Journal Post 13

This is the most current layout of the poster that I have. I like this better than the previous layout but I still have work to do on the picture placement as well as rewording the text sections. All the information I have received on poster presentations is that the more concise they are the better. I think the line graphs work really well to make the case of the thin-client and to show the difference in energy output of the respective information worker categories.

Poster Layout (work in progress)

Journal Post 12

The main focus of this week has been working on the poster layout and adding graphs compiled from the data I have gathered. The graphs that I added to the poster layout were:

• kWh per week
• Metric tons of CO2 emitted per week
• kWh per year
• Metric tons of CO2 emitted per year
• Each was compiled with data for 6 users, 10 users, and entire UM campus (1% and 10%)

I also added the introduction section and materials section to the poster layout. I will do a post of the poster layout as it is right now. The layout is not properly formatted and is still a work in progress. 

I have also completed the user survey and will also embed that in the blog and send an email to each user asking them to respond. 

The spreadsheet has not been updated as there is no additional data to add until Monday when the last set of thin-client data is pulled. I will also be redoing the control test on one user starting Monday. 

This week I will be going to the training for the UMCUR conference and assisting at The University of Montana Science Fair.

The rest of my time working on the project will be finalizing the data and putting together the results, conclusions and discussions part of the poster. 

Journal Post 11

My abstract has been accepted for the UMCUR conference! I will have to do one more rewrite which will be do April 4th. Due to some unforeseen circumstances the last wave of the thin-client rollout was postponed until Monday of this week so the data should be in the spreadsheet by next Tuesday.  The things that have been accomplished since the last post include:

• Spreadsheet updated to include a new comparison sheet that includes up to date extrapolation data for the entire UM campus. The spreadsheet can be found (here). 
• The data includes thin-client and repurposed extrapolation for 1%, 2%, 5%, and 10% of entire university.
• I have taken the total number of active ports that was given to me by UM IT and cut it in half to account for controllers, printers and laptops that may be using the active ports. 
• The data is accurate as of now but I do need to measure the energy output of one of the regular workstations again. 
• I have started to put together a layout for the poster presentation that can be found (here). It does not have any color is a work in progress. 
• I am using Adobe Illustrator and spent some of the weekend familiarizing myself with the software.
• The questionnaire is almost finished now and should be ready to be passed out by the time the last wave of thin-client testing is finished. The questions I have right now can be found (here).

It is getting down to crunch time now with the presentation only 1 month away. I feel very confident that the data that i have gathered will make for an interesting presentation and findings that will be useful to the university system as a whole. The week ahead will be a busy one for me as I will attend the training for the UMCUR presentation as well as finalizing the thin-client data. In the coming weeks the data should become more concrete and my poster presentation should start to take more of it's final form. Check back next week (hopefully this weekend) to see the progress I have made and to get a better understanding of the findings and presentation strategy I will be taking. 

Journal Post 10

The data has all been collected for the repurposed machines! Last week was the start of the testing using the thin clients. So far 4 users have used the thin clients and and data has been collected for two of the users. Following up with the to do list from the last post I have accomplished most of what is on the list:

• Submitted abstract for UMCUR presentation
• Revised spreadsheet with all new data gathered and also separated the data in to more relevant categories. The spreadsheet can be found (here). There are also some preliminary graphs and calculations on it. 
• I found a conversion method from electricity used to Metric tons of CO2 emitted from the EPA website (here)

• The revision of the questionnaire has proved more challenging than I originally thought it would. The problem I am having is how to word questions that have to do with performance of a machine and what the user experience was like.
• I have combined the repurposed user data and the server data including average kWh with 1,2,5,10 users on the same server.

•  I also received a rough estimate of how many computers the university has on campus. The number is actually the number of active network ports but it was the most reliable data that could be provided by the IT department. 
•  That number was then used to extrapolate the data that had been gathered to the entire campus. (This process is not yet complete and the numbers will change from what is one the spreadsheet at the current moment)

Now that we are almost done collecting the data it is time to get down to putting it all together and figuring out the poster layout.  I will be using Adobe Illustrator to design the poster and will use the university printing center to produce it. What I still need to accomplish is:

• Figure out a way to show the carbon footprint reduction (if any) for the repurposed machines.
• make an appointment with Lisa
• Create survey and have the users respond to it
• Work more on extrapolating data so that it is more realistic 
• 8 year plan for roll out
• What is the end result and how do I show it on the poster presentation
• Get more familiar with Illustrator
• Create layout for poster presentation (sans data and text for now)
• Look into information on energy output per calculation speed (joules per bits)
• Take pictures of thin-client and regular workstation for all users.

That is all I have for now but I will post with an update soon.

