Report IOps or MBps, you should always keep I/O latency in the picture. Generally speaking, you should pay more attention to IOpsĪnd I/O latency when it comes to small-sized I/O requests, and be moreĬoncerned with MBps when dealing with large-sized I/O requests. Previously, a larger I/O request takes longer to complete. Path, it is in reserve proportion to the size of the I/O request. I/O latency-also known as I/O response time-measures howįast an I/O request can be processed by the disk I/O subsystem. Note that for I/O requests of a fixed block size, MBps is simply I/Os give you better throughput because they incur less disk seek time penalty than That is, the larger the I/O requests, the higher the MBps. For a given I/O path, MBps is in direct proportion to the size of the So, the bigger the pipeline isĪnd the more pressure it can handle, the more megabytes of data it can push Pipeline is and how much pressure it can sustain. If you view the I/O path as a pipeline, MBps measures how big the MBps measures how much data can be pumped through the disk To process a 256KB I/O request than it does an 8KB request. Reverse proportion to the size of the I/O requests. For a given disk I/O path, this metric is generally in IOps measures how many I/O requests can be satisfied by theĭisk I/O path in a second. In general, you can use three key metrics to quantify the To address the question of which is faster, you must beĬlear about what you mean by "faster". To simplify discussions in the rest of this article, a driveįrom a directly-attached storage will be referred to as a DAS drive, andĪ drive presented from a SAN as a SAN drive. But it is the most commonly deployed type in the enterprise, The second point is that switch-based Fibre-Channel SAN is Significantly alter the performance characteristics of a drive presented from the Every single component on this I/O path has the potential to
#APPLE XSAN STORAGE SOFTWARE#
There are layers of software on this path, including various drivers, firmware,Īnd APIs. Typically include switches, inter-switch links, front-side adapters, disk arrayĬache and processors, disk controllers, and disk drive media. Part of the SAN starts from the host bus adapters (HBA) all the way to theĮventual disk media on the disk array.
Switch-based Fibre-Channel SAN as an example, the I/O path that is considered Two points are worth highlighting further.įirst, a SAN is not a monolithic entity. This article assumes this most common view. Rather, they view the fabric and theĭisk arrays - practically everything behind the drive and beyond the host serverĬomprising the SAN. To most other folks, however, the distinction between the switched fabricĪnd the disk arrays means very little. To the storage professionals, it could narrowly mean the switched fabricīetween the host servers and the sophisticated disk arrays that actually storeĭata.
A Storage Area Network (SAN) may mean two different things.
0 kommentar(er)
