Tag Archives: data

Power Consumption of Flash Drives

Historical Introduction to Flash Drive Price Versus Memory Capacity

Flash drive manufacturers typically advertise their products based on memory capacity, not on power consumption efficiency. One might believe that the two necessarily go hand in hand, with newer memory storage devices making considerable gains in both domains every year.

This is certainly the case with memory capacity. $200 could buy you 8 MB of memory in 2000 (year that USB 2.0 flash drives were first introduced to the public), as opposed to 2GB of memory in 2005, and 128 GB of memory in 2010.  Today, the average-sized 4 GB flash drive sells for around $10. Perhaps these increases in memory capacity reflect changes in everyday flash drive use – particularly in the domains of computer maintenance, law enforcement, business, and entertainment, where 4 GB of memory represents optimal balance between cost and desired memory capacity.

However, this does not imply that the power consumption efficiency of flash drives has seen similar gains. Our investigation attempted to establish whether or not larger drives are associated with smaller power consumptions.

Source: www.ehow.com

Results and Conclusions

The graphs in figures 1-3 show the power consumption of the laptop at rest and of the laptop with the flash drive when it is plugged in, as it opens files, and as it is ejected. Tables 1-3 summarize the data (in a previous blog post).

The data does not point to there being a significant difference among the flash drives and the amount of power they consume. Though we believed that flash drives with greater capacity would be more efficient and would consume less power than would smaller ones, especially older ones like the 256 MB, we did not observe a trend in our experiment. The differences that we did observe may be due to normal fluctuations in power consumption of each flash drive and may not be statistically significant. Some background tests in which a file was opened from the same flash drive more than once show that the power consumption reading on the watts up? PRO fluctuates and is not the same each time. It is clear from the data that a computer into which a flash drive has been plugged in consumes more power overall than a computer that has nothing running on it (figure 1). Ejection of a flash drive consumes about the same amount of power regardless of the size of the flash drive, though the 2 GB drive consumed the most power (+7.6 W), and the 2 GB micro flash drive consumed the least (+5.8 W). Whether this is a real trend, however, is uncertain because it was not observed when opening documents from flash drives.

In order to check whether the power consumption recorded when opening files is due to the flash drive or the computer, the same file was also opened from the computer. The data are inconsistent, however, and it is difficult to tell which device is consuming power, or whether there is a trend among different flash drive sizes. Power consumption was greater when opening files from the computer for only some of the files; for others, it was greater when opening them from a flash drive. This suggests that there is no difference among flash drives, and also that we cannot tell which device is consuming power, a confounding variable that must be eliminated in future experiments.

Overall, power consumption increases when a flash drive is plugged into a laptop, but it is unclear whether larger-capacity flash drives consume more or less power, whether they are merely plugged in or opening files. Future experiments should focus on a more rigorous statistical treatment of data, more of which should be collected using many different types of flash drives. Each type of flash drive should be tested multiple times (preferably three or more) and its power consumption should be recorded. Standard deviations should be determined, which can help eliminate or, at the least, pinpoint some of the uncertainties regarding fluctuations in power readings. It would also be interesting to test particular sizes of flash drives and determine if there is a difference in power consumption among different brands.

Energy consumption of Irene’s room

The energy consumption of my room turned out to be less than I thought. As expected, the refrigerator took up a large amount of my overall consumption. I was surprised at how little power my computer ended up using as I waited for the Watt’s Up Pro’s value to level out. One thing that surprised me was that the power strip and my chargers did not show a reading when tested while not in use. I had thought that the lights on the power strip and charger may have used some energy, but I guess the amount is negligible.

An interesting thing to note is how the wattage appeared to change when measurements using the Watts Up Pro were taken from devices plugged directly into the Watts Up versus when the values were taken from the power strip connected to the wall. In addition to this preliminary data, I will be recording how long I am using these appliances over the next 3 days to determine my average energy consumption.

Appliance Wattage Notes
laptop 21-22 Increased from average of 17-22W after extended use, high 43 when turned on
phone charger 4.1
alarm clock 1.1
phone + power strip 4.7
laptop + powerstrip 23-24 Started at 32, slowly decreasing, flatlines around 23-24
laptop+ phone+powerstrip 26.9
fan- high setting 30
fan-low setting 20.1
fan-medium setting 23.5
fridge 283 Slowly decreasing over time
camera charger 3.5
camera + powerstrip 3.9

Applications of mind-reading research

Group 13: Data!

Quick recap: We’re investigating the science of “mind-reading.” Each of us selected a specific topic to research. Maddy is looking into how fMRI works. Adam is examining current research being done. I (Jackie) am going to talk about applications of all this technology.

