Computer Summary Page

From Unofficial BOINC Wiki

Contents 1 General 1.1 "Computer Summary" Page As Visible To Other Participants 2 Line by Line 2.1 IP Address (Private) 2.2 Domain Name (Private) 2.3 Local Standard Time (Private) 2.4 Name (Private) 2.5 Created 2.6 Total Credit 2.7 Recent Average Credit 2.8 CPU Type 2.9 Number of CPUs 2.10 Operating System 2.11 Memory 2.12 Cache 2.13 Swap space 2.14 Total disk space 2.15 Free disk space 2.16 Measured Floating Point Speed 2.17 Measured Integer Speed 2.18 Average Upload Rate 2.19 Average Download Rate 2.20 Average Turnaround Time 2.21 Maximum Daily Work Unit Quota per CPU 2.22 Results 2.23 Number of Times Client Has Contacted Server (Private) 2.24 Last Time Contacted Server 2.25 % of Time Client Is On (Private) 2.26 % of Time Host Is Connected (Private) 2.27 % of Time User Is Active (Private) 2.28 Average CPU Efficiency 2.29 Result Duration Correction Factor 2.30 Location (Private) 2.31 "Merge This Computer" Link 2.32 "Delete This Host" Link 3 Notes, Oddities, and "Watch Out For This" 4 Also See

[edit] General

This is the "Computer Summary" Page for one of the computers I am using in the SETI@Home Project. I used to show several examples so you could compare and contrast. However, the BOINC Web Site Software and other gremlins caused all of the Computer ID numbers to change. To the point where I surrendered. So, now I will just show you this one.

This page shows all of the relevant details about the computer including the name I gave it on my network. Also listed are things like the total memory size, disk space, etc.

You might be wondering why all of this information is gathered and tracked. Well, it is quite simple. When you are a Participant within a Project the Project Team does not want to send work to your computer that it cannot handle. In the case of this specific machine it has lots of RAM memory and a fast processor. But the disk size is not all that great by today's standards. But this was not intended to be a work machine so the size of the hard drive was not of much concern to me.

Other than the slightly small hard drive (the other reason that it was small is that it is a 10,000 RPM drive) this is still a pretty fast machine.

However, this small disk space may mean that this machine is not a machine that would handle a large Work Unit that needs plenty of disk space (we don't seem to have that problem with the BOINC Powered Projects that are currently issuing work, but that could change at any time).

[edit] "Computer Summary" Page As Visible To Other Participants

When the "Computer Summary" Page is viewed by any Participant that is not logged into the Account of the owner of that computer will only see a restricted amount of information about the computers within that Account.

As shown in this example, much of the data is simply not shown. You can validate this by logging out of your Account, then accessing your "Account Data" Page and then "drilling-down" to your computers.

Of course, you can always make your computers "hidden" so that other Participants cannot see them at all. But, when you ask for help, hidden computers takes away one of the pc technical support information tools that lets other Participants work on the problem.

[edit] Line by Line

Ok, with all that out of the way, let us now look to the wealth of information that is contained within these records.

The examples shown have all been taken from the SETI@Home Project, that is not to slight the other projects, but, if you have seen one example, well, you have seen them all. The only thing that changes is the color of the background and possibly some of the numbers …

[edit] IP Address (Private)

At any rate, if you click on the "Show IP address" link you will be shown a page that contains the IP Address of this computer. And a report of how many times this particular IP Address has been used to report for this specific computer. My IP Addresses are dynamically assigned, but, computers being the silly things they are, well, most of the time this is not going to change. Especially since I do not turn off my computers unless the lights go out. So, when the IP Address needs to be renewed, it is most likely going to be the same as it was before.

Note:

This information is only available when you are viewing your own computers. No other Participant can see your IP address. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Domain Name (Private)

For those computers that are in a domain, the name of that domain will be listed. In most cases this field will carry the name of the computer. For computers in a business network this field will likely be filled with the name of the domain within which the computer operates.

Since most of us using home networks use the simple (and in theory faster) workgoups, this field will be set to the name of the computer instead of the name of the workgroup.

Note:

This information is only available when you are viewing your own computers. No other Participant can see your domain name. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Local Standard Time (Private)

This field contains the time zone so that the Project Team will know when you are sleeping and when they can come by and pay you for the tooth under the pillow. I doubt that, but this does allow the project people to realize that if your computer is causing troubles that they may not get attention on the problem until later when you wake up. You can see a time zone conversion chart to see if this information is correct.

