Advanced Information Systems: The Cloud Physical Layer PDF

# Advanced Information Systems ## The Cloud Physical Layer ### Physical Layer Overview - The physical layer comprises physical compute, storage, and network resources - Compute systems execute software of providers and consumers - Storage systems store business and application data - Networks conne...

# Advanced Information Systems ## The Cloud Physical Layer ### Physical Layer Overview - The physical layer comprises physical compute, storage, and network resources - Compute systems execute software of providers and consumers - Storage systems store business and application data - Networks connect compute systems with each other and with storage systems - Networks also connect multiple data centers or multiple clouds to one another ### Introduction to Compute System - A computing platform (hardware, firmware, and software) that runs platform and application software - Executes the provider's as well as the consumers' software - Typically x86-based servers or hosts - Compute systems are provided to consumers in two ways: - Shared hosting: Multiple consumers share compute systems - Dedicated hosting: Individual consumers have dedicated compute systems - Typically providers use compute virtualization and offer compute systems in the form of virtual machines ### Key Components of a Compute System | Component | Description | |---|---| | Processor | An IC that executes software programs by performing arithmetical, logical, and input/output operations | | Random-Access Memory | A volatile data storage device containing the programs for execution and the data used by the processor | | Read-Only Memory | A semiconductor memory containing boot, power management, and other device-specific firmware | | Motherboard | A PCB that holds the processor, RAM, ROM, network and I/O ports, and other integrated components, such as GPU and NIC | | Chipset | A collection of microchips on a motherboard to manage specific functions, such as processor access to RAM and to peripheral ports | ### Software Deployed on Compute Systems | Software | Description | |---|---| | Self-service portal | Enables consumers to view and request cloud services | | Platform software | Includes the software that the provider offers through PaaS | | Application software | Includes the applications that the provider offers through SaaS | | Virtualization software | Enables resource pooling and creation of virtual resources | | Cloud management software | Enables a provider to manage the cloud infrastructure and services | | Consumer software | Includes a consumer's platform software and business applications | ### Types of Compute Systems - Tower compute system - Rack-mounted compute system - Blade compute system #### Tower Compute System - Built in an upright enclosure called a “tower” - Has integrated power supply and cooling - A group of towers occupies significant floor space, requires complex cabling, and generates noise from cooling units - Deploying in large environments may involve substantial expenditure #### Rack-mounted Compute System - Designed to be fixed on a frame called a "rack" - A rack is a standardized enclosure with mounting slots for vertically stacking compute systems - Simplifies network cabling, consolidates network equipment, and reduces floor space use - Administrators may use a console mounted on the rack to manage the compute systems #### Blade Compute System - Comprises an electronic circuit board with only the core processing components - Multiple blades are housed in a blade chassis - The chassis provides integrated power supply, cooling, networking, and management - Blades are interconnected via a high speed bus - Modular design increases compute system density and scalability ### Introduction to Storage System - A storage system is the repository for saving and retrieving electronic data - Providers offer storage capacity along with compute systems, or as a service - Storage as a Service enables data backup and long-term data retention - Cloud storage provides massive scalability and rapid elasticity of storage resources - Typically, a provider uses virtualization to create storage pools that are shared by multiple consumers ### Types of Storage Devices | Device | Description | |---|---| | Magnetic disk drive | Stores data on a circular disk with a ferromagnetic coating, provides random read/write access, most popular storage device with large storage capacity | | Solid-state (flash) drive | Stores data on a semiconductor-based memory, has very low latency per I/O, low power requirements, and very high throughput | | Magnetic tape drive | Stores data on a thin plastic film with a magnetic coating, provides only sequential data access, low-cost solution for long term data storage | | Optical disc drive | Stores data on a polycarbonate disc with a reflective coating, has Write Once and Read Many