On Bandwidth (Jul 08 2009)

Message boards : Technical News : On Bandwidth (Jul 08 2009)
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4

AuthorMessage
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 916351 - Posted: 9 Jul 2009, 23:17:54 UTC - in response to Message 916346.  

Tweenday Two (Dec 27 2007)

[i]In reality, we have a 1GBit connection to the world via Hurricane Electric, but alas this is constrained by a 100 Mbit fiber coming out of the lab down to campus - it will take some big $$$ to upgrade that, which may happen sooner or later (as it would not only benefit us).

It will be very interesting to find out what campus finds. Might not need to dig any trenches at all -- he says hopefully as Murphy stands menacingly just out side his field of view.
ID: 916351 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 916352 - Posted: 9 Jul 2009, 23:26:00 UTC - in response to Message 916346.  

Bit Ceiling (Oct 08 2008) - Richard Haselgrove

I retract - I am certainly not an authoritative source. But I may have had an unconscious memory of

Tweenday Two (Dec 27 2007)

In reality, we have a 1GBit connection to the world via Hurricane Electric, but alas this is constrained by a 100 Mbit fiber coming out of the lab down to campus - it will take some big $$$ to upgrade that, which may happen sooner or later (as it would not only benefit us).

I mentioned how we seem to have a 60Mbit ceiling...

... We have gigabit all over our server closet (more or less - some older servers are 100Mbit going into the 1Gbit switch).

But I do remember quite clearly how Matt eventually found that the 60 MB cap was the internal processing limit of the CISCO routers then in use. Upgrading the routers got us to the effective 95 MB cap we have now.
ID: 916352 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 916398 - Posted: 10 Jul 2009, 2:18:31 UTC - in response to Message 916246.  

If it is political, that there isn't better comms to the building, maybe we need to shame the 'powers that be' into action.
A few thousand e-mails from the us users might wake them up. Or ask the students to have a demonstration on our behalf.

I think if we want to have a "demonstration" it'd be best to mail in a $5 bill rather than try to "shame" the administration into spending more money on what is clearly an under-funded project flying under the radar.

The minimum donation the Berkeley system supports is 10 U.S. dollars. Mailing in a $5 bill would create a bureaucratic dilemma, that's the worst kind.

That $5 amount has been mentioned several times recently, it is certainly tempting to think in terms of how little each participant would need to contribute to make this a well-funded project. But the world seldom works as we would prefer, I'm just trying to keep the record straight so practicalities aren't forgotten.
                                                                Joe
ID: 916398 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 916414 - Posted: 10 Jul 2009, 3:24:33 UTC - in response to Message 916398.  

If it is political, that there isn't better comms to the building, maybe we need to shame the 'powers that be' into action.
A few thousand e-mails from the us users might wake them up. Or ask the students to have a demonstration on our behalf.

I think if we want to have a "demonstration" it'd be best to mail in a $5 bill rather than try to "shame" the administration into spending more money on what is clearly an under-funded project flying under the radar.

The minimum donation the Berkeley system supports is 10 U.S. dollars. Mailing in a $5 bill would create a bureaucratic dilemma, that's the worst kind.

That $5 amount has been mentioned several times recently, it is certainly tempting to think in terms of how little each participant would need to contribute to make this a well-funded project. But the world seldom works as we would prefer, I'm just trying to keep the record straight so practicalities aren't forgotten.
                                                                Joe

$5 is a metaphor, I meant "donate."

I should know that when dealing with engineering-types that it is better to specify the amount to three decimal places.
ID: 916414 · Report as offensive
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 30608
Credit: 53,134,872
RAC: 32
United States
Message 916420 - Posted: 10 Jul 2009, 4:01:20 UTC - in response to Message 916414.  

If it is political, that there isn't better comms to the building, maybe we need to shame the 'powers that be' into action.
A few thousand e-mails from the us users might wake them up. Or ask the students to have a demonstration on our behalf.

I think if we want to have a "demonstration" it'd be best to mail in a $5 bill rather than try to "shame" the administration into spending more money on what is clearly an under-funded project flying under the radar.

The minimum donation the Berkeley system supports is 10 U.S. dollars. Mailing in a $5 bill would create a bureaucratic dilemma, that's the worst kind.

That $5 amount has been mentioned several times recently, it is certainly tempting to think in terms of how little each participant would need to contribute to make this a well-funded project. But the world seldom works as we would prefer, I'm just trying to keep the record straight so practicalities aren't forgotten.
                                                                Joe

$5 is a metaphor, I meant "donate."

