have a unique IP address for communication over the network. This requires modification of the host OS’s network configuration files.  Typically, the compute node will have one NIC connected to a local switch. All VM and host OS traffic uses this NIC for network communication.  To allow unique IP addresses for each VM (and communication to/from the VM) a network bridge must be created. This allows VMs with randomly generated MAC addresses to attach to a specific NIC interface –Ex: Bridge br0 attaches to NIC interface p2p1 (or eth0).  Since the VM has a bridged NIC connection, it can request (globally) unique IP addresses from the DHCP service hosting physical machines in the local network.  It’s also possible for management of VM IP address by subnet with two NIC interfaces on the cloud compute node. 28

29. Local Cloud Architectures – Managing Resources in IaaS  Head node communication with compute nodes:  Once a VM is requested (web interface, dynamic scaling, etc.), the head node’s client process polls all available compute nodes Number of current active VMs Amount of free RAM CPU utilization (system load)  The head node uses this information to determine placement of the new VM. (load balancing) Based on data polled from available compute nodes, the head node sends a request to the compute node best suited to support the new VM. Users connect and use the virtualized resources provided by the VM. The use has the ability to power down the VM. –Quotas can be placed on users of local IaaS cloud architectures. 29

30. Local Cloud Architectures – Managing Resources in IaaS  VM load balancing vs. consolidation  Issues (potentially) with VM load balancing Energy consumption of compute nodes  It is possible to reduce consumption of the local IaaS cloud: VM load consolidation  VM load consolidation, implemented in the IaaS middleware, allows new and active VMs to execute on as few compute nodes as possible. Based on historical data, the cloud middleware can alternate between load balance and load consolidate. –Ex: load balance requests during peak hours for increased performance –Ex: load consolidate during non-peak hours for energy savings Trade-off: –Load balance: consumes more energy, higher performance per VM –Load consolidation: consumes less energy, potentially decreased performance per VM 30

31. Opportunities and Challenges  The use of the cloud provides a number of opportunities:  It enables services to be used without any understanding of their infrastructure.  Cloud computing works using economies of scale: It potentially lowers the expense for start up companies, as they would no longer need to buy their own software or servers. Cost would be by on-demand pricing. Vendors and Service providers claim costs by establishing an ongoing revenue stream.  Data and services are stored remotely but accessible from “anywhere”. 31

32. Opportunities and Challenges  In parallel there has been backlash against cloud computing:  Use of cloud computing means dependence on others and that could possibly limit flexibility and innovation: The others are likely become the bigger Internet companies like Google and IBM, who may monopolise the market.  Security could prove to be a big issue: It is still unclear how safe out-sourced data is and when using these services ownership of data is not always clear.  There are also issues relating to policy and access: If your data is stored abroad whose policy do you adhere to? What happens if the remote server goes down? How will you access files? There have been cases of users being locked out of accounts and losing access to data. 32

33. Advantages of Cloud Computing  Lower computer costs:  You do not need a high-powered and high-priced computer to run cloud computing's web-based applications.  Since applications run in the cloud, not on the desktop PC, your PC does not need the processing power or hard disk space demanded by traditional desktop software. 33

34. Advantages of Cloud Computing  Reduced software costs:  Instead of purchasing expensive software applications, you can get most of what you need for free-ish! most cloud computing applications today, such as the Google Docs suite.  better than paying for similar commercial software which alone may be justification for switching to cloud applications. 34

35. Advantages of Cloud Computing  Instant software updates:  Another advantage to cloud computing is that you are no longer faced with choosing between obsolete software and high upgrade costs.  When the application is web-based, updates happen automatically available the next time you log into the cloud.  When you access a web-based application, you get the latest version without needing to pay for or download an upgrade. 35

36. Advantages of Cloud Computing  Unlimited storage capacity:  Cloud computing offers virtually limitless storage.  Your computer's current 2 TB hard drive is small compared to the hundreds of PB available in the cloud.  Increased data reliability:  Unlike desktop computing, in which if a hard disk crashes and destroy all your data, a computer crashing in the cloud will not affect the storage of your data. if your personal computer crashes, all your data is still out there in the cloud, still accessible 36

37. Advantages of Cloud Computing  Universal document access:  That is not a problem with cloud computing, because you do not take your documents with you.  Instead, they stay in the cloud, and you can access them whenever you have a computer and an Internet connection  Documents are instantly available from wherever you are, as long as you have an internet connection  Latest version availability:  When you edit a document at home, that edited version is what you see when you access the document at work.  The cloud always hosts the latest version of your documents as long as you are connected, you are not in danger of having an outdated version 37

38. Advantages of Cloud Computing  Easier group collaboration:  Sharing documents leads directly to better collaboration.  Many users do this as it is an important advantages of cloud computing multiple users can collaborate easily on documents and projects  Device independence.  You are no longer tethered to a single computer or network.  Changes to computers, applications and documents follow you through the cloud.  Move to a portable device, and your applications and documents are still available. 38

39. Disadvantages of Cloud Computing  Requires a constant Internet connection:  Cloud computing is impossible if you cannot connect to the Internet.  Since you use the Internet to connect to both your applications and documents, if you do not have an Internet connection you cannot access anything, even your own documents. 39

40. Disadvantages of Cloud Computing  Does not work well with low-speed connections:  A low-speed Internet connection, such as that found with dial-up services, makes cloud computing painful at best and often impossible.  Web-based applications require a lot of bandwidth to download, as do large documents.  Features might be limited:  Many web-based applications are not as full-featured as their desktop-based applications. For example, you can do a lot more with Microsoft PowerPoint than with Google Presentation's web-based offering 40

41. Disadvantages of Cloud Computing  Can be slow:  Even with a fast connection, web-based applications can sometimes be slower than accessing a similar software program on your desktop PC.  Everything about the program, from the interface to the current document, has to be sent back and forth from your computer to the computers in the cloud. 41

42. Disadvantages of Cloud Computing  Stored data might not be secure:  With cloud computing, all your data is stored and replicated in the cloud. The questions is How secure is the cloud?  Can unauthorized users gain access to your data?  Stored data can be lost:  Theoretically, data stored in the cloud is safe, replicated across multiple machines.  But on the off chance that your data goes missing, you may have no physical or local backup. 42

43. Disadvantages of Cloud Computing  HPC Systems:  Not clear that you can run compute-intensive HPC applications that use MPI/OpenMP!  Scheduling is important with these types of applications  General Concerns:  Each cloud systems uses different protocols and different APIs may not be possible to run applications between cloud based systems  Amazon has created its own DB system (not SQL), and workflow system (many popular workflow systems out there) so your normal applications will have to be adapted to execute on these platforms. 43

44. The Future  SaaS clouds used in 77% of all organizations  IaaS clouds used in 67% of all organizations  75% of Industry data resides in local or public clouds  $80B in 2015  Consumer mobility will continue to drive cloud usage  Device connectivity (Internet of Things) and application data sharing becoming mainstream  Cloud will drive increased adoption of Software Defined Networks (SDNs) to create more efficient network architectures  Cloud technologies will reengineer enterprise IT 44