RAID and SAN Technologies Quiz

Cloud Physical Layer Overview

• The physical layer encompasses physical compute, storage, and network resources.
• Compute systems execute software for providers and consumers.
• Storage systems hold business and application data.
• Networks connect compute systems to each other and storage systems.
• Networks link multiple data centers or clouds.

Introduction to Compute System

• A computing platform (hardware, firmware, and software) runs platform and application software.
• It runs both provider and consumer software, typically on x86-based servers or hosts.
• Compute systems are offered via shared hosting (multiple consumers share systems) or dedicated hosting (individual consumer dedicated systems).
• Providers frequently use compute virtualization and offer systems as virtual machines (VMs).

Key Components of a Compute System

• Processor: An integrated circuit (IC) that performs arithmetic, logical, and input/output operations.
• Random-Access Memory (RAM): A volatile data storage device for program execution and data use by the processor.
• Read-Only Memory (ROM): A semiconductor memory that contains boot, power management, and device-specific firmware.
• Motherboard: A printed circuit board (PCB) that houses the processor, RAM, ROM, network and I/O ports, and other integrated components like GPUs and network interface cards (NICs).
• Chipset: A group of microchips on the motherboard that manage specific functions like processor access to RAM and peripherals.

Software Deployed on Compute Systems

• Self-service portal: Enables cloud users to view and request services.
• Platform software: Provider software offered via Platform as a Service (PaaS).
• Application software: Applications offered via Software as a Service (SaaS).
• Virtualization software: Enables resource pooling and virtual resource creation.
• Cloud management software: Facilitates provider management of cloud infrastructure and services.
• Consumer software: User platform software and business applications.

Types of Compute Systems

• Tower compute system: A vertically oriented enclosure containing integrated power supply and cooling. Larger environments may require substantial expenditure.
• Rack-mounted compute system: Designed for mounting on a standardized rack, simplifying cabling and utilizing space. Administrators manage systems using consoles mounted on the rack.
• Blade compute system: An electronic circuit board assembly with core processing components within a chassis, including integrated power supply, cooling, and management. Modular design enhances density and scalability.

Introduction to Storage System

• A storage system is the repository for saving and retrieving electronic data.
• Providers offer storage capacity—often combined with compute— or as a service. Storage as a Service (StorageaaS) enables data backup and long-term retention.
• Cloud storage boasts massive scalability and rapid elasticity. Virtualization creates shared storage pools.

Types of Storage Devices

• Magnetic disk drive: Data storage on circular disks with ferromagnetic coatings. Common, fast access, large capacity.
• Solid-state (flash) drive: Semiconductor memory storage, offering low latency, low power consumption, and high throughput.
• Magnetic tape drive: Plastic film with magnetic coating. Primarily sequential access, cost-effective for long-term storage.
• Optical disc drive: Polycarbonate disks with reflective coatings. Commonly used for write-once/read-many applications (CDs, DVDs, BDs) and long-term storage.

Redundant Array of Independent Disks (RAID)

• Data is written across multiple disk drives, combined into a logical unit (RAID group).
• RAID improves performance by serving I/Os from multiple drives concurrently.
• RAID provides data protection from drive failure utilizing techniques like striping, mirroring, and parity.

RAID Techniques

• Striping: Spreads data across multiple drives to utilize them in parallel.
• Mirroring: Simultaneously stores the same data on two different drives to create copies.
• Parity: Employs mathematical operations on data strips and stores results, safeguarding data from drive failures.

Common RAID Levels

• RAID 0: Striped set with no fault tolerance.
• RAID 1: Disk mirroring.
• RAID 1+0: Nested RAID combining striping and mirroring.
• RAID 3: Striped set with parallel access and 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

• Block-level access: Storage is handled as blocks of data.
• File-level access: Data accessed through file systems on the storage.
• Object-level access: Data stored in terms of objects and their metadata

Storage System Architecture

• Storage system architectures depend on data access methods.
• Common storage options include: Block-based, File-based, Object-based, and Unified.