Here’s what I’ve found so far:

Applications of mind-reading technology

  • Applications of “mind-reading” technology
    • Overview, previous approaches, limitations – deCharms (2008) discusses how real-time brain imaging (e.g., with fMRI) allows access to both subjective experience (to an extent) and to objective observations and quantitative measurements of brain activity. He outlines some past approaches to “mind-reading” as well as limitations to current approaches. This leads to a discussion of the applications of current neuroimaging research:
  • Lie detection
    • Langleben (2008) argues that blood oxygenation level-dependent (BOLD) fMRI could be sensitive to differences between lies and truth. The key, he claims, is that BOLD fMRI can only compare states rather than positively identify deception. He discusses how many popular science articles conflate how much fMRI can do.
    • Mertens & Allen (2008) discuss whether ERP-based procedures could detect deception, instead of or in addition to fMRI.
    • Moreno (2009) discuss ethical issues in lie detection and how the law should be influenced by cognitive neuroscience, specifically in cases where neuroimaging could be used to determine truth, lies, and guilt.
  • Pain detection
    • Marquand et al. (2010) suggest that supervised machine learning algorithms can be used to decode fMRI data. They use this kind of technique to show that fMRI can be used to predict participants’ subjective pain ratings and propose that it will be a useful method for producing qualitative predictions about brain states.
  • Brain-computer interfaces
    • Direct brain communication in paralysis, motor restoration in stroke – Birbaumer & Cohen (2007) evaluate the use of EEG and fMRI in brain-computer interfaces, focusing on applications for paralyzed patients and for motor restoration in the case of stroke. Although currently, our understanding of the information flow in the brain that is required for such interfaces to work is incomplete, such interfaces will eventually be able to be used for direct brain communication and will allow otherwise “locked-in” patients to interact with the world.
    • Daly & Wolpaw (2008) also discuss advances in the analysis of brain signals and training patients to control those signals, focusing on EEG techniques specifically for patients with severe motor disabilities.
  • Pattern analysis and future research
    • Norman et al. (2006) argue that fMRI data can be used in conjunction with sophisticated pattern-classification algorithms to decode the exact information represented in a patient’s brain at a particular moment in time. They discuss factors that would boost the performance of this method — it is possibly the most promising research toward actual mind-reading.
  • In the press: Where the reporters think this research is headed.
    • Biever (2008) – to record and read people’s dreams
    • Debrosse (2010) – as a counter-terrorism technique
    • Masterman (2009) – in court, for lie-detection
    • Graham-Rowe (2011) – brain-computer interfaces useful to the disabled
    • BBC News (2005) – to read unconscious thoughts, attitudes, preferences

References (credible)
Birbaumer, N., & Cohen, L. (2007). Brain-computer interfaces: communication and restoration of movement in paralysis. Journal of Physiology, 579.3: 621-636.

Daly, J., & Wolpaw, J. (2008). Brain-computer interfaces in neurological rehabilitation. The Lancet Neurology, 7: 1032-43.

deCharms, C. (2008). Applications of real-time fMRI. Nature Reviews Neuroscience, 9: 720-729.

Langleben, D. (2008). Detection of deception with fMRI: Are we there yet? Legal and Criminological Psychology, 13: 1-9.

Marquand, A., Howard, M., Brammer, M., Chu, C., Coen, S., & Mourão-Miranda, J. (2010). Neuroimage, 49(3): 2178-2189.

Metens, R., & Allen, J. (2007). The role of psychophysiology in forensic assessments: Deception detection, ERPs, and virtual reality mock crime scenarios. Psychophysiology, 45 (2): 286-298.

Moreno, J. (2009). Future of neuroimaged lie detection and the law. Akron Law Review, 717-737.

Normon, K., Polyn, S., Detre, G., & Haxby, J. (2006). Beyond mind-reading: multi-voxel pattern analysis of fMRI data. Trends in Cognitive Sciences, 10(9): 424-430.

Popular science news

BBC News. (2005, April 25). Bran scan ‘sees hidden thoughts’. BBC News: Health. Retrieved April 19, 2011, from http://news.bbc.co.uk/2/hi/health/4472355.stm.

Biever, C. (2008, December 12). ‘Mind-reading’ software could record your dreams. NewScientist: Tech. Retrieved April 19, 2011, from http://www.newscientist.com/article/dn16267-mindreading-software-could-record-your-dreams.html.

Debrosse, J. (2010, March 15). Mind-reading technology being researched for foil terrorist attempts. McClatchy – Tribune Business News. Retrieved April 19, 2011, from ABI/INFORM Dateline. (Document ID: 1983690351).

Graham-Rowe, D. (2011, April 12). Dialing with your thoughts. Technology Review. Retrieved April 19, 2011, from http://www.technologyreview.com/communications/37357/?a=f.

Masterman, J. (2009, April 23). Current debates about fMRI research methods bear on policy questions. Science progress. Retrieved April 19, 2011, from http://www.scienceprogress.org/2009/04/fmri-mindreading-studies/.