CSE 721
Winter 2009
Sample Problems for Final Examination
1. Consider 4096processor systems with the following topologies:
i) 3D torus (with
wraparound, bidirectional links), ii) hypercube (with bidirectional links). Order the
systems with respect to the following:
(a)
D
av
, the average distance from a fixed processor to all other processors in the
system (including itself).
(b) Ratio of
D
av
to diameter of system.
(c) Aggregate peak communication bandwidth of system (proportional to product of
number of links and width of links) if the bisection width of the system is limited
by board constraints to be no more than 2
16
wires.
2. A parallel prefix sum is to be computed for a vector of size 1,024,000. Assume that
the startup time for a pointtopoint communication is 100
μsecs
, the perword trans
mission time is 0.1
μsecs
, and time per arithmetic operation is 0.01
μsecs
. Estimate
the speedup achievable for the parallel prefix operation on 2, 8, 64, 256 and 1024
processors, for:
(a) a hypercube system
(b) a ring connected system
3. A company is planning to build distributedmemory parallel computers for clients who
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Winter '11
 Saday
 Central processing unit, Bisection Width, parallel prefix sum, parallel prefix, distributedmemory parallel computers

Click to edit the document details