This information is used by Einstein@Home to select a download-server close to you.

Note:

This information is only available when you are viewing your own computers. No other Participant can see your time zone. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Name (Private)

The Computer Name is the name given by the Participant to the computer. This should be the "friendly" name assigned to the machine. If this is a pre-built machine the name may not be quite so friendly.

Since most people only have one computer this will be the name of the machine and you will have no problems at all in identifying your machine. But for those of us that have a "farm" of computers this allows us to "map" the computer to the real machine we have at home. And if your are tracking your computers at a remote location, you can look at the numbers (like the Recent Average Credit) and see which machine may warrant some attention.

In the example shown, you can see that I have names that reflect some of the characteristics of the computer.

Note:

This information is only available when you are viewing your own computers. No other Participant can see your [[Computer Name. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Created

This indicates the date on which the computer was introduced into the world of BOINC Powered Projects.

[edit] Total Credit

The Total Credit line shows a value that reflect the processing done over the "life" of the computer while processing Work Units within the BOINC Powered Project. This is the accumulation of the Total Credit earned by this specific machine.

Note:

If the computer has been "merged" or has just been created by the Scheduling Server (even if there is another computer of the same name in the Account), it is possible that this will not be the real "Total Credit" earned by the machine. In other words, the bookkeeping may be off a little. Your Total Credit for the Account will be unaffected by this error.

[edit] Recent Average Credit

The Recent Average Credit line shows a value that reflect the processing done (the number of "Cobblestones") for the preceding week or two by this specific computer. Note that if the computer has not been returning Results this value may be "stuck" showing an artificially high value.

This value is like a speedometer reading. See these Glossary entries for more details:

Recent Average Credit, Computation_of_Credit, Cobblestone, Whetstone_Benchmark, Dhrystone_Benchmark

[edit] CPU Type

The name of the type of the Central Processing Unit (CPU) that is installed in the computer. One of the reasons for keeping this data is the possible need to determine if some of the science that was done may have been compromised by the Physical Processor.

For Apple computers, you can look up the CPU name listed in this field and associate it to actual "CPU Type Names" using the "Apple CPU Type to Computer Type Cross-Reference Matrix".

[edit] Number of CPUs

This line lists the number of CPUs that are installed in this computer. The Intel® Hyper-Threaded processors will show up here with two processors. Most people that are not using the high-end computers at work will likely only have a single processor and this number will be one.

[edit] Operating System

This lists the Operating System that is being used along with other identification data indicating the exact version, Service Pack level, and "build" level. Each of the major Operating Systems will have their own format for the data. In the examples you will see information for at least the [[WP: Win}}, and Apple Mac OS X Operating Systems.

[edit] Memory

This line contains the information about how much Physical Memory is available on the computer. This can be a critical factor for some Science Applications because of the amount of room they need in Physical Memory (the new project Rosetta@Home is an example) space vs. that fairly standard 4 Giga-Bytes of Virtual Memory. If your system's memory is too small, then it is possible that the Science Application would not function effectively. On systems with too small an amount of Physical Memory would "thrash" paging data on and off the disk drive and slowing the processing of the Work Unit.

[edit] Cache

The Cache size, like the memory size may impact the efficiency of the Science Application as well. In this case, the memory to Cache to processor becomes the limiting factor. Knowing about the size of Cache available can be critical for some Science Applications. It is unfortunate that the storage is only one number in that modern processors have, usually, two levels of Cache and the level one cache can be separated into instruction (or micro-operations) and data cache.

We are also seeing the arrival of level 3 Caches in some of the very high end systems. So, better knowledge of the cache and its configuration may be important. This architectural issue may also allow the project the capability of sending special versions to the Participant's Computers to take the best advantage of the architecture. It could be as simple as the use of optimization "flags" on the compile and linking process, to the use of different compilers. For example, Microsoft's Visual Studio may make the best (fastest) version for AMD processors while the Intel® compiler does best for their processors.