Journal Post 9

It has been a long time since my last post so I have quite a bit to update. The main work that has been done over the month of February has been gathering the data on the repurposed workstations. The data that has been gathered has not been what I was expecting. The repurposed machines seem to have the same (if not more) energy output than the previous workstations the users were using. All of the repurposed tests have been finished at this point and the data can be found in the spreadsheet (here). We also received the thin clients that we will be using for the third part of the test this week they are HP t5335 thin clients. They come with 36 Watt power supplies and the full specs can be found in this pdf (HP t5335 specs). The things that still need to be accomplished include;

• Submit abstract for UMCUR presentation (poster session)
• Finish thin client testing
• Revise end of study questionnaire
• Continue analysis of data 
• Calculate metric tons of CO2 based on avg KwH measurements
• Expand data to entire university system
• Figure out how many total workstations there are on the university campus (and what category each falls into). 
• Define categories for study group
• Kiosk, Faculty (power user), Data entry
• Combine server and users data to get power output with server
• Will be done using different amount of users on single server
• Based on data collected server runs at same power level no matter how many users are logged in simultaneously 
• Create graphs to so each user on different type of workstation

Today I will be working all day to start to make sense of the data that I have collected. I will be creating multiple spreadsheets and updating the calendar to reflect the current status of the data collection. When I make significant progress I will post again with links to the spreadsheets as well as the updated calendar. the end survey should also be completed by the end of the day. I will be more diligent at staying up to date with the blog in the future. 

Journal Post 8

It has been a busy couple of weeks for me so I have not been able to update the blog as often as I had hoped. I have done quite a bit of work on my research project since the last post, one of the main updates to the blog are the new calendar that I have embedded which shows the tentative schedule for testing including the weeks that each user will be testing each system. I have been using the repurposed system exclusively at work and have been keeping notes on the overall functionality of the the system and user experience. I will post the results when i have finished testing. The main things that I have accomplished since my last post include;

• I have a new project participate
• Griz Central student kiosks
• On 24/7 all they do is access internet and print
• Started David's testing on repurposed machine on 1/31/2012
• Started control test on Mary's PC on 1/24/2012
• Completed control test on 1/31/2012 
• results in spreadsheet (here)
• Researched thin clients and decided on either of two models 
• HP Smart Client t5335z
• HP Smart Client t5565z
• Specifications can be found (here)
• researched definitions of the information (knowledge) worker
• Almost everything I found referenced Peter Drucker (Peter Drucker Wikipedia)
• I need to find more information 
• Ubuntu 9.10 USB boot drives are stable and functioning properly 
• Sound is still not working
• All programs and applications installed successfully on server and are available to all users
• Microsoft Project and Visio were not able to be installed due to licensing issues
• We will attempt to install directly on profile that needs access to these programs
• I am currently working on a survey to hand to all participates to describe user experience after using repurposed machine
• I will post survey as soon as it is finished and results as they come in 
• Updated spreadsheet with computer models and PSU sizes 
• Decide whether I want to do a poster or oral presentation
• I love the visual idea of the poster but I would also like to be able to put my project into words so....

I should have more time this week to work on the project because I have taken Wednesdays off from work with that sole purpose.

Tentative Calendar for Remaining Tests

Journal Post 7

I have been doing quite a bit of work since I last posted.

• successfully create all USB boot media
• Using Ubuntu 9.10 I have created boot USB drives that boot directly into the Windows desktop environment on the server 
•  user never sees the Ubuntu desktop until the user logs off of the server
•  I have also removed the components from all of the Dell Vostro GX520's
• Created user environment on the server
• Added users to Remote Desktop Users group
• Installed applications
• Adobe Reader, java, Adobe Flash, Google Chrome, Mozilla Firefox, TechSmith Camtasia Relay
• Setup personal Microsoft Outlook
• Tested applications in real world environment
• Mapped network drives
• Installed printer and printed test page
• Removed Kill-A-Watt monitoring device from PTS Student PC and collected final data on January 19th, 2012 at 1:30 p.m.
• data in spreadsheet (here)
• Installed Kill-A-Watt monitoring device on repurposed PTS Student PC at 2:30 p.m. January 19th, 2012
• preliminary data in spreadsheet (here)
• Sent out email to all volunteers asking what specialized software will be needed 

Next steps

• install the monitoring devices on the two full PC's that I have not monitored as 
• setup repurposed PC's for users besides myself and monitor them
• get energy readings from server
• setup meeting with adviser 

Journal Post 6 - Persistence Achieved!

