Computer Architecture (2 Problems)

In summary, the first question asks for the average access time for a block of data on a hypothetical disk drive with 32 sectors per track, 3600 RPM speed, and 7 millisecond average seek time. The answer should be in milliseconds and reduced to a final value.The second question is about transaction processing and its focus on I/O rate, specifically the number of disk accesses per second. The CPU performance is given as 800 MIPS and each transaction requires 2 disk reads and 2 disk writes. The database software and operating system have their own instructions for processing a transaction and performing a disk read or write access. The I/O bus and memory have a data transfer rate of 1000 MB/sec and each read/write
  • #1
JackFrank
1
0
need help answering these 2 questions

Q1.
A hypothetical disk drive has the specifications shown below. Find the average access time for a block (sector) of data.

I want the answer in milliseconds, and you need to reduce your answer to a simple final value.

• 32 sectors per track
• 3600 RPM speed
• 7 millisecond average seek time
________________________
Q2.
Transaction processing is chiefly concerned with I/O rate: number of disk accesses per
second, as opposed to data rate, and the I/O speed versus CPU speed in terms of Transactions Per Second (TPS). Assume the following information:

• CPU has the performance of 800 MIPS.
• Each transaction requires 2 disk reads plus 2 disk writes.
• The database software executes 40000 instructions to process a transaction.
• The operating system uses 15000 instructions for each disk read or write access.
• The I/O bus and memory are capable of sustaining 1000 MB/sec of data transfer without interfering with the CPU.
• Each read/write access transfers 100 bytes.

a. What is the peak performance of the CPU in terms of TPS?
b. What is the peak performance of the I/O bus system in terms of TPS?

Thank you in advance.
 
Physics news on Phys.org
  • #2
Okay, so what are your thoughts on these questions?

EDIT: Welcome to Physics forums, by the way. However, the way these forums works is that you need to show some work before we're able to help you out (i.e. we don't give out answers / do your homework for you)
 
  • #3


Q1. To find the average access time for a block of data, we can use the formula:
Average Access Time = Average Seek Time + Average Rotational Latency + Average Transfer Time

Average Seek Time = 7 milliseconds
Average Rotational Latency = (1/2) * (1 / (RPM / 60)) = (1/2) * (1 / (3600 / 60)) = 0.0083 milliseconds
Average Transfer Time = (1 / (RPM / 60)) = (1 / (3600 / 60)) = 0.0083 milliseconds

Therefore, the average access time = 7 + 0.0083 + 0.0083 = 7.0166 milliseconds

Q2.
a. To find the peak performance of the CPU in terms of TPS, we can use the formula:
TPS = (CPU Speed / (Number of instructions per transaction * Number of transactions)) * 1000

Number of instructions per transaction = 40000 + 2 * 15000 = 70000
Number of transactions = 2 (since each transaction requires 2 disk reads and 2 disk writes)

Therefore, TPS = (800 * 10^6 / (70000 * 2)) * 1000 = 57142.857 TPS

b. To find the peak performance of the I/O bus system in terms of TPS, we can use the formula:
TPS = (Data transfer rate / (Number of bytes per transaction * Number of transactions)) * 1000

Number of bytes per transaction = 2 * 100 = 200 (since each transaction transfers 100 bytes for each read/write access)
Number of transactions = 2 (since each transaction requires 2 disk reads and 2 disk writes)

Therefore, TPS = (1000 * 10^6 / (200 * 2)) * 1000 = 2500000 TPS
 

1. What is computer architecture?

Computer architecture refers to the design and organization of a computer system, including its hardware and software components. It encompasses the overall structure, functionality, and performance of a computer system.

2. What are the main components of computer architecture?

The main components of computer architecture include the central processing unit (CPU), memory, input/output devices, and communication channels. These components work together to process and store data, and to communicate with other devices.

3. What is the difference between Von Neumann and Harvard architectures?

Von Neumann and Harvard architectures are two different approaches to organizing the memory and data flow in a computer system. In Von Neumann architecture, the same memory is used for both data and instructions, while in Harvard architecture, separate memories are used for data and instructions. This allows for faster data access in Harvard architecture, but it is more complex and expensive to implement.

4. What is pipelining in computer architecture?

Pipelining is a technique used in computer architecture to increase the efficiency of instruction processing. It involves breaking down a program into smaller parts and overlapping the execution of these parts. This allows for multiple instructions to be processed simultaneously, resulting in faster overall processing speed.

5. How does cache memory affect computer architecture?

Cache memory is a small, high-speed memory that stores frequently used data and instructions for faster access. It is located between the CPU and main memory. By storing frequently used data and instructions in cache memory, the computer system can retrieve them quickly, improving overall performance. However, the addition of cache memory can also increase the complexity and cost of a computer system.

Similar threads

  • Engineering and Comp Sci Homework Help
Replies
4
Views
3K
  • Engineering and Comp Sci Homework Help
Replies
1
Views
1K
  • Sticky
  • Programming and Computer Science
Replies
13
Views
4K
  • Programming and Computer Science
Replies
1
Views
975
  • Programming and Computer Science
Replies
29
Views
3K
  • Engineering and Comp Sci Homework Help
Replies
2
Views
2K
  • STEM Academic Advising
Replies
1
Views
2K
  • Computing and Technology
Replies
1
Views
4K
  • Computing and Technology
Replies
22
Views
4K
Back
Top