aboutsummaryrefslogtreecommitdiffstats
path: root/csci5451
diff options
context:
space:
mode:
Diffstat (limited to 'csci5451')
-rw-r--r--csci5451/ass1.tex47
1 files changed, 42 insertions, 5 deletions
diff --git a/csci5451/ass1.tex b/csci5451/ass1.tex
index 0124d95..16bd360 100644
--- a/csci5451/ass1.tex
+++ b/csci5451/ass1.tex
@@ -6,14 +6,51 @@
\subsection{Overlapping Intervals}
\subsection{Non-Overlapping Intervals}
\section{Question 2}
- \subsection{}
+ \subsection*{A}
Shared memory has all of the processors access one large pool of memory while distributed memory has each processor have a section of the memory.
- \subsection{}
+ \subsection*{B}
Distributed memory programs communicate via pipes and message queues while shared memory programs share memory with locks in place to prevent unintended behavior and race conditions.
- \subsection{}
+ \subsection*{C}
Shared memory is in most devices nowadays with even phones having multiple cores and a single bank of shared memory. Large rendering farms and HPC will have distributed memory to split rendering or to maximize performance.
- \subsection{}
+ \subsection*{D}
Distributed is easier to scale because the cluster can have nodes added or removed easily without affecting the other nodes.
\section{Question 3}
-
+ \subsection*{A}
+ Concurrency: 8\\
+ Critical path length: 4\\
+ Maximum achievable speedup: \(15/4\) \\
+ Minimum number of processes needed: 8\\
+ Maximum achievable speedup if the number of processes is limited to: \\
+ 2- \(15/8\)\\
+ 4- \(15/5\)\\
+ 8- \(15/4\)\\
+ \subsection*{B}
+ Concurrency: 8\\
+ Critical path length: \\
+ Maximum achievable speedup: \\
+ Minimum number of processes needed: \\
+ Maximum achievable speedup if the number of processes is limited to: \\
+ 2- \(15/8\)\\
+ 4- \(15/5\)\\
+ 8- \(15/4\)\\
+ \subsection*{C}
+ Concurrency: 8\\
+ Critical path length: 7\\
+ Maximum achievable speedup: \(14/7\)\\
+ Minimum number of processes needed: \\
+ Maximum achievable speedup if the number of processes is limited to: \\
+ 2- \(14/10\)\\
+ 4- \(14/8\)\\
+ 8- \(14/7\)\\
+ \subsection*{D}
+ Concurrency: 2\\
+ Critical path length: 8\\
+ Maximum achievable speedup: \(15/8\)\\
+ Minimum number of processes needed: 2\\
+ Maximum achievable speedup if the number of processes is limited to: \\
+ 2- \(15/8\)\\
+ 4- \(15/8\)\\
+ 8- \(15/8\)\\
+ \section{Question 4}
+ \subsection*{A}
\end{document} \ No newline at end of file