Cloud storage is often mis-advertised as a specific product, however it can generally be considered to be the storage of data online hosted by your cloud vendor of choice. Each vendor has their own mechanisms by which to store your data, usually across multiple distributed and connected resources, and can present it to you in the model or protocol to satisfy your requirement. SolNode supports the online storage models listed below.
Each storage model will generally have a specific use-case it specialises in and further options around the protocol(s) in which it can be presented to your application. If you are uncertain as to which model, protocols or performance offerings may be best to suit your use-case, feel free to reach out and we will be happy to assist you to better understand the options available to satisfy your requirements.
Most computer users will already be familiar with utilising file storage, as it is usually the place where you save all of your important documents and all kinds of media. The mechanism behind your ‘shared drive’ or ‘My Documents’ is most likely utilising a flavour of file storage. The logical method behind file storage is often referred to as hierarchical storage as it emulates the way files are stored in a filing cabinet where traditionally a person collates any related pages of documents into folders which are then named and stored with a logical naming-convention/mechanism to ensure ease of retrieval at a later date.
Applying this analogy to digital storage, files/data is given a name, then placed in named folders/directories, or even nested under more folders/directories to form a set path, stored within a logical unit called a file-system. The net result is files organised into a hierarchy, with folders/directories and sub-folders/sub-directories. A user wanting to access the file data only needs to follow the hierarchical path to retrieve the data from the target folder/directory. A limited set of metadata is associated with each file, including attributes like the file name, created date and also the last date it was modified.
File data lends itself to being shared among users on a network which can be centralised and open to a wide audience across a business by implementing a Network Attached Storage (NAS) device. End users can then access their file data, and other shared file data, from remote locations, enabling modern flexible working arrangements and cross-business collaboration.
– Deployment in file-based Network Attached Storage (NAS) system
– Securely sharing files among users across a network (local or distributed)
– Repositories for storing documents, content and media
– Archival data stores
– CIFS/SMB and NFS related requirements.
Block storage is the mechanism that underpins your file storage, in that, the filesystem in which your files are stored in sits on top of the block storage. Block storage is made up of many physical chunks of storage media, usually referred to as a hard-disk, SSD or a LUN, upon which your filesystem is created across and your files stored. Every server, personal computer, portable device and even USB storage device, uses a combination of block and file storage together to achieve their use-case. Some server applications require the usage of a specific filesystem or specific parameters which will drive your use-case for block storage.
In the cloud and enterprise, block storage is commonly provided over Storage Area Networks (SAN) which would traditionally provide a host access over fibre-channel protocols, but in recent years, has also moved in the direction of ethernet using iSCSI. Utilising these specialised networks, a range of storage arrays are ready to provide a LUN/raw-volume to be presented to the your server. The operating system can then format the hard drive with the appropriate file system based on the application requirements. Block storage is not generally suitable itself for direct access across a wide range of clients, rather used by one or several clustered servers that will utilise the storage to run critical applications.
– Ideal for use with applications that require consistent I/O performance and low-latency connectivity like databases.
– Running mission-critical applications with high disk performance requirements like Mail, ERP and collaborative platforms.
– Applications that require direct block access and are not compatible for use with a filesystem.
– Clustered applications that require traditional block-level shared storage.
Object storage manages data as objects instead of files or blocks, storing them in a flat address space, called a storage pool. Each object contains the data itself along with a variable amount of associated metadata and a globally unique identifier called an Object ID. Object Storage also has the ability to create flexible metadata as defined by the user which could include the type of application the data is associated with, its privacy classification, what level of protection or backup to apply to the data and when and where to replicate the data. With extending the metadata comes the opportunities to perform further analysis on the use and function of data within the storage pool. Data objects are retrieved from the storage pool when an application presents the Object ID to the front end interface or Rest API. Object Storage is also usually accessible over a web-based channel, or HTTP(s), making it truly ‘cloud-native’.
– Storage of large unstructured data like video, images and music files.
– Disaster recovery, backup and archiving solutions replacing old tape based solutions.
– Big Data storage for BI analytics around financial, healthcare, scientific industries.
– “S3” compatible storage applications (eg. CloudBerry)