Storage systems have become a single, complex IT domain, and can have different meanings for different people. Also, how can we define these systems? Storage systems are the hardware that stores the data.
For example, it may be the server of a small business with an office often users or less. The storage system would then consist of hard disks inside this server, where are the information of the users. In a large enterprise context, storage systems can be large SAN enclosures that contain hard drives and have space that has been slotted and boxed in different ways to provide redundancy and performance.
The technology of constantly evolving storage systems
Today, storage technology encompasses all kinds of storage media. These may include Write Once Read Many systems – single-write, multi-read (WORM), tape library systems or tape library systems, and virtual tape library systems. In recent years, SAN and NAS have brought remarkable reliability. What is the difference between the two systems?
NAS (network storage) units are stand-alone units with their own operating system, file system, and which manage the connected hard disks. These units come in various forms and sizes to meet most needs and operate as file servers.
For a period of time, large-scale storage systems were not accessible to small businesses. Serial ATA (SATA) -based SAN-based SANs have proven to be a cost-effective way to provide high-capacity storage space. These network units also offer virtual tape library backup systems, literally, RAID arrays that act as magnetic tape recorders to completely erase the tape storage media.
Other storage technologies such as iSCSI, DAS (Direct Connection Storage), Proximity Storage (data that is attached to removable media), and CAS (Addressable Content Storage) are all methods that provide availability. Architects in data storage know that having a simple backup is not enough.
At present, in intense information contexts, backups, whether incremental and scheduled each evening or weekly and complete, are obsolete in a few hours, or even within a few minutes of their creation.
In large data warehouses, backing up constantly changing data is not an option. The only solution for these huge structures is to have mirrored storage systems, ie identical servers with exactly the same storage space.
3 things to consider when choosing a system
A careful analysis of the operational environment is necessary. Most would say that having absolutely no failures is the best environment, which is true for both users and administrators. The harsh reality is that data loss disasters occur every day despite the implementation of risk mitigation policies and programs.
When considering your storage requirements, consider:
Recovery Time: What is the maximum time period your customer has before the data is available again? In other words, how long can you or your client survive without the data? This will establish the performance requirements of the equipment,
The quality of the recovered data: Is the recovery of the original data required or will old backed up data be sufficient? This concerns the backup scheme that is used. If the data on your storage system is changing rapidly, the original data will be the most useful,
The amount of data to archive: The recovery of large amounts of data will take time to transit through a network. . In DAS (Direct Connection Storage) configurations, the recovery time will depend on the equipment as well as the I / O performance of the hardware. Unique Data Protection Schemes Storage System Manufacturers Seek Unique Ways to Process Quantities data while still being able to provide disaster redundancy. Some large SANs incorporate a complex device organization at the block level, essentially creating a low-volume file system from the RAID perspective. Other SANs have an internal block-level transaction logging system so that the SAN control processor tracks all block-level writes to individual disks. Using this transaction log, the SAN can recover unexpected power outages or outages.
How can recoverability be improved?
Some computer specialists in the field of storage systems propose to add more intelligence to the control board of the RAID array so that it is “aware of file systems”. This technology would provide more recoverability in case of disaster, the goal being to give the storage network more self-healing capabilities.
Other ideas in this direction suggest having a storage pool through which multiple computers can access information without being dependent on the file system of a particular system. In organizations with multiple hardware and systems platforms, a transparent file system will provide access to data regardless of the system that wrote the data.
Other computer scientists have a very different approach to the redundancy of the storage network. The RAID concept is used on a large number of systems, however, computer scientists and engineers are looking for new ways to provide better data protection in the event of a failure. The goals that drive this type of RAID development are data protection and redundancy without sacrificing performance.
With this information about the technologies used in the architectures of storage systems, we come to the end of the first part of this article. Next week, we will discuss the issue of how to prevent system failures and what to do when you are struck by data loss. So see you soon…