Advanced Information Systems Lecture 7: Introduction to Virtualization PDF

# Advanced Information Systems ## Lecture 7: Introduction to Virtualization ### Virtualization Refers to the logical abstraction of physical resources, such as compute, network, and storage that enables a single hardware resource to support multiple concurrent instances of systems or multiple hard...

# Advanced Information Systems ## Lecture 7: Introduction to Virtualization ### Virtualization Refers to the logical abstraction of physical resources, such as compute, network, and storage that enables a single hardware resource to support multiple concurrent instances of systems or multiple hardware resources to support a single instance of a system. - Enables a resource to appear larger or smaller than it actually is - Enables a multitenant environment improving utilization of physical resources ### Benefits of Virtualization - Optimizes utilization of IT resources - Reduces cost and management complexity - Reduces deployment time - Increases flexibility ### Virtual Layer Overview - Virtualized compute, network, and storage forms the virtual layer. - Enables fulfilling two characteristics of cloud infrastructure: - Resource pooling - Rapid elasticity - Specifies the entities operating at this layer: - Virtualization software - Resource pools - Virtual resources ### Virtual Layer: Virtualization Process and Operations A diagram shows the process of creating a virtualized environment. **Step 1:** Deploy virtualization software on: - Compute systems - Network devices - Storage devices **Step 2:** Create resource pools: - Processing power and memory - Network bandwidth - Storage **Step 3:** Create virtual resources: - Virtual machines - Virtual networks - LUNS **Output:** Virtual resources are packaged and offered as services ### Compute Virtualization Software #### Hypervisor Software that is installed on a compute system and enables multiple OSs to run concurrently on a physical compute system. - **Hypervisor kernel:** - Provides functionality similar to an OS kernel. - Designed to run multiple VMs concurrently - **Virtual machine manager (VMM):** - Abstracts hardware - Each VM is assigned a VMM. - Each VMM gets a share of physical resources A diagram shows a hypervisor kernel and its associated virtual machines. #### Types of Hypervisor - **Bare-Metal Hypervisor**: - It is an operating system. - Installed on a bare-metal hardware. - Requires certified hardware - Suitable for enterprise data centers and cloud infrastructure - **Hosted Hypervisor**: - Installed as an application on an OS. - Relies on OS, running on a physical machine for device support. - Suitable for development, testing, and training purposes A diagram shows the difference between a bare-metal hypervisor and a hosted hypervisor. ### Network Virtualization Software - Abstracts physical network resources to create virtual resources: - Virtual LAN/virtual SAN - Virtual Switch - Network virtualization software can be: - Built into the operating environment of a network device - Installed on an independent compute system - Fundamental component for deploying software-defined networks - Hypervisor's capability ### Storage Virtualization Software - Abstracts physical storage resources to create virtual resources: - Virtual volumes - Virtual disk files - Virtual arrays - Storage virtualization software can be: - Built into the operating environment of a storage device - Installed on an independent compute system - Fundamental component for deploying software-defined storage - Hypervisor's capability ### Introduction to Resource Pool A logical abstraction of the aggregated computing resources, such as processing power, memory capacity, storage, and network bandwidth, that are managed collectively. - Cloud services obtain computing resources from resource pools. - Resources are dynamically allocated as per consumer demand. - Resource pools are sized according to service requirements. A diagram shows resources for aggregation, a resource pool, service A, consumer A and consumer B. ### Example: Pooling Processing Power and Memory Capacity A diagram shows the pooling of processing power and memory capacity into one resource pool. It illustrates two consumers, who are each assigned their respective processing power and memory capacity. - The processor pool's processing power is 12000MHz - The memory pools capacity is 18GB ### Example: Pooling Storage in a Disk-Based Storage System A diagram shows the pooling of storage in a block-based storage system. It illustrates the storage pool, which is divided into four drives with a total capacity