I have been working all weekend to find a way to make the VDI Blaster software persistent when booting from a USB flash drive. In the end I could not find a solution without running the Echo Management server that is provided with the software. Using the Echo Management software proved to be a major undertaking and I decided to look into other options. I tried 3 versions of Puppy Linux but could not achieve persistence on those either. After scouring many forums and message boards I came across a tool that allows you to make Ubuntu distributions persistent called Universal  USB Installer. I used Ubuntu 9.10 and finally I have achieved the persistence I was looking for. I have now gone ahead and created a launcher in the startup folder that will automatically launch the RDP session when the USB flash drive is booted so the users will not need to even click on the connection button in order to see the Windows desktop that will be running on the server. This was a long and painstaking process but in the end I believe that it is the only way this project can succeed. I am very happy to have this part behind me.

I have also updated the spreadsheet with the data to incorporate more hardware and usage statistics. It should now also be accessible to any body who would like to see it. Here is the link : spreadsheet update

Journal Post 5

Pulled data from PTS student PC

• Data in spreadsheet (here)

Had a meeting with adviser today discussed plan for the next couple of weeks. The things that I need to get done are:

• Add model specifications and KWH average to spreadsheet
•  Add model specifications to spreadsheet
• Create persistent VDI connection
• Using a flash drive to boot the VDI software does not hold any configuration that I have done
• May have to contact tech support
• Server configuration: Office Software, Email configuration, User profile, drive mapping etc..   
• Search for low wattage power supply.
• So far I have not found anything that will work with Dell Optiplex GX520
  

Journal Post 4

Removed Kill-A-Watt from repurposed Dell Optiplex GX5200 at 6:15 p.m. on January  6th, 2012

• Data in spreadsheet (here) 

I now have a good amount of preliminary data and will start to analyze what I have gathered.

Tomorrow I will bring the repurposed PC to campus and begin to test it in a real world environment.

Journal Post 3

Removed Kill-A-Watt from Dell Optiplex GX520 (active) at 7:09 p.m. on January 5th, 2012

• data in spreadsheet  (here)

Started monitoring "stripped" Dell Optiplex GX520 using VDI Blaster software at 1:05 p.m. January 6th, 2012
Booted "stripped" Dell Optiplex GX520 using VDI Blaster software

• Created live USB flash drive per instructions from DevonIT
• Disabled SATA and PATA controllers in the BIOS 
• Changed boot sequence to only boot from USB
• Froze at "Welcome" screen for 2-3 minutes
• Once fully booted adding RDP server was simple process
• Once added the RDP link it automatically placed on the desktop 
• This froze for 4-5 minutes when applying the settings
• The RDP client was extremely responsive and boted directly into the server once all the settings had been applied 
• Web browser and applications (MS Word, Excel, Outlook, Dropbox) pulled up quickly 
• Frame rate on video is acceptable but not ideal

I am currently using the repurposed legacy Dell Optiplex GX520 with the HDD and CD/DVD drive removed (see previous post) to write this post and update all the data

More data and observations to come later today

Journal Post 2

Removed Kill-A-Watt energy monitor from Penny's computer at 10:31 a.m. on January 5th, 2012

• results in spreadsheet (here)

Started Dell Optiplex GX520 energy monitoring (active) at 12:40 p.m. January 5th, 2012

• Ran a 4 hour movie in the DVD drive 
• Downloaded 1 GB file 
• Created a spreadsheet
• Streamed 1 hour worth of internet video

Used FTP client to download VDI Blaster Software

• Burned the .iso image I received onto 2 GB USB flash drive
• I will test it tonight or tomorrow

Removed components from additional Dell Optiplex GX520

• Removed DVD drive
• Removed hard drive

Before

PSU tag

After

Passive heat sink for CPU  

Power supply with fan

I will have the results of the active GX520 monitoring posted tomorrow

Journal Post 1

I will be using this blog to keep a journal of the progress that I make and the findings that I come across. I have been working on this project for a little while now so I this post will have all of my progress up until today. From now on I will post everyday and maybe more.

This is the spreadsheet that is referenced Spreadsheet Data

Tuesday December 6, 2011

Met with Tom to discuss project

Discussed who test study group will be comprised of

-          Faculty

o   Tom 

o   Penny 

-          Staff or staff supervised workstations

o   David 

o   Debbie 

o   PTS Student PC

o   COT outreach student PC

Received login information for server

Discussed Android OS RDP client

-          PocketCould by WYSE

Discussed legacy OS

-          PuppyLinux

-          Other small Linux OS with RDP client

-          VDI Blaster software

Repurposed PC’s retrieved from PTS surplus and IT Central

-          Dell Optiplex GX520

o   2-4 available

-          4 USB flash drives

o   2 x 2GB

o   2 x 4GB

Determined the five steps that need to be taken;

                Determine study group of 6 people

o   David Opitz

o   Penny Jakes

o   Tom Gallagher

o   Mary Opitz student PC

o   Debbie Crawford

o   PTS student PC (Sam and I)

-          Standard workstation measurements

o   David

·   Started Friday December 16^th at 4:45 p.m. 

o   Penny

·   Started December 7^th at 3:05 p.m.

o   Tom

·   Started Wednesday December 7^th at 9:27 a.m.

