Thanks to visit codestin.com
Credit goes to www.scribd.com

0% found this document useful (0 votes)
10 views11 pages

DSTN Problems

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
10 views11 pages

DSTN Problems

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 11

• Example:

– 512 bytes/sector

– 300 sectors/track

– 20,000 tracks/surface

– 2 surfaces/platter

– 5 platters/disk

– Capacity = 512 x 300 x 20000 x 2 x 5 = 30.72GB


• Average seek time=4ms
• Rotation speed= 15,000 RPM
• 512 bytes per sector
• No. of sectors per track=500
• Want to read a file consisting of 2500 sectors.
• Calculate the time to read the entire file
– A) File is stored sequentially.
B) File is stored randomly.
We know Disk access time = Avg Seek time + Avg rotation latency + Transfer Time

15,000 Rotation require 60 Sec


1 Rotation we can do in 4 ms
So avg rotation latency = 2 ms

Case 1 : Data are store in continuous way


To read 2500 sector we need to read 5 track and in 1 rotation we are reading 1 track
So to read 5 track we need 5* 4 ms = 20 ms
So transfer time will be 20ms
Since data is store in continuous manner we only have to deal with Avg seek time and Avg
rotation latency for first time
So disk access time = 4+2+20 ms = 26 ms

Case 2 : Data store in random way


As we know 1 track or 500 sector can be read in 4 ms
So 2500 sector we can read in 20 ms or 0.02 sec
Since data is store in random way every time we have to deal with seek time = 2500 * 4ms =
10,000 ms or 10 sec
Similarly every time we have to deal with rotation latency = 2500 * 2ms = 5,000 ms or 5 sec
So total disk access time = 0.02 sec + 10 sec + 5 sec = 15.02 sec
Compute the capacity of the disk with the following parameters 512 bytes per
sector, 1200 sectors per track, 80,000 tracks per surface, 2 surfaces per platter
and 7 platters per disk. Upper Side of top platter and lower side of bottom platter
are not used for storage.

The Equation to computer the capacity of a disk is Capacity = (Bytes/sector) x


(sectors/track) x (tracks/surface) x (surfaces / platter) x (platters / disk).

Applying the equation C = 512x1200x80000x2x6 = 589.824GB


Consider the data storage design where 5 disks are maintained in 3 different sites.
Due to some technical issues Disk 2 and Disk 3 losses data. Given below RAID
configuration, recover the data in disks 2 and 3. Note: Disk 4 is the redundant
parity storage for Disks 1 and 2. Disk 5 is the redundant parity storage for Disks 2
and 3.

Data in Disk 1 Disk 2 Disk 3 Disk 4 Disk 5

1001 ???? ???? 1110 1111

Disk 2 contents = XOR (Disk 1 and Disk 4), as 4th disk is the parity for the 1st and

= XOR (1001 ⊕ 1110)


2nd disks as per A.

= 0111 (This answer is used in following calculation)


Disk 3 contents = XOR (Disk 2 and Disk 5), as 5th disk is the parity for the 2nd and

= XOR (0111 ⊕ 1111)


3rd disks as per B.

= 1000
Consider an application that generates 20,500 IOPS, with 80 percent of them
being Reads. Calculate the disk load in RAID 5?

RAID 5 disk load = 0.8× 20500 + 4 × (0.2 × 20500) [because the write penalty

for RAID 5 is 4]

= 16400+ 4 × 4100

= 16400 + 16400

= 32800 IOPS
An SSD has 8 dies, each die has 32 planes, each plane has 4096 blocks, each block
has 512 pages, and each page is of 32 KB size. If 5% is saved for bad block
mapping, what will be usable size of the SSD.

8*32*4096*512*32*0.95 KB = 14.72TB (precisely considering 1KB = 1024 Bytes)


or 15.80TB (Approximately considering 1KB = 1000 Bytes)
A Unix-style i-node has 20 direct pointers and one single, one double and one
triple indirect pointers. Disk block size is 1 Kbyte, disk block address is 64 bits.
What is the maximum possible file size?

Each direct pointer points to 1KB of data, for a total of 20*1KB.


The indirect pointer points to an indirect block that contains pointers to data.
The indirect block is a 1KB block filled with 8-byte pointers, for a total of 128
pointers. Thus, the indirect pointer can refer to 128*1KB of data.
Similarly, the double-indirect pointer points a double-indirect block containing
128 pointers to indirect blocks, each of which points to 1KB*128 for a total of
128*128*1KB of data.
Finally, the triple-indirect pointer refers to 128*128*128*1KB.
The total data is thus 20*1KB + 128*1KB +128*128*1KB + 128*128*128*1KB.
= 2GB
We have two files each of 200 blocks with block size is 16 kB and block pointer
size is 2kB. If 80 blocks are identical from both the files, then compute the space
saved in block based deduplication

In normal scenario space required is 2*200*16=6400 kb. In deduplication 80


blocks are identical. So for second file space required is (200-80)*16+80*2=2080
kb. So space saved is 6400-(3200+2080)=1120 kB.
Assume you have a installed capacity of 100 TB. Provisioned capacity is of 175 TB.
What is the over provisioned percentage? If actual used capacity is 40 TB and on
an average 10 TB is consumed every month and you want a warning when 10 TB
or less space is available, then after how many month you should get the
warning?

Over provisioned percentage is 75%. Left space is 60 TB warning on 10 TB left.


Consumption rate is 10 TB per month. So warning on (60-10)/10=5 months.
We have a disc that has 10 tracks per platter with 8 sectors. The drive
supports 10 writable platters on a common spindle. Each sector stores four
512 byte blocks. There is a read write head for every platter. The heads can
be switched in 1ms. The track traversal is sequential and is at the rate of
10ms per track. How a 7.5 KB file could be stored ideally?

Ideal way of storing the file is to use consecutive sectors so that the
movement of head is minimum. Each sector can store 4x512 = 2096 Bytes.
No of sectors required is = (7.5x1024/2096) = 3.67~ 4 sectors.
Which one of the three RAID setups is more efficient in terms of capacity
utilization?
a. RAID 5 (22+1)
b. RAID 6 (42 +2)
c. RAID 6 (19+ 2)
Answer Should have proper justification

NOTE: RAID 5 (7 + 1) means that 7 disks are used for data storage and 1 disk is
used for parity storage
Formula Used : Capacity Utilization by parity (in %)
= Proportion of the disks utilized for parity * 100
= No.of.Disk used for Parity * 100 / Total No.of.Disks in RAID

a. RAID 5 (22 + 1): 100* (1/23) = 4.35 % of capacity used for Parity
b. RAID 6 (42 + 2): 100*(2/44) = 4.55% of capacity used for parity.
c. RAID 6 (19 + 2): 100*(2/21) = 9.76% of capacity used for parity

Hence RAID 5 (22+ 1) is better w.r.t to capacity utilization and proportion of space
sacrificed for Parity

You might also like