In multi-core CPU systems, several CPUS (or cores) can access a common memory space through a shared bus or crossbar switch. In such multi-core CPU systems, multiple threads are created to run concurrently in different cores. Interestingly, multiple threads can also be created in a single CPU system. However, the mechanism of running the multiple threads in a single CPU system is slightly different compared to using multi-core CPU systems. Explain how a single CPU system supports running multiple threads. Use the editor to format your answer

In multi-core CPU systems, several CPUS (or cores) can access a common memory space through a shared bus or crossbar switch. In such multi-core CPU systems, multiple threads are created to run concurrently in different cores. Interestingly, multiple threads can also be created in a single CPU system. However, the mechanism of running the multiple threads in a single CPU system is slightly different compared to using multi-core CPU systems. Explain how a single CPU system supports running multiple threads. Use the editor to format your answer

Systems Architecture

7th Edition

ISBN:9781305080195

Author:Stephen D. Burd

Publisher:Stephen D. Burd

Chapter11: Operating Systems

Section: Chapter Questions

Problem 21VE

See similar textbooks

Similar questions

Consider the following three CPU organizations:CPU SS: A 2-core superscalar microprocessor that provides out-of-order issue capabilities on 2 function units (FUs). Only a single thread can run on each core at a time.CPU MT: A fi ne-grained multithreaded processor that allows instructions from 2 threads to be run concurrently (i.e., there are two functional units), though only instructions from a single thread can be issued on any cycle. CPU SMT: An SMT processor that allows instructions from 2 threads to be run concurrently (i.e., there are two functional units), and instructions from either or both threads can be issued to run on any cycle. Assume we have two threads X and Y to run on these CPUs that include the following operations: Thread X Thread Y A1 – takes 3 cycles to executeA2 – no dependences A3 – conflicts for a functional unit with A1A4 – depends on the result of A3 B1 – take 2 cycles to executeB2 – confl icts for a functional unit with B1B3 – depends on the result of…
Explain the trade-offs between increasing the number of hardware threads in a dynamic multithreading processor and the complexity of managing them. What are the practical limits to the number of threads that can be effectively supported?
a) Suppose two threads T1 and T2 are running concurrently in the same process with a single CPU. Suppose T1 does “Load X into register R1” machine instruction. Then the CPU scheduler switches the CPU to T2. Will T2 see the same vaue of R1 as loaded here by T1? (YES or NO)? Explain in less than 50 words. Be specific. (b) Suppose two threads T1 and T2 are running concurrently in the same process with a single CPU. Is it possible to have a moment (time instant) where both threads have made a request to Memory Unit to read some data (ex. as part of a “Load … into register …” instruction) and Memory Unit has not yet completed either of these two requests? Explain in less than 50 words.
Explore the techniques used in modern CPUs to dynamically adjust the number of active threads and their priorities based on workload characteristics.
Uniprogramming typically provides better utilization of system resources than multiprogramming. (T/F). A fetched instruction is normally loaded into the Instruction Register (IR). (T/F) If a kernel is single-threaded, system calls from any thread can block the entire task .(T/F) Solid state drives are less reliable than conventional Magnetic hard drives. (T/F) The less-privileged processor execution mode is often referred to as kernel mode. (T/F) The Dining Philosopher’s Problem illustrates basic problems in deadlock and starvation. (T/F)
(Exercise 4.3) Which of the following components of program state are shared across threads in a multithreaded process? a. Register values b. Heap memory c. Global variables d. Stack memory (Exercise 4.4) Can a multithreaded solution using multiple user-level threads achieve better performance on a multiprocessor system than on a single processor system? Explain. (Exercise 4.6) Is it possible to have concurrency but not parallelism? Explain. (Exercise 4.9) A system with two dual-core processors has four processors available for scheduling. A CPU-intensive application is running on this system. All input is performed at program start-up, when a single file must be opened. Similarly, all output is performed just before the program terminates, when the program results must be written to a single file. Between startup and termination, the program is entirely CPU-bound. Your task is to improve the performance of this application by multithreading it. The application runs on a system that…
. Consider a system that has multiple processors where eachprocessor has its own cache, but main memory is shared among allprocessors.1. a) Which cache write policy would you use?2. b) The cache coherency problem. With regard to the systemjust described, what problems are caused if a processor has acopy of memory block A in its cache and a second processor,also having a copy of A in its cache, then updates mainmemory block A? Can you think of a way (perhaps morethan one) of preventing this situation, or lessening itseffects?
Suppose that when run using a single thread, a particular program spends 20 seconds executing code that cannot be parallelized, and 280 seconds executing code that is highly parallelizable. Using Amdahl’s law, give an upper bound on the speedup (ratio of old to new execution time) that might be achieved for this program by exploiting parallel execution on a highly-parallel computer with thousands of processors/cores.
When a program is adapted to run on multiple processors in a multiprocessor system, the execution time on each processor is comprised of computing time and the overhead time required for locked critical sections and/or to send data from one processor to another. Assume a program requires t = 100 s of execution time on one processor. When run p processors, each processor requires t/p s, as well as an additional 4 s of overhead, irrespective of the number of processors. Compute the per-processor execution time for 2, 4, 8, 16, 32, 64, and 128 processors. For each case, list the corresponding speedup relative to a single processor and the ratio between actual speedup versus ideal speedup (speedup if there was no overhead).
In uniprocessor systems, crucial sections are used to protect shared memory against corruption caused by concurrent process access. When it comes to clusters and grids, why is it that we don't always have to follow such procedures?
Describe the CPU's support for multiple threads and briefly go over the three distinct forms of multithreading.
Explain the concept of thread-level speculation and its role in dynamic multithreading. How can it be utilized to further enhance processor performance?