The Object Storage structure
Our Object Storage infrastructure is based on the latest solutions within a storage environment for archiving petabyte (1.000.000 GB) of non-structured data. It is based on a technology which guarantees high storage power, simplicity in operations and total control of data.
The design of the infrastructure consists in a distributed intelligence system which does not present any SPOFs (Single Point Of Failure).
The resilience and the self-healing capacity are crucial features of the Object Storage system. Updates in the technology or on the hardware do not require any data migration or downtime. The parallel architecture ensures high performance levels for large volumes of traffic and requests per second.
The structure of our Storage has been created in such a way to guarantee self-healing, by creating multiple copies of the data on the Servers.
When a file is uploaded in the storage space, the system automatically creates two copies of such file and saves the copies on randomly selected disks on different Servers, making sure that 3 copies of the data is always available in the system. In case of brakeage or damage of a storage disk, the system will automatically activate the self-healing system, recreating a copy of the lost set of data on a third disk. All the files uploaded in the storage space, are not replicated on the same disks, each copy is made on a different disk of a different Server. This way a safer storage network is created.