capability (CD, DVD, BD), low-cost solution for long-term data storage | ### Redundant Array of Independent Disks (RAID) A storage technology in which data is written in blocks across multiple disk drives that are combined into a logical unit called a RAID group. - Improves storage system performance by serving I/Os from multiple drives simultaneously - Provides data protection against drive failures - Three key techniques used for RAID: striping, mirroring, and parity #### RAID Technique: Striping A RAID technique to spread data across multiple drives in order to use the drives in parallel. #### RAID Technique: Mirroring A RAID technique to store the same data simultaneously on two different drives, yielding two copies of the data. #### RAID Technique: Parity A RAID technique to protect striped data from drive failure by performing a mathematical operation on individual strips and storing the result on a portion of the RAID group. ### Common RAID Levels | RAID Level | Description | |---|---| | RAID 0 | Striped set with no fault tolerance | | RAID 1 | Disk mirroring | | RAID 1+0 | Nested RAID (striping and mirroring) | | RAID 3 | Striped set with parallel access and a dedicated parity disk | | RAID 5 | Striped set with independent disk access and distributed parity | | RAID 6 | Striped set with independent disk access and dual distributed parity | ### Data Access Methods | Access Method | Description | |---|---| | Block-level Access | Storage system handles blocks of data and allows lower level access | | File-level Access | Storage system handles complete files and allows higher level access | | Object-level Access | Storage system handles objects and provides API to manage data | ### Storage System Architecture Storage system architectures are all based on the data access methods. Common storage system option are: - Block-based - File-based - Object-based - Unified #### Block-based Storage System - Enables creating and assigning storage volumes to compute systems - Compute system discovers the volumes as local drives - Required file system can be created on the volumes #### File-based Storage System - A dedicated, high-performance file server with storage (also known as Network-attached Storage) - Enables clients to share files over an IP network - Supports data sharing for UNIX and Windows users - Uses a specialized OS that is optimized for file I/O #### NAS Deployment Options - The two common NAS deployment options are: - Traditional NAS (scale-up NAS) - Scale-out NAS - Traditional NAS - Capacity and performance of a single system is scaled by upgrading or adding NAS components - Scale-out NAS - Multiple processing and storage nodes are pooled in a cluster that works as a single NAS device - Addition of nodes scales cluster capacity and performance without disruption #### Object-based Storage System - Stores file data in the form of objects based on data contents and attributes - Uses a flat, non-hierarchical address space - Object contains user data, related metadata, and user-defined attributes - Objects are uniquely identified using object ID #### Unified Storage System - A storage system that combines block, file, and object storage technologies into a single system ### Introduction to Networking - Networking enables data transfer and sharing of IT resources between nodes across geographic regions - Cloud consumers require a reliable and secure network to connect to a cloud and access cloud services - Network connectivity also enables resource aggregation and service mobility across cloud data centers - Multiple clouds may be inter-connected to enable workloads to be moved or distributed - For example: cloud bursting in a hybrid cloud model ### Types of Network Communication Based on the nodes connected by a network, the network communication is broadly categorized as: - Compute-to-compute communication - Compute-to-storage communication - Inter-cloud communication #### Compute-to-compute Communication - Interconnecting physical compute systems enables compute-to-compute communication - Compute-to-compute communication typically uses IP-based protocols - Compute systems connect to a network through physical network card(s) - Physical switches and routers are common interconnecting devices #### Compute-to-storage Communication ##### Storage Area Network (SAN) A network that interconnects storage systems with compute systems, enabling the compute systems to access and share the storage systems. - Based on the protocols they support, SANs can be classified as: - Fibre Channel SAN (FC SAN) - Internet Protocol SAN (IP SAN) - Fibre Channel over Ethernet SAN (FCoE SAN) ###### FC SAN A SAN that uses Fibre Channel (FC) protocol to transport data, commands, and status information between compute and storage systems. - FC provides block-level access to storage - FC offers data transfer speeds up to 16 Gbps - Theoretically, an FC SAN can connect