[edit] Swap space

Because the Operating System almost certainly does not have enough Physical Memory to "back-up" the Virtual Memory space, it uses part of the disk drive to hold the contents of the pages of Virtual Memory that are not currently mapped into the Physical Memory space. Here again we have the situation where the size of the Work Unit may require more resource than is available on the Participant's Computer.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Total disk space

Depending on your Preferences in the BOINC System, this may be a limiting factor in the amount of space the projects can consume due to the "Use no more than x% of total disk space" value in your General Preferences.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Free disk space

The remaining space on the disk drive may preclude the storage of the Work Units that the project has to send. So this, like all of the other size parameters will potentially affect the work that the Data/Scheduling Server will send to the Participant's Computer]].

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Measured Floating Point Speed

This line shows the most recent measurement of the processor's capabilities to perform Floating-Point Operations (mathematical operations) using the "Whetstone" benchmark. The value listed is the result of the running the "Benchmark" program that rates the processor's speed. The speed of the processor's ability to perform integer mathematics is not measured by this Benchmark.

[edit] Measured Integer Speed

This line shows the most recent measurement of the processor's capabilities to perform integer mathematical operations using the "Dhrystone" benchmark. The value listed is the result of the running the "benchmark" program that rates the processor's speed. The speed of the processor's ability to perform Floating-Point Operations (mathematical operations) is not measured by this Benchmark.

[edit] Average Upload Rate

This is the latest measured upload rate established. Like many of the other parameters this allows the projects to screen the participants so that work that cannot be done by a computer is not issued the work. In this case, the issue is how fast can the Participant's computer move data into the project's Data Server. Some models, like the ones for Climate prediction.net, may result in data files that are tens of megabytes in size.

If the Participant is on a dial-up network, well, lets just say it is not practical.

This is kept separately from Download rate because many of the connections, particularly ones like DSL and Cable Modems, have different upload and download speeds. Normally, the upload speed is a fraction of the download speed. Which makes sense, most people retrieve data from the Internet, they only send small files up, but pull lots of files down …

[edit] Average Download Rate

This is the latest measured download rate established. Like many of the other parameters this allows the projects to screen the Participants so that work that cannot be done by a computer is not issued the work. In this case, the issue is how fast can the Participant's Computer get data from the Project's Data Server.

[edit] Average Turnaround Time

This is a measurement of how long it takes for a Work Unit/Result to be returned to the Project after it is sent to you. This value is related to the size of your Work Buffer, and is used in some Projects to determine which computers are assigned Work Units that are being re-sent.

[edit] Maximum Daily Work Unit Quota per CPU

This is the maximum number of Work Units that each CPU on this computer can download for a specific Project. This value decreases by 1 (one) each time your computer has a download error, a computational error, or fails to return a work unit by it's Result Deadline. The value doubles upon each properly returned Work Unit, up to a project specified maximum.

[edit] Results

This link in this line, ("'(number-of-results)'"), will connect to the "Results For Computer" Page that lists the Results that this machine has produced.

This number may vary over time with increasing values (normal expectation) and with decreasing values.

Now why would this value go down? Simple, to keep the on-line database small the BOINC Server Software runs a program called the DB Purge that keeps the database "lean and mean" by removing records that are "old" and are not really necessary.

This allows us to also remove old computers from our list of computers. Before a computer can be removed all of the Results associated with that computer must also have been removed.

[edit] Number of Times Client Has Contacted Server (Private)

This line contains the count of the number of times that the BOINC Manager has contacted the Scheduling Server. This is another predictive metric in that higher numbers of contacts means that the system is productive and may allow "tuning" of the amount of work that might be issued by the Project.

Of course, there is always the possibility that the reason that there is a high number of contacts is because of consistent communication errors, causing the abandonment of many Work Units that the Participant's Computer attempts to download.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Last Time Contacted Server

This is just a tracking value that shows when was the last time that this computer contacted the Scheduling Server.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit]  % of Time Client Is On (Private)

This is the percent of total time you run the BOINC Client Software, and is used when deciding on how much work (if any) to give you.

If for example you're running 24/7, this should be very close to 100%, while if you only run BOINC Client Software for 8 hours/day it will drop to 33%.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit]  % of Time Host Is Connected (Private)

This is the percent of time the BOINC Client Software is running it has got an Internet connection. For permanently connected clients this should be close to 100%.

Unfortunately, it sometimes also is close to 100% even if you're using dial up, so is not always meaningful.

