0 votes 0 votes Can someone tell me what is the actual meaning of cache access / memory access time ? Is it the time to fetch a word/byte from cache or is it the time to search the cache or is it the sum of both? If it's the time to fetch a byte then why do we add this time in hierarchical access ? It's not the searching time right? CO and Architecture cache-memory multilevel-cache + – Xylene asked Jul 6, 2017 • edited Jul 6, 2017 by Xylene Xylene 484 views answer comment Share Follow See all 0 reply Please log in or register to add a comment.
0 votes 0 votes Its the sum of both A_i_$_h answered Jul 7, 2017 A_i_$_h comment Share Follow See all 4 Comments See all 4 4 Comments reply Bikram commented Jul 8, 2017 reply Follow Share Yes, search time is implicit in access time. When we mention access time it includes search time unless specified explicitly ... @xylene In general access time in hardware, storage means the average time interval between a storage peripheral (usually a disk drive or semiconductor memory) receiving a request to read or write a certain location and returning the value read or completing the write. 0 votes 0 votes Bikram commented Jul 8, 2017 reply Follow Share @Xylene Now the question is why we add this time in hierarchical access ? That's because in case of hierarchical access when there is a miss in level 1 then only we move to level 2 . So there some time needed to reach in level 1 and some time needed to search if that required block is found in level 1 or not ! if not found we have to go to level 2 and search again .. so this ( reach to level 1 + first time search in level 1 ) time is need to add in case of hierarchical access . If this is simultaneous access then this time is not needed because in case of simultaneous we can reach level 1 and level 2 at the same time . Hope i am clear to you. 0 votes 0 votes Xylene commented Jul 8, 2017 reply Follow Share Sir, suppose there is a miss in L2 cache and a hit in main memory then can we access the word directly from main memory or should be transfer it to cache and then access? 0 votes 0 votes Bikram commented Jul 8, 2017 reply Follow Share @xylene If there is a miss in L2 cache and hit in Main memory then we bring back the word in cache as it requested by CPU and then access. We don't access that word directly from memory , it should be bring into cache because cache is close to cpu .memory access is take large time so for large data block each memory access takes long time ..hence we need that data block in cache so that time to access is less . After access if it again need in near future then keep in cache , if not require then keep in memory . 0 votes 0 votes Please log in or register to add a comment.
0 votes 0 votes it is the time required to fetch a byte from cache . search time generally mention explicitly . if not mention we ignore search time. we can say search time is implicit in access time , it is there but not mention. Bikram answered Jul 7, 2017 • edited Jul 8, 2017 by Bikram Bikram comment Share Follow See 1 comment See all 1 1 comment reply Bikram commented Jul 7, 2017 reply Follow Share Also read what Memory access time means , it is how long it takes for a character in memory to be transferred to or from the CPU , www.pcmag.com/encyclopedia/term/37400/access-time so it is same as fetching a byte from memory . 0 votes 0 votes Please log in or register to add a comment.