A B C D E F


G H I J


K L


M N O P


Q R S T U V


W X Y Z


Controller Cache The cache is an amount of volatile memory installed onto a storage controller. It speeds up disk access by temporarily storing frequently accessed data in fast memory rather than on disk, which is much slower. Generally, the more cache memory available, the greater the efficiency when reading and writing data to disk or array. For a SAN storage solution, it is highly advisable to have a separate battery backup for the controller cache, as this can prevent data being lost in the event of total power failure.


Copy-on-Write A system that creates a snapshot of any changes made to the data. The snapshot is stored (on a tape or a disk, or in a SAN) as a set of reference points, and can be used to recover the original data in the event of disk error or other processes that causes data loss or corruption.


CPU Cache High speed memory used by the CPU to store the most frequently used instructions. This enables the CPU to bypass the system bus and boost overall system speed. A cache controller checks whether data is already stored in the cache. Level 1 cache is stored on the CPU, while Level 2 cache is held in separate chip next to the CPU. Some CPUs have both Level 1 and Level 2 cache built into them and this is known as Level 3 cache.


Crash Consistent If a server crashes or is abruptly switched off, any applications still running will not exit gracefully. This leaves the data files in an open state and can result in them being difficult to re-open or even corrupted.


DAS (Direct Attached Storage) Storage system in which the storage device (such as a RAID array) is connected directly to a server or workstation, with no network device (such as a switch) between them. DAS can offer a low-cost storage solution, but if your school has a large IT infrastructure or you want a more flexible storage solution, then a networked storage system like a SAN is a better option.


DAT (Digital Audio Tape) An audio tape system adapted for data storage and backup. DAT evolved into the Digital Data Storage (DDS) format and DDS tapes can store up to 160GB (native). There are plans to increase native capacity to more than 300GB.


Deduplication An operation often performed during the backup process, whereby duplicated data is removed. Data deduplication can operate at the file, block and bit level. This process can save your school storage space, time and money. See expanded explanation.


Degraded A RAID array is considered to be degraded when a proportion of the disks in the array have failed, and data has to be reconstructed from the parity information. This can reduce performance, and depending on the amount of redundancy built into the array, even make the data unsafe.


DR (Disaster Recovery) A process or plan which enables an organisation to recover as quickly as possible from a system breakdown, such as a server


A SPOTLIGHT ON… SAN Connection Technologies


SANs and servers are connected via a network of high-speed host bus adapters and switches. There are basically two options when it comes to choosing which data transport technology to deploy: Fibre Channel or iSCSI.


Fibre Channel Fibre Channel is a set of data transport standards designed for high-speed data transfer between workstations, supercomputers, displays and storage devices like SANs. Its name is slightly misleading, as Fibre Channel can utilise fibre optic cable, coaxial cable or even twisted pair copper wire. Fibre Channel can be used peer-to-peer, switched or in an arbitrated loop, which can connect up to 126 devices. Fibre Channel offers data speeds of 1Gb/s, 2Gb/s, 4Gb/s and 8Gb/s, with 16Gb/s speed on the development roadmap. Fibre Channel supports a number of protocols including SCSI and IP. Fibre Channel has traditionally been used by large enterprises, and historically, an expensive technology to deploy. Despite its advantages, Fibre Channel exceeds the requirements of the vast majority of school environments, and for most schools, iSCSI is a more viable (and affordable) option.


iSCSI iSCSI is based on standard Ethernet switches and components, and so the cost of deployment is much lower than for Fibre Channel. iSCSI is also built on the TCP/IP protocol and so is ideal for LANs, WANs and the internet. Because iSCSI utilises ubiquitous network technologies, it is cheaper, less complex to install and easier to manage than Fibre Channel. Performance-wise, your school is unlikely to notice any difference between Fibre Channel and iSCSI. In fact, it’s estimated that a school would need to have at least 70 hard disk drives in its SAN array before noticing any difference in performance.


Multipath I/O (Multipath Input/Output) Whatever SAN connection technology you choose, it’s important that a high degree of redundancy is built into the fabric. This ensures that there is no single path or connection between server and SAN, because any component failure along the path could break the connection. A way to protect against this is to use a Multipath I/O configuration. This uses multiple switches, controllers and other redundant components to ensure high availability.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13