data is distributed across various disks/disk arrays managed by AMPs.Each data row is physically stored in data blocks in the disk arrays.To Identify the data block to which the row belongs (to be read/written), Master Index and Cylinder index are used.
Once correct AMP has been found, Master Index for that AMP is used to identify the Cylinder Index that is to be referenced.The cylinder index for that cylinder then identifies the correct data block.
Hi Rmm, Master Index,Cylinder index are use to find the rows of data. Amps uses the MI and CI to access the row of data quickly. when we try to retrieve the data PE creates Table ID for Table. Master Index knows every table Id and Row Hash and Cylinder numbers. Amp uses Table Id to find the MI. By using MI Amp finds the Cylinder Number. By using Cylinder number Amp find the Cylinder Index. In CI knows Table Id and Row ID and Row Hash values. By using CI Amp will find the row of data. The Process is AMP find the MI. Then By using MI AMP find the CI. the By using CI Amp find the row of data
Teradata uses logical address system, unlike other DMBS which uses physical address systems. This allows teradata to locate the data easilly even if the data gets reallocated/defragmentated. MI and CI are two structures of this logical address which teradata uses to locate the data on the data block.