Distributed Storage Model - Concept and Principles
A Distributed Storage model is a "multi-node" architecture that contains a set of multiple physical systems where storage capacity and computing power are grouped together to behave as one single storage system (cluster).
Generally, each node in the cluster is rack-mountable enterprise appliance containing memory, CPU, networking, Ethernet or low-latency InfiniBand interconnects, disk controllers and storage media.
In other words, this is equivalent to storing data on multitude of servers, which behave as one storage system although data is distributed between these servers.
BMC Discovery Modelling
A new Distributed Storage Model has been created to show this grouping. Each physical node is represented as a storage system and a cluster is represented as storage system group.The Distributed Storage systems are modelled and mapped as follows:
Mapping of physical components to BMC Discovery node
From a BMC Discovery perspective, the Distributed Storage system components are represented as Physical and Logical components.
Physical components represent the hardware components of a physical system such as Disk Drives, Chassis, Fiber ports, Network cards, Processors, etc.
Physical Components | BMC Discovery Node |
---|---|
Disk Drive | DiskDrive |
Chassis | HardwareContainer |
Fiber port | FibreChannelPort |
Processor | StorageProcessor |
Network card | NetworkInterface |
IP address of the Physical System | IPAddress (IPv4/IPv6) |
Example:
Mapping of the Logical components to BMC Discovery node
Logical components represent the storage cluster that groups all the physical systems as wells as the logical components such as Storage Pools, File Systems and Storage Volumes.
Logical Components | BMC Discovery Node |
---|---|
Storage Cluster | StorageSystemGroup |
Pool | StoragePool |
Volume / LUN | StorageVolume |
File System (CIFS / NFS) | FileSystem |
IP address of the Cluster | IPAddress (IPv4/IPv6) |
Example:
The following patterns are based on the Distributed Storage Model starting with the May 2019 TKU update:
Comments
Log in or register to comment.