Seek Time & Memory Hierarchy

¬ SEEK TIME

The disk address is specified in terms of the surface number, track or cylinder number, and the sector number. The read/write heads need to be first positioned on the track on which the data are to be recorded or from which data needs to be read.
Information is always written from the beginning of a sector and can be read only from the beginning of the desired track. Thus, the disk access time depends on the (i) Seek Time & (ii) Latency Time

Seek time: The seek time measures the amount of time required for the read/write heads to move between tracks over the surfaces of the platters. Seek time is one of the most commonly discussed metrics for hard disks, and it is one of the most important positioning performance specifications. The seek time would vary depending upon the position of the access arms at the time the read/write command was received, i.e., if the access arm was positioned on the outermost track and the current read operation required it to be positioned on the 5th track, then the time taken to position the access arm on track 5 is the seek time. It is obvious from this example that moving from outermost to the innermost track or vice versa would result in the maximum seek time. The average seek time in most systems is 10-100 milliseconds.

Typical full seek time is 8 ms. Average seek time (from one random position to another random position) is 1/3 full seek time.

While seek time is usually given as a single number, in fact there are three different seek time specifications you should examine for a drive, as they represent the drive's performance when doing different types of seeks:

(i) Average: As discussed, this is meant to represent an average seek time from one random track (cylinder) to any other. This is the most common seek time metric, and is usually 8 to 10 ms, though older drives had much higher numbers, and top-of-the-line SCSI drives are now down to as low as 4 ms!

(ii) Track-to-Track: This is the amount of time that is required to seek between adjacent tracks. This is similar in concept (but not exactly the same as) the track switch time and is usually around 1 ms.

(iii) Full Stroke: This number is the amount of time to seek the entire width of the disk, from the innermost track to the outermost. This is of course the largest number, typically being in the 15 to 20 ms range. In some ways, combining this number with the average seek time represents the way the drive will behave when it is close to being full.
MEMORY HIERARCHY:
Memory in the computer system is required for storage and subsequent retrieval of instructions and data. Fast access to instructions and data stored in memory is essential in order to meet the throughput requirements of the processor. Memory system can be considered to consist of three groups of memories:
Ø Internal processor memory consists of a small set of high-speed registers
Ø Primary storage or main memory is a large memory, which is fast, but not as fast as internal processor memory.
Ø Secondary storage or auxiliary memory or backing store is much larger in size than the main memory but is slower than the main memory.

A balanced provision for different kinds of memory – CD-ROM, hard-disc, DVD, main semiconductor memory (DRAM), cache memory (SRAM, SSRAM or SDRAM), processor registers – is necessary in order to meet the performance/cost requirements of a computer system.

650 MB – 1 TB, 100 ms - ??
The total memory capacity of the computer can therefore be visualised as being a hierarchy of components consisting of all storage devices employed in a computer system from the slow but high-capacity auxiliary memory to a relatively faster main memory, to an even smaller and faster cache memory accessible to the high-speed processing logic. Thus, as one goes up in the hierarchy, the following conditions occur:
(i) Decreasing cost per bit (ii) Increasing capacity (iii)Increasing access time
(iv) Decreasing frequency of access of the memory by the processor

CPU Registers:
CPU contains several registers for storing data and instructions. But these can store only few bytes. Internal processor memory consists of a small set of high-speed registers that are internal to a processor and are used as temporary locations where actual processing is done.
Main Memory:
Primary storage or main memory is a large memory, which is fast, but not as fast as internal processor memory. This memory communicates directly with the CPU and is mainly based on integrated circuits.

Cache Memory:
Another kind of memory that is increasingly being used in modem computers is called the cache memory. Cache memories are small, fast memories placed between the CPU and the main memory. Due to it’s expensiveness of this memory it is used in small quantities only. The memory read or writes operation is first checked with the cache and if the desired data is available in the cache it is used by the CPU directly.

Secondary Storage Memory:
Devices that provide backup storage (like magnetic tapes and disks) are called secondary storage or auxiliary memory. It typically stores all the system programs and other software.
For e.g. CD-ROM, hard-disc, DVD