MAC OS X Archives - Latest News from Backup Technology

Server backup can be tricky. The backup cloud must support all kinds of operating systems and applications. It should be remembered that servers operate in a high traffic environments. They are always connected to the remote backup server in the cloud via the Internet and continuously service the data requirements of the end user. The server platform can include all flavours of Windows, Linux, Unix, Mac OS X, AS-400, Solaris or all kinds of applications such as Exchange/ Outlook, Lotus Notes/Domino, Groupwise, SQL, MySQL, PostgreSQL, Oracle, IBM DB2, Sharepoint or more.

Agentless server backup is preferable. The agentless software requires only a single instance installation of the agent on a network. All servers connected to the network can then be identified and the agent can be configured to accept backup requests from the servers. The entire backup process can be managed and controlled from a single console.

IT Administrators can schedule backups and decide on the granularity of the backup. For instance the IT Administrator can decide that Windows and Linux systems must be backed up continuously and other systems may be backed up as per a fixed schedule. A single backup can be the backup of all data in a server, a share or volume or directory. The backup may also be of a single file or registry. Backups can also be scheduled to start and stop within a specific backup window.

Multiple servers can be backed up in parallel. The multi threaded backup application ensures that the software can process large volumes of data and support concurrent backup and restore requests. The only limiting factor would be the amount of bandwidth that is available with your organization.

The data being backed up to the server can be compressed or deduplicated using compression algorithms that have been tested and proven. The lossless algorithms can reduce the size of the raw data being transmitted from the server and common file elimination protocols can ensure that the same file is not transmitted twice. Versions of files may be stored with appropriate version numbers for ease of discovery. Retention policies can be specified to ensure that only relevant and active files continue to be stored in expensive storage media and other files are relegated to cheaper storage systems to save on costs.

Backup Technology’s Backup for Servers is high performance software that is designed to operate efficiently in high traffic environments with many of the features that have been discussed above. For more information please visit, http://www.backup-technology.com

RAID (Redundant Array of Inexpensive Disks) is based on virtualization technology to store data. In RAID system, many drives are used to make one logical unit. There is a variety of RAID levels and each of them is optimized for particular condition. As a matter of fact, RAID levels are not standardized through industry groups, and therefore, companies often have distinctive numbers and accomplishments.

Commonly used RAID levels are:

RAID 0
RAID 5
RAID 6
RAID 10

Software to carry on RAID functionality is usually positioned on controller card. In Mac OS X and Windows Server 2012, there are software for RAID functionality.

RAID Level 0

In this level, data is divided into blocks. If multiples disks are used, RAID 0 provides greater I/O performance. This level gives satisfactory performance in write and read operations. Storage virtualization is easy to implement. The drawback of RAID 0 is the inability for fault-tolerance. In case, one of the drives get failed, available data in RAID 0 will be lost. Consequently, it cannot be used for critical systems. It is perfect for non-critical data storage where the basic aim is to write or read data at highest speed; such as video editing station. For critical system, it is better to use cloud backup technology so that important data remain backed up to the cloud.

RAID Level 5

It is known as secure level and needs three or more drives. Across the drives, data blocks are banded. With the help of parity data, the computer calculates the other blocks of data even if data does not exist any longer. In other words, RAID 5 has the capability to survive. If one drive fails and replaced, data can be accessed easily. RAID 5 offers fast transactions of read data; however transactions of write data is a bit slower.

Drawback of RAID 5 is that failure of one drive influences on others as well. When one drive fails, data restoring process needs one day or more depending on the controller speed and load. When another disk fails during data restoring process, it is impossible to retrieve the lost data.

RAID Level 6

It is similar to RAID 5 however two drives are used for data parity. Consequently, RAID 6 needs minimum 4 drives and bears the loss of two drives at the same time. It is true that failure of two drives simultaneously is rare. RAID 6 has fast transaction of read data. The drawback of this level is slow speed of write data just because of parity. When read operation is selected, it reads the data from drive-1 parity and then drive-2 parity. Similarly, write operation means to write parity one and then write parity two; which is time consuming.

RAID Level 10

It is known as uncomplicated level for calculation. It has mirror sets and each of them writes data two times to develop mirroring. As a result, write performance gets slow. RAID 10 is an easy and fast process for rebuilding data to new drive from existing mirrors. In this level of virtualization, maximum storage capacity is utilized for mirroring and makes it expensive method.

Conclusion

It is the basic needs of enterprises to keep their important data secure. Though there are various performance levels of RAID, but each level has some drawbacks. In RAID array, all drives have same age factor so failure of one drive can lead to other. To keep data safe from natural disasters, drive failure and/or hacking, the best option is a cloud backup service.