Also remember, this function is currently only defined on Microsoft Windows®, so if you run Linux or Apple Mac OS X or another non-Windows-OS it will probably display -100%.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit]  % of Time User Is Active (Private)

This is the percent of time the BOINC Client Software is running, that you allow one or more BOINC Powered Projects to run, and is together with "% of time client is on", used in deciding how much work (if any) to give you.

If for example you run 24/7, this will be very close to 100%.

If for example you run the BOINC Client Software only 8 hour/day, but you crunch all the time you let the BOINC Client Software run, it will again be very close to 100%.

But, if you run the the BOINC Client Software 24/7, but use the Preference Setting "Do work only between the hours of" 20-06, meaning 10 hours/day, it will be 41.6%

Note:

In Version 4.19 and earlier the BOINC Client Software, the definition was % of total time, meaning was always smaller than "% of time client is on".

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] Average CPU Efficiency

New for v4.70 and later.

The fraction of CPU Time compared to real time a Science Application runs, used for calculating how much work to give out.

If for example it takes 2 wall-hours to get 1 hour of CPU Time (as displayed in BOINC Manager), this will be close to 0.5.

On computers only used for web-surfing, checking emails, and the occasional file-download and listening to music, it should be close to 1.

If for example you only run the BOINC Client Software at the same time you run CPU-demanding games, it can be close to 0.

[edit] Result Duration Correction Factor

New for v4.70 and later, the Result Duration Correction Factor was added to comensate for fluctuations in Benchmark performance.

The fraction of CPU Time a result takes, compared to the expected time based on Benchmark, used for calculating how much work to give out.

If for example a "perfect" benchmark, where the estimated time is 4 hours and it actually takes 4 hours, this value should be 1.

If the benchmark says 4 hours, but you're using an optimized application so it takes only 2 hours to crunch a wu, it will be 0.5.

Similar, if benchmark says 4 hours, but you've got a very slow CPU so it actually takes 10 hours, the "Result duration correction factor" will be 2.5.

[edit] Location (Private)

The locations, or Venues are:

• "Home"
• "School"
• "Work"

This is the Venue that the computer has assigned to get specific preferences that might override the "General Preferences". To change the venue, you just have to select one of the specific Venues "Home", "School", or "Work" and then click the "Update" button. Then the next time that the BOINC Client Software on your computer contacts the Scheduling Servers (or you force an "Update" to occur to make the contact, see the "File Menu" for details) this preference will be changed in your computer.

The control is a standard list box that you all know and love.

Note:

This information is only available when you are viewing your own computers. No other Participant can see this value. For an example of what others can see, look at the example in the "Account Data" Page.

[edit] "Merge This Computer" Link

This link takes you to a page where you can merge the computers (just to be confusing we call your computer a Host in some places) if the BOINC Software on a particular Project's Web Server creates several computer identities for a single computer. What you can do then is to go into the "Merge this host" page and fix their mistake.

There are some limits though. If the information about the machines is too different the "merger" will not work. One of the changes that will cause this is if you change the Operating System on the machine. At that point it would become a "new" computer as far as the BOINC Software is concerned.

You can merge two computers if, and only if, the text descriptions are identical including the CPU Type, CPU speed, and Operating System name and version. Changing any of these attributes will make the computers too different to be merged. (Only available on your own computers.)

[edit] "Delete This Host" Link

If you want to remove a computer from the list, for example due to the computer not being used any longer, or after a new host-id has been (wrongly) generated but with a different Operating System computer support service name or CPU-name so it can't be merged, you can do so here.

Rest assured, your total credit will not decrease even if you delete an old host.

You can only delete a computer after "Results" have dropped to zero. (Only available on your own computers.)

[edit] Notes, Oddities, and "Watch Out For This"

1. None

[edit] Also See

Project	Application	Deadline	Quorum
Climateprediction.net (CPDN)	hadsm3	347 Days, 5 Hours, and 20 Minutes	1
	Sulphur Cycle	150 Days, 5 Hours, and 20 Minutes	1
Einstein@Home	einstein	14 Days	3
LHC@Home	sixtrack	Variable, around 5 days	3
Predictor@Home	mfold	7 Days	3
	Charmm	1 day
Rosetta@Home	rosetta	7 or 28 Days	1
SETI@Home	setiathome enhanced	Variable, 4 to 60 days	3
SIMAP@Home	simap	10 Days	2
SZTAKI Desktop Grid	search	4 Days	4
World Community Grid	faah	unknown	unknown
	rosetta	unknown	unknown