Parallel_Systems_3 Shared Memory Machines 3. [2 points] Cons…

Parallel_Systems_3 Shared Memory Machines 3. Consider a shared address space architecture.  The address space is divided into two partitions:  Partition A is NCC-NUMA  Partition B is CC-NUMA  What guarantees are needed from the architecture to ensure that it provides sequential consistency memory model to the programmer? 

Parallel_Systems_4b M.E.Lock The context for this question i…

Parallel_Systems_4b M.E.Lock The context for this question is the same as the previous question. 4. Consider the following lock algorithm: while ((L == locked) or (T&S(L) == locked)) {    while (L == locked); // spin    delay (d); // different delays for different processors  }  // success if we are here    b. What are the program characteristics that would negatively impact the program’s performance with this lock algorithm implementation? In what way does it reduce performance? 

Parallel_Systems_5b M.E.Lock The context for this question i…

Parallel_Systems_5b M.E.Lock The context for this question is the same as the previous question. 5. Consider the ticket lock algorithm from lecture 4 (slide 108): b. What are the downsides of the ticket lock algorithm for systems with write-update cache coherence? Briefly explain your answer.

Potpourri_6b Potpourri The context for this question is the…

Potpourri_6b Potpourri The context for this question is the same as the previous question. 6. I/O ring data structures can be used to facilitate data transfer between paravirtualized guest OSs and the hypervisor. Consider a guest OS making a hypercall to the hypervisor to write some data to disk. The process begins with the guest making the write request, followed by the hypervisor processing the request and responding. Finally, the guest consumes this response.      b. Access to the ring data structure is through several pointers. For the disk write, describe in order the sequence of pointer modifications from request production to response consumption. Assume that there is space in the I/O ring for the Guest OS to enqueue a new request. For each modification, state: who makes the modification (guest or hypervisor).  which pointer is modified.  what the pointer modification is. 

Parallel_Systems_4c M.E.Lock The context for this question i…

Parallel_Systems_4c M.E.Lock The context for this question is the same as the previous question. 4. Consider the following lock algorithm: while ((L == locked) or (T&S(L) == locked)) {    while (L == locked); // spin    delay (d); // different delays for different processors  }  // success if we are here    c. What are the program characteristics that would enhance the program’s performance with this lock algorithm implementation? In what way does it improve performance?