What It All Means
A glossary to make sense of server jargon
By Staff -- Broadcasting & Cable, 5/17/2004
Confused by some of the terms associated with video servers? You're not alone. Here's a quick primer on what some of the more obscure phrases mean.
Cache: A memory cache uses high-speed static memory, or Static Random Access Memory (SRAM), to store frequently accessed video files or other content and applications. This speeds up operations that otherwise would need to access content on Dynamic Random Access Memory (DRAM), which has slower access speeds. Some memory caches are built into the architecture of computer microprocessors. Another approach is disk-caching, with content stored in a buffer in the server's main memory.
Fibre Channel:A data-transfer architecture designed for mass storage and other devices that need to transfer very high-bandwidth content, such as video and audio files. It uses fiber optic cables to connect devices and supports transfer rates of 100 megabits per second (100 Mbps) in two directions.
Gigabit Ethernet:Another form of data transfer, it supports rates of 1 gigabit (or 1,000 megabits) per second (1 Gbps).
Gigabyte:(G or GB) Equal to 1,024 megabytes.
Material Exchange Format: (MXF) A metadata wrapper that is placed around a video or audio file and allows it to be transferred from one video server to another, regardless of the format a server uses.
Megabyte:(MB) Equal to 1,048,576 bytes of information.
Metadata:Description information related to a given video or audio file. It typically includes when the content was recorded, who was involved with the material's acquisition, where it was recorded, and possible keywords describing the content. Think of it as CliffsNotes for a file.
Networked Attached Storage architecture: In this storage approach, a network-attached storage (NAS) device is dedicated to nothing more than file sharing. The server handles all the processing of data but does not actually store any content itself. As a result, more hard-disk storage space can be added to a network without existing servers having to be shut down. Not to be confused with SAN (see below).
RAID array architecture:Redundant Array of Independent (or Inexpensive) Disks, uses two or more drives to provide backup and improved performance. There are four different grades of RAID. Level O spreads data across multiple disks but without redundancy; level 1 provides disk mirroring for redundancy; level 3 is the same as level O but has one disk dedicated to error correction; and level 5 provides excellent performance and good fault tolerance.
Storage Area Network architecture:(SAN) A high-speed subnetwork of shared storage devices—in this case, video servers. The advantage of a SAN is that stored data does not reside directly on any of the servers, leaving more network capacity available to the end user and making it possible to have more server power dedicated to applications. The servers typically reside on a local area network (LAN) or wide area network (WAN).
Terabyte:(TB) Equal to 1,024 gigabytes.
