Matches in DBpedia 2014 for { <http://dbpedia.org/resource/Independent_Reference_Model> ?p ?o. }
Showing items 1 to 12 of
12
with 100 items per page.
- Independent_Reference_Model abstract "The Independent Reference Model (I.R.M) is a conceptual model used in the analysis of storage system: disk drives, caches, etc.The motivation for coming up with this model (and others like it) is to compensate for the lack of "traces" in such storage devices.A "trace" is simply a logging of a set of data regarding the performance of the storage device, focusing on the I/O requests: how many read/write operations, the size a each request, the exact address (LUN-wise), and a time-stamp.Accurate, valid and detailed traces of actual storage systems are very difficult to get for the purpose of academic analysis (for reasons that are beyond the scope of this article), and that is why such models are a necessity.Usually, the data that is available is much coarser and low in quality, in comparison to a full trace: For example, the data might record for every unit of time T, the number of I/Os that took place at each LUN (or track), along with the total hit/miss ratios.For example: For a disk drive with 4 tracks, A, B, C and D and after 15 minutes of work, the I/O requests were the following: 7600, 20, 50, 6000 for A, B, C and D respectively.It is easy to see why this data is insufficient to determine the actual workload:Consider a second even more simple example: Two tracks, A and B, which each have 1,000 I/Os during 15 minutes.To answer the simple question, "How hard did the disk work during those 15 minutes?" then consider these two following scenarios:(I) The disk first received and responded to all 1,000 I/O requests at track A, and later, all 1,000 I/Os of track B.(II) The disk received and responded to an I/O request from a different track interchangeably: First at A, then at B, then A again, alternating A/B up to 1000 times.It is easy to see that, at each of these scenarios, the amount of work done by the disks is very different (in the first scenario, the case being that the disk did a minimal amount work, not having to travel between tracks more than once, and in the second scenario, a maximal amount of work).".
- Independent_Reference_Model wikiPageID "25112132".
- Independent_Reference_Model wikiPageRevisionID "520933021".
- Independent_Reference_Model hasPhotoCollection Independent_Reference_Model.
- Independent_Reference_Model subject Category:Information_storage.
- Independent_Reference_Model comment "The Independent Reference Model (I.R.M) is a conceptual model used in the analysis of storage system: disk drives, caches, etc.The motivation for coming up with this model (and others like it) is to compensate for the lack of "traces" in such storage devices.A "trace" is simply a logging of a set of data regarding the performance of the storage device, focusing on the I/O requests: how many read/write operations, the size a each request, the exact address (LUN-wise), and a time-stamp.Accurate, valid and detailed traces of actual storage systems are very difficult to get for the purpose of academic analysis (for reasons that are beyond the scope of this article), and that is why such models are a necessity.Usually, the data that is available is much coarser and low in quality, in comparison to a full trace: For example, the data might record for every unit of time T, the number of I/Os that took place at each LUN (or track), along with the total hit/miss ratios.For example: For a disk drive with 4 tracks, A, B, C and D and after 15 minutes of work, the I/O requests were the following: 7600, 20, 50, 6000 for A, B, C and D respectively.It is easy to see why this data is insufficient to determine the actual workload:Consider a second even more simple example: Two tracks, A and B, which each have 1,000 I/Os during 15 minutes.To answer the simple question, "How hard did the disk work during those 15 minutes?" then consider these two following scenarios:(I) The disk first received and responded to all 1,000 I/O requests at track A, and later, all 1,000 I/Os of track B.(II) The disk received and responded to an I/O request from a different track interchangeably: First at A, then at B, then A again, alternating A/B up to 1000 times.It is easy to see that, at each of these scenarios, the amount of work done by the disks is very different (in the first scenario, the case being that the disk did a minimal amount work, not having to travel between tracks more than once, and in the second scenario, a maximal amount of work).".
- Independent_Reference_Model label "Independent Reference Model".
- Independent_Reference_Model sameAs m.09gg088.
- Independent_Reference_Model sameAs Q6017092.
- Independent_Reference_Model sameAs Q6017092.
- Independent_Reference_Model wasDerivedFrom Independent_Reference_Model?oldid=520933021.
- Independent_Reference_Model isPrimaryTopicOf Independent_Reference_Model.