138k views
2 votes
Assume that block size is always a power of 2 in bytes and that the maximum block size is 128 KB (131,072 bytes). The disk has 134,217,728 (227) bytes per track.

1. If the median file size is 10,450 bytes on a 10,000 RPM drive with a mean seek time of 5 msec, what block size gives the optimal space efficiency, and what is the maximum raw disk throughput?
2. If the median file size is several Megabytes in size, then given the parameters listed above, what block size give the optimal space efficiency, and what is the maximum raw disk throughput?

1 Answer

7 votes

Final answer:

To determine the optimal block size and maximum raw disk throughput, consider the median file size and the maximum block size. If the median file size is 10,450 bytes, a block size of 8 KB may be optimal. The maximum raw disk throughput cannot be calculated without the data transfer rate. If the median file size is several Megabytes, a larger block size, such as 128 KB, would be more space efficient.

Step-by-step explanation:

To determine the block size that gives the optimal space efficiency, we need to consider the median file size and the maximum block size. The optimal block size is one that minimizes internal fragmentation, meaning it closely matches the file size without wasting space. In this case, the median file size is 10,450 bytes, so a block size that is close to this value would be optimal. Since the block size is a power of 2, we can start by trying a block size of 8 KB (8,192 bytes).

The maximum raw disk throughput is determined by the speed at which the disk can read and write data. The key factor here is the data transfer rate, which is not provided in the question. Without this information, we cannot calculate the maximum raw disk throughput.

If the median file size is several Megabytes in size, a larger block size would give the optimal space efficiency. For example, if the median file size is 2 MB, a block size of 128 KB (the maximum block size) would be optimal. With a larger block size, there is less internal fragmentation, resulting in higher space efficiency.

User Jorge Guerola
by
8.6k points