approximately 15 million nodes ###### FC SAN Components | Component | Description | |---|---| | Network adapters | Provide physical interface to a node for communicating with other nodes, Examples: FC HBAs and storage system front-end adapters | | Cables and connectors | Optical fiber cables are predominantly used to provide connectivity, Connectors enable cables to be swiftly connected to and disconnected from ports | | Interconnecting devices | FC switches and directors, Directors have a modular design, a higher port count, and better fault-tolerance, Switches either have a fixed port count or a modular design | ###### Fabric Connect and Addressing - A fabric created with FC switches connects all nodes and enables them to communicate - Each switch in a fabric contains a unique domain identifier (ID) - Each network adapter is physically identified by a 64-bit World Wide Node Name (WWNN) - Each adapter port is physically identified by a 64-bit World Wide Port Name (WWPN) - Each adapter port in a fabric has a unique 24-bit FC address - Fabric assigns FC addresses to adapter ports dynamically ###### Fabric Port Types - **N_Port:** Node port, connects to a compute system or storage system - **F_Port:** Fabric port, connects to another FC switch - **E_Port:** Expansion port, connects to an FC switch in another fabric ###### Zoning An FC switch function that enables node ports within a fabric to be logically segmented into groups and to communicate with each other within the group. - Both node ports and switch ports can be zone members - **Benefits:** - Provides access control - Restricts RSCN traffic ###### Types of Zoning - **WWN zoning:** Uses WWN to define zones - **Port zoning:** Uses port numbers to define zones - **Mixed zoning:** Uses a combination of WWN and port numbers to define zones ###### IP SAN A SAN that uses Internet Protocol (IP) for the transport of storage traffic. It transports block I/O over an IP-based network. - **Key drivers of IP SAN are:** - Leveraging an existing IP-based network instead of building a new FC SAN infrastructure - Many robust, mature security options are available for IP networks - Many long-distance, disaster recovery (DR) solutions already leverage IP-based networks - Two primary IP SAN protocols are: iSCSI and FCIP ###### iSCSI Networking iSCSI encapsulates SCSI commands and data into IP packets that are transported over an IP-based network. - **iSCSI network components are:** - **iSCSI initiators:** Initiate data transfers to iSCSI targets, Examples: iSCSI HBA - **iSCSI targets:** Serve data to iSCSI initiators, Examples: storage system with iSCSI port (Native iSCSI), iSCSI gateway (Bridged iSCSI) ###### iSCSI Name - iSCSI name is a unique iSCSI identifier that identifies initiators and targets in an iSCSI network - **The two common types of iSCSI names are:** - **iqn:** iSCSI Qualified Name - Example: iqn.2014-02.com.example:optional_string - **eui:** Extended Unique Identifier - Example: eui.0300732A32598D26 ###### FCIP Networking FCIP is an encapsulation of FC frames into IP packets that are transported between disparate FC SANs over an IP-based network through FCIP tunnel. - **An FCIP entity (e.g. FCIP gateway) exists at either end of an FCIP tunnel:** - Encapsulates FC into IP - Transfers IP packets to remote gateway - Decapsulates FC from IP - Widely used in disaster recovery implementations ###### FCoE SAN A converged enhanced Ethernet (CEE) network that uses the FCoE protocol to transport FC data along with regular Ethernet traffic over high speed Ethernet links. FCoE encapsulates FC frames into Ethernet frames. - Transfers both compute-to-compute and FC storage traffic using the same network components - Reduces complexity of managing multiple discrete networks - Reduces the number of adapters, cables, and switches, along with power and space consumption required in a data center - Based on an enhanced Ethernet standard that ensures lossless ###### FCoE SAN Components: CNA and S/W FCoE Adapter | Component | Description | |---|---| | Converged network adapter (CNA) | Provides functionality of both NIC and FC HBA in a single device, Encapsulates FC traffic onto Ethernet frames (FCoE traffic), Consolidates both FC and regular Ethernet traffic over CEE links | | Software FCoE adapter | A software on the compute system performs FCoE processing, Supported NICs transfer both FCoE and regular Ethernet traffic | ###### FCoE SAN Components: FCoE Switch and Storage Port | Component | Description | |---|---| | FCoE switch | Contains Fibre Channel Forwarder (FCF), Ethernet Bridge, and a set of ports for FC, Ethernet, or FCoE connectivity, FCF encapsulates FC frames into Ethernet frames (FCoE frames) and decapsulates FCoE frames to FC frames | | FCoE storage port | Connects to FCoE switch, enabling end-to-end FCoE environment | #### Inter-cloud Communication - Cloud bursting - Cloud migration - Cloud federation