MicroNet Technology 5 manual Glossary

C-Glossary

up a lane. The lanes are designed for multiplexing, and PCIe devices that utilize multiple lanes in this configuration are designated with an “x” multiple to denote the number of lanes they can address simultaneously. The slots are modular- multiple lane slots (such as 16x) can accommodate devices of smaller lane count (such as 1x, 4x, 8x) but usually not the other way.

PCI-XAn enhanced version of PCI version 2.2. It supports one PCI slot per bus when running at 133 MHz, two slots when running at 100 MHz and four slots when running at 66 MHz. It is intended to provide throughputs in excess of 1 GB/sec using a 64-bit wide 133 MHz implementation.

Performance an important criterion on which a customer judges a RAID controller. There are a number of popular benchmarking utilities that are available to measure the I/O performance of a controller. Some of these utilities simulate specific real-life applications and provide the user a score indicating the controller’s overall performance in that niche. Others allow the user to specify tests with specific I/O characteristics and generate throughput numbers corresponding to each specification. The nature of the tests a user may wish to conduct on a controller depends on the application space in which that controller is anticipated to be deployed.

Physical Drive A single tangible drive is referred to as a physical drive.

Primary Storage Main memory i.e., RAM is frequently referred to as primary storage.

RAID Abbreviation of Redundant array of independent disks. It is a set of disk array architectures that provides fault-tolerance and improved performance.

RAID Level There are a number of RAID formats that are widely used. Some of the well- known uni-level types are RAID 0, RAID 1, RAID 3, RAID 5 and RAID 6. The prevalent complex types are RAID 10 and RAID 50. ,

RAID 0 or simple striping. Data is distributed across two or more disks. No data redundancy is provided. Advantage: Striping can improve the I/O throughput by allowing concurrent I/O operations to be performed on multiple disks comprising the RAID 0 array. Disadvantage: this RAID type does not provide any data redundancy, and a single disk failure would damage the entire volume.

RAID 1 An array that uses a single pair of disks. Both disks in the pair contain the same data It provides the best data protection but can’t improve system performance. And storage space for the same data capacity should be double than in general cases. Hence storage cost doubles. The capacity of RAID 1 will be the size of the smaller HDD, so we suggest you connect HDDs of the same sizes to save HDD space. Advantage: RAID 1 ensures that if one of the disks fails, its contents can be retrieved from the duplicate disk. Furthermore, a RAID 1 array can also improve the throughput of read operations by allowing separate reads to be performed concurrently on the two disks.

RAIDBank5 Owner’s Manual 42