of 4000 GB. This pool is shared among three consumers, each of whom is assigned a specific amount of storage space from the storage pool. - The storage pool's capacity is 4000 GB. ### Example: Pooling Storage Across Block-Based Storage Systems A diagram shows the pooling of storage across block based storage systems. It illustrates a storage pool shared among four block-based storage systems, which are in turn assigned to three consumers. Each consumer utilizes their assigned storage space from the pool. - The storage pool capacity is not specified. ### Example: Pooling Network Bandwidth of Network Interface Cards (NICs) A diagram shows the pooling of network bandwidth, assigned to two consumers through a hypervisor. Each consumer is assigned a different bandwidth amount from the network bandwidth pool. - The network bandwidth pool is 3000 Mbps. ### Identify Pool Specifies a range of network identifiers (IDs) such as virtual network IDs and MAC addresses. - IDs are allocated from the identity pools to the elements of cloud services - An identity pool may map to a particular service or to a group of services. A diagram shows two identity pools, each assigned to a corresponding consumer. ### Virtual Machine (VM) A logical compute system that, like a physical compute system, runs an OS and applications. - Created by a hypervisor installed on a physical compute system. - Comprises virtual hardware, such as virtual processor, memory, storage, and network resources. - Appears as a physical compute system to the guest OS. - Hypervisor maps the virtual hardware to the physical hardware. - Provider provisions VMs to consumers for deploying applications. - VMs on the same compute system or cluster run in isolation. ### VM Hardware A diagram shows the elements that may be included in the hardware of a virtual machine: - Processor - SCSI/IDE Controllers - NICs - RAM - Graphics card - Disk drive - Serial/Parallel ports - HBAS - Mouse - Floppy/Optical drives and controllers - Keyboard - USB Controller ### VM Files From a hypervisor's perspective, a VM is a discrete set of files such as: - **Configuration file:** Stores information such as VM name, BIOS information, guest OS type, memory size. - **Virtual disk file:** Stores the contents of the VM's disk drive. - **Memory state file:** Stores memory contents of a VM in a suspend state. - **Snapshot file:** Stores the VM settings and virtual disk of a VM. - **Log file:** Keeps a log of VM activity and is used in troubleshooting. ### File System to Manage VM Files - **Hypervisor's Native File System**: - Clustered file system deployed on local or external storage. - Enables multiple hypervisors to perform concurrent reads and writes. - Enables high availability to protect against hypervisor or compute system failure. - **Shared File System:** - Enables storing VM files on remote file servers or NAS devices. - Hypervisors have built-in NFS or CIFS clients. A diagram shows the native file system and shared file system in relation to hypervisors and physical compute systems. ### VM Console - VM console is an interface to view and manage VMs on a compute system or a cluster. - VM console may be: - Installed locally on a compute system. - Web-based. - Accessed over a remote desktop connection. - Used to perform activities such as: - Installing a guest OS and accessing VM BIOS. - Powering a VM on or off. - Configuring virtual hardware and troubleshooting. ### VM Template - A master copy of a VM with standardized virtual hardware and software configuration that is used to create new VMs. - Created in two ways: - Converting a VM into a template. - Cloning a VM to a template. - Steps involved in updating a VM template are: - Convert the template into a VM. - Install new software or OS/software patches. - Convert the VM back to a template. ### Virtual Appliance - Preconfigured virtual machine(s) preinstalled with a guest OS and an application dedicated to a specific function. - Used for functions, such as providing SaaS, routing packets, or deploying a firewall. - Simplifies the delivery and operation of an application. - Simplifies installation and eliminates configuration issues. - The application is protected from issues in other virtual appliances. - Typically created using Open Virtualization Format (OVF). ### VM Network - A logical network that provides Ethernet connectivity and enables communication between VMs within a compute system. A diagram shows the general setup for VM networks and their connection to clients. ### VM Network Components | Component | Description | | ------------- | ------------- | | Virtual switch | A logical OSI