.   +---------------------+      
+---|---------------------|-------.    .---_||||||_-----------------.   
|:--|-----\\\\\//////-----|--. o   ----                          (\)| 
|#|||------         ------|--.||      [  Peter A. H. Peterson       | 
|-::|-----|.-,   ,-.|-----|===========[  Assistant Professor of CS  |  
|o+o|----(| @ |   @ |)----|:::::::::::[  329 Heller Hall            | 
||o||-----#  '-'    #-----|:::::::::::[  1114 Kirby Drive           | 
|||||------\"==="  /------|===========[  Duluth, MN 55812           | 
|||||-------`-----'|------| | o       [  pahp@d.umn.edu             | 
||/ |.----/'\'-----'/\---.|-+-'       [  pedro@tastytronic.net_  (/)|   
||^_|     '''|     |''    |-----'''----||||-||||||||||-||||||||||.  |  
||| +---------------------+-----------'                          | :| 
|:|                                                              |:.|
| '------------.  [back to research]                             |.:|
| home         |                                                 |:.| 
'------------. |  Power and Security Tradeoffs                   |.:|
.------------' |                                                 |:.|      
| cv           |  With support from Intel, we are currently      |.:|      
| .------------'  offering an undergraduate research course in   |:.|      
| '------------.  trade-offs between power and security.         |.:|      
| research     |  Virtually all security software requires       |:.|      
'------------. |  extra computation, and in doing so, consumes   |.:|      
.------------' |  more power than it otherwise would have. In    |:.|      
| interests    |  our course, student groups use the LEAP to     |.:|      
| .------------'  investigate various angles on these            |:.|      
| '------------.  trade-offs in the design of Banana Computer's  |.:|      
| blog         |  next-generation zPad. In the process,          |:.|      
'------------. |  students learn valuable lessons about power,   |.:|      
.------------' |  security, engineering trade-offs, research,    |:.|      
| reading      |  and high-quality experimental design.          |.:|      
| .------------'                                                 |:.|      
| '------------.  We feel that a strong understanding of         |.:|      
| quotes       |  experimental design is essential for all       |:.|      
'------------. |  computer scientists and engineers. Our         |.:|      
.------------' |  experience is that most students -- even       |:.|      
| contact      |  those with operating systems backgrounds --    |.:|      
|              |  do not know how to set up and execute a        |:.|      
|''''''''''''''|  practical and defensible experimental          |.:|      
| 001010101101 |  evaluation program. The complexity of modern   |:.|      
| 011011011001 |  software and hardware, in combination with a   |.:|      
| 010011011011 |  generally higher-level student perspective,    |:.|      
| 110011010111 |  makes for many difficulties in carrying out    |.:|      
| 100100001100 |  an effective experimental program.             |:.|      
| 101001100111 |                                                 |.:|      
| 001101100101 |  In general, we find that students do not       |:.|      
| 001111000000 |  understand the potential sources of error in   |.:|      
| 001010101101 |  their experiments, nor do they know how to     |:.|      
| 011011011001 |  isolate variables in order to confidently      |.:|      
| 010011011011 |  identify what effects they are actually        |:.|      
| 110011010111 |  measuring. Additionally, the probabilistic     |.:|      
| 100100001100 |  nature of complex systems means that students  |:.|      
| 101001100111 |  need to have a fairly strong grasp of          |.:|      
| 001101100101 |  statistical concepts in order to compile       |:.|      
| 001111000000 |  meaningful results. In addition to performing  |.:|      
| 001010101101 |  meaningful  experiments with the LEAP, we      |:.|      
| 011011011001 |  hope to educate students in these and other    |.:|      
| 010011011011 |  experimentation issues.                        |:.|      
| 110011010111 |                                                 |.:|      
| 100100001100 |                                                 |:.|      
| 101001100111 |                                                 |.:|      
| 001101100101 |  [back to research]                             |:.|      
| 001111000000 |                                                 |.:|      
|'''''''''''''''-------------------------------------------------'::'. 
|::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::| 
`=======---------.______.-------:::::::-:------------------------. ? | 
                                                                  ---'