-          Zero-client build measurements

-          Legacy build measurements

-          Server measurements

Wednesday December 7^th 2011

-          Started measurements for Penny and Tom

o   Tom 9:27 a.m.

o   Penny 3:05 p.m.

-          Discussed using thin-clients instead of Android devices due to Android resolution and power/USB connection

o   On most Android devices there is only one USB connection and it cannot be used simultaneously to charge/power the device and used as an input device.

-          I received a “New Internet Computer” (NIC) device from Tom to test the functionality of it.

o   As far as I can tell right now the only way to boot is with a CD and that is hardly functional.

o   Will test further

Wednesday December 14^th 2011

-          Took electrical measurements from Tom’s full PC at 9:27 a.m.

o   Data in spreadsheet

Tuesday January 3^rd 2012

-          Took electrical measurements for David’s full PC at 8:05 a.m. on January 3^rd 2012

o   Data in spreadsheet 

-          Started measurements for PTS student PC at 3:28 p.m. on January 3^rd 2012

-          Looked into VDI Blaster software that was acquired by Tom for the project

o   My first impression is that the software is what we need

o   VDI Blaster has built in RDP Client and can be installed on flash drive

-          Tried WYSE PocketCloud from UMwpa wireless internet on campus to server in HB04 and it worked flawlessly.

o   Was able to navigate and open applications

Wednesday January 4^th 2012

-          Met with Tom

o   Discussed repurposed machine

§  Will download VDI Blaster software via instructions in email

§  Get measurements on machine before removing any parts

·         Started taking measurements January 4^th at 10:15 a.m.

-          Discussed zero/thin client

o   Tom will procure zero/thin clients for project

-          Discussed presentation and journal

o   Options are poster or oral presentation

o   Journal will now all be done using Blogger

-          Removed Kill-a-watt from Tom’s computer at 9:31 a.m. on January 4^th 2012 and took readings

o   Data in spreadsheet

-          Took readings from server at 9:40 a.m. on January 4^th 2012

o   Data in spreadsheet

-          Started measurements on Dell Optiplex GX520 that was not being used

o   Started the computer with Kill-A-Watt and left it idle

o   Removed Kill-A-Watt from Dell Optiplex GX520

·         Data in spreadsheet

Abstract

This is the abstract that I wrote for the project proposal. The entire proposal can be found here: Sustainable Computing Through Desktop Virtualization.

Sustainable Computing Through Desktop Virtualization

Abstract

Green IT or green computing involves implementing environmental and sustainable practices in computing and information technology. Green IT revolves around four core concepts; virtualization, workforce mobility, e-waste recycling and energy efficiency (Green computing). The University of Montana’s Climate Action Plan outlines strategies for achieving campus-wide carbon neutrality by the 2020.  Desktop computers in an office environment are one of the biggest overall contributors to the office workers overall carbon footprint. Energy-wise IT is presented in the Climate Action Plan as a strategy for achieving greater energy efficiency and conservation. (Peacock, 2010).  The Desktop Virtualization project will study the implementation of desktop virtualization through thin-client computing in the workplace to achieve the following goals (1) achieve greater energy efficiency (2)Investigate repurposing of legacy computers as thin-client devices to extend the computing lifecycle and reduce e-waste (3) reduce the carbon footprint of the information worker. It will measure energy conservation through the replacement of standard desktop PCs with thin-client computing devices and virtualized desktops as a way to reduce the overall carbon footprint of the information worker. It will also study the use of legacy computers as thin-client devices to extend the computing lifecycle and reduce e-waste. The Desktop Virtualization project will take action to meet the growing need to reduce energy consumption and make the University of Montana Climate Action Plan a reality.

Intoduction

My name is Lance Pellerin and I'm in the last semester of the Information Technology Systems program at The University of Montana - College of Technology. In order to fulfill my internship requirement I decided to do a research project on sustainability and green IT. With the help of my adviser and some research I came up with a project that revolves order the four key components of green IT (reuse/recycle, virtualization, mobility and energy efficiency). I wrote a proposal for The Davidson Honors College and I was lucky enough to be chosen. Over the next couple of months I will be updating this blog with my progress and results.