kiamkhai wrote:Dear TM1 Developers
PLease give me some tips the actual meaning of in-memory and how does it works for TM1?
You might want to take a look at some of the introductory sections of the manuals which cover this in general terms.
Put simply, if a user queries a TM1 cube, all of the values are stored in, and retrieved from, the server's memory. TM1 doesn't need to read the data from the server's hard disk drive. Consolidations and rule-based values are calculated "on demand" (that is, only when the user asks for them) and are also retained in memory. The only time that TM1 reads values from the hard disk drive is when the server session is starting, and the only time it writes to the disk is when you tell the server session to do a data save which stores all of the values to the disk. (Though changes to the values are normally logged in real time a transaction log file so that they can be restored should the server session crash.)
This makes TM1 much faster than applications which need to keep reading values from a hard disk.