Types of Storage Architectures

• Block-based Storage: Enables creation and assignment of storage volumes to compute systems, allowing compute systems to treat volumes like local drives.
• File-based Storage: Dedicated servers with high performance file storage, often attached to networks. Enables filesharing and supports various operating systems.
• Object-based Storage: Stores data as objects, using non-hierarchical addressing. Contains objects with user data, associated metadata, and uniquely identified attributes.
• Unified Storage: Combines block, file, and object storage functions in a single system.

NAS Deployment Options

• Traditional NAS: Scaled by upgrading or adding components to a single system.
• Scale-out NAS: Uses multiple nodes in a cluster to offer scalable capacity and performance.

Introduction to Networking

• Networking enables data transfer and resource sharing across geographic locations.
• Cloud consumers require secure and dependable networks for accessing cloud services and resources.
• Network connectivity supports resource aggregation and service mobility across cloud data centers.
• Multiple clouds can have interconnections to enable workload distribution.

Types of Network Communication

• Compute-to-compute Communication: Connects physical compute systems via IP-based protocols. Uses Switches and Routers (common interconnecting network devices).
• Compute-to-storage Communication: Interconnects storage systems with compute systems via networks. A SAN enables and facilitates access to storage from systems.
• Fibre Channel SAN (FC SAN): Uses Fibre Channel (FC) protocol to transfer data, commands, and status between compute and storage systems; block-level access; high speeds (up to 16 Gbps).
• Internet Protocol SAN (IP SAN): Uses IP for storage traffic. Leverages existing IP networks; widely used in disaster recovery and long-distance connectivity. Uses iSCSI and FCIP (other protocols).
• Fibre Channel over Ethernet SAN (FCoE SAN): Combines FC storage traffic and Ethernet networking, streamlining management and lowering cost.
• Inter-cloud Communication: Facilitates communication between geographically disparate cloud environments.

FC SAN Components

• Network adapters: Provide physical communication interfaces. Examples: FC Host Bus Adapters (HBAs) and storage system front-end adapters.
• Cables and connectors: Primarily use optical fiber for high-speed connectivity.
• Interconnecting devices: Typically FC switches (modular designs with multiple ports for better fault tolerance).

Fabric Connect and Addressing

• Fabric connects all nodes and facilitates communication. Unique switch identifiers (IDs).
• Identifiers for network adapters (WWNN) and ports (WWPN). 24-bit FC address used for ports in a fabric.

Fabric Port Types

• Different port types (N-Port, F-Port, E-Port).

Zoning

• Zoning (FC switch feature): Enables segmentation and logical grouping of node ports within a fabric for communication and access control.

Types of Zoning

• WWN Zoning: Groups compute and storage based on their unique WWN identifiers.
• Port Zoning: Groups ports based on their physical locations on switches or their usage.
• Mixed Zoning: Hybrid approach combining aspects of WWN and port zoning.

iSCSI Networking

• iSCSI encapsulates SCSI commands in IP packets, facilitating transport over IP networks.
• iSCSI Initiators (e.g., HBA): Client devices.
• iSCSI Targets (e.g., storage systems with iSCSI): Storage devices.
• iSCSI Names are unique identifiers used for initiators and targets. iqn (iSCSI Qualified Name) and eui (Extended Unique Identifier) are common types.

FCIP Networking

• FCIP encapsulates FC frames into IP packets between disparate FC SANs over an IP network. Using IP network for transport.
• An FCIP entity (e.g., gateway) encapsulates/decapsulates at either end. Often used for disaster recovery.

FCoE SAN

• FCoE SAN combines FC storage traffic with Ethernet, reducing complexity and cost in managing multiple networks. Standardizes use of Ethernet for FC traffic for both compute and storage elements.

FCoE SAN Components

• Converged Network Adapters (CNAs) combine NIC and FC HBA functionality.
• Software FCoE adapters support both FCoE and Ethernet traffic.
• FCoE Switches provide the mechanism for routing and switching FCoE packets between nodes.
• FCoE storage ports connect storage devices to the FCoE switch.

Inter-cloud Communication

• Communication between geographically or logically diverse cloud environments and/or providers.