Layer 2 Ethernet switch created in a compute system. Connects VMs locally and also directs VM traffic into a physical network. Forwards frames to a virtual switch port based on destination address. A distributed virtual switch can function across multiple physical compute systems. | | Virtual NIC | Connects a VM to virtual switch and functions like a physical NIC. Has unique MAC and IP addresses. Forwards the VM's network I/O in the form Ethernet frames to the virtual switch. | | Uplink NIC | A physical NIC connected to the uplink port of a virtual switch. Functions as an inter-switch link between virtual and physical Ethernet switches. Not addressable from the network. | ### Logical Unit Number (LUN) Abstracts the identity and internal functions of storage system(s) and appear as physical storage to the compute system. - Mapping of virtual to physical storage is performed by the virtualization layer. - Provider provisions LUN to consumers for storing data. - Storage capacity of a LUN can be dynamically expanded or reduced. - LUN can be created from: - RAID set (traditional approach) - Storage pool ### Creating LUNs from a RAID Set LUNs are created from a RAID set by partitioning the available capacity into smaller units, which are spread across all the physical disks that belong to a RAID set. - This is suited for applications that require predictable performance. A diagram shows the creation of LUNs from a RAID set, including a virtual volume, a physical disk, and a compute system. ### Creating LUNs from a Storage Pool Two types of volumes are created from storage pool: - **Thin LUN:** - Does not require storage to be completely allocated at creating time. - Consumes storage as needed from the underlying storage pool in increments called thin LUN extents. - **Thick LUN:** - Physical storage is completely allocated at creating time. A diagram shows how thin and thick LUNs are created from a storage pool. ### Use of Thin LUN - Thin LUNs are appropriate for applications that can tolerate performance variations. - In some cases, performance is improved when using a thin volume due to striping across large numbers of drives in the pool. - Environments where cost, storage utilization, space, and energy efficiency are paramount. - For applications where storage space consumption is difficult to forecast. - An environment that needs optimized self provisioning. ### Virtual Network - A software-based logical network that is either a segment of a physical network or spans across multiple physical networks. - Appears as a physical network to the connected nodes. - Virtual networks share network components without leaking information between them. - Network traffic is routed only when two nodes in different virtual networks are communicating. - All types of networks can be virtualized, such as compute network, SAN, and a VM network. ### Virtual Network Example A diagram shows the configuration of two VM networks, which are connected via a physical switch and an IP router. ### Virtual LAN (VLAN) - A virtual network created on a LAN that enables communication between a group of nodes with a common set of functional requirements, independent of their physical location in the network. - A VLAN is identified by a unique 12-bit VLAN ID. - Configuring a VLAN: - Define VLAN on physical and virtual switches and assign a VLAN ID. - Configure VLAN membership based on port, MAC address, protocol, IP subnet address, or application. ### Private VLAN (PVLAN) - A sub-VLAN that segregates the nodes within a standard VLAN, called the primary VLAN. A PVLAN can be configured as either isolated or community. - Enables a provider to support a larger number of consumers. - Provides security between nodes on the same VLAN. - Simplifies network management. A diagram shows the configuration of a PVLAN. ### Stretched VLAN - A VLAN that spans multiple sites and enables Layer 2 communication between a group of nodes over a Layer 3 WAN infrastructure, independent of their physical location. - Layer 2 WAN frames are encapsulated in Layer 3 WAN packets. - Enables movement of VMs across locations without changing their network configuration. A diagram shows the configuration of a stretched VLAN. ### Virtual SAN (VSAN) - A logical fabric created on a physical FC or FCoE SAN, enabling communication between a group of nodes with a common set of requirements, independent of their physical location in the fabric. - A VSAN has its own fabric services, configuration, and set of FC addresses. - Traffic disruptions in one VSAN do not affect other VSANs. - A VSAN may be extended across sites similar to a stretched VLAN.