I should know that when dealing with engineering-types that it is better to specify the amount to three decimal places.

Never decimal places, always significant digits! :)

ID: 916420 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 916424 - Posted: 10 Jul 2009, 4:09:41 UTC - in response to Message 916414.  

If it is political, that there isn't better comms to the building, maybe we need to shame the 'powers that be' into action.
A few thousand e-mails from the us users might wake them up. Or ask the students to have a demonstration on our behalf.

I think if we want to have a "demonstration" it'd be best to mail in a $5 bill rather than try to "shame" the administration into spending more money on what is clearly an under-funded project flying under the radar.

The minimum donation the Berkeley system supports is 10 U.S. dollars. Mailing in a $5 bill would create a bureaucratic dilemma, that's the worst kind.

That $5 amount has been mentioned several times recently, it is certainly tempting to think in terms of how little each participant would need to contribute to make this a well-funded project. But the world seldom works as we would prefer, I'm just trying to keep the record straight so practicalities aren't forgotten.
                                                                Joe

$5 is a metaphor, I meant "donate."

I should know that when dealing with engineering-types that it is better to specify the amount to three decimal places.


I prefer 6 digit precision myself.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 916424 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 916442 - Posted: 10 Jul 2009, 6:17:54 UTC - in response to Message 916420.  
Last modified: 10 Jul 2009, 6:18:08 UTC


The minimum donation the Berkeley system supports is 10 U.S. dollars. Mailing in a $5 bill would create a bureaucratic dilemma, that's the worst kind.

That $5 amount has been mentioned several times recently, it is certainly tempting to think in terms of how little each participant would need to contribute to make this a well-funded project. But the world seldom works as we would prefer, I'm just trying to keep the record straight so practicalities aren't forgotten.
                                                                Joe

$5 is a metaphor, I meant "donate."

I should know that when dealing with engineering-types that it is better to specify the amount to three decimal places.

Never decimal places, always significant digits! :)

Whenever I start thinking floating point (and money is generally floating point), I automatically shift to scientific notation, so 3 decimal places is 4 significant digits.
ID: 916442 · Report as offensive
Profile Astro-AL

Send message
Joined: 31 Mar 00
Posts: 18
Credit: 95,868,034
RAC: 80
United States
Message 916806 - Posted: 11 Jul 2009, 14:36:01 UTC - in response to Message 916153.  

I couldn't see where Matt responded to your comments. Have you heard anything about your suggestion.
ID: 916806 · Report as offensive
Profile Dr. C.E.T.I.
Avatar

Send message
Joined: 29 Feb 00
Posts: 16019
Credit: 794,685
RAC: 0
United States
Message 916822 - Posted: 11 Jul 2009, 16:15:34 UTC


> Thanks for the Postings Matt - as you can see - it's appreciated by all here @ the Project . . .

< Think-Tank [suggestion as mentioned to Eric Korpela] we should be able to fix practically everything that needs fixin' . . .

@ Berkeley - accolades to each of you for all that you do [as well, have done]




"I am he as you are he as you are me and we are all together . . ."




BOINC Wiki . . .

Science Status Page . . .
ID: 916822 · Report as offensive
Speedy
Volunteer tester
Avatar

Send message
Joined: 26 Jun 04
Posts: 1639
Credit: 12,921,799
RAC: 89
New Zealand
Message 917073 - Posted: 12 Jul 2009, 5:15:14 UTC
Last modified: 12 Jul 2009, 5:15:57 UTC

Thanks to the team that keeps our crunchers fed with work. Here's my thought on the comment about turning downloads off. 1st Turn data recorders off, split the tapes that have just been filled with fresh data to save trashing data collected. Is the gigabit Internet connection a flat rate or is it a set rate per petabyte? from the Current Budget fall 2008 it says that to get a 1Gbit connection to the lab it will be $80000. I'm not sure if this is the connection that the lab can only use 100Mbit of.
ID: 917073 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 917077 - Posted: 12 Jul 2009, 5:53:08 UTC - in response to Message 917073.  

Thanks to the team that keeps our crunchers fed with work. Here's my thought on the comment about turning downloads off. 1st Turn data recorders off, split the tapes that have just been filled with fresh data to save trashing data collected. Is the gigabit Internet connection a flat rate or is it a set rate per petabyte? from the Current Budget fall 2008 it says that to get a 1Gbit connection to the lab it will be $80000. I'm not sure if this is the connection that the lab can only use 100Mbit of.

SETI buys their bandwidth delivered to one of the internet exchanges (PAIX) so it is very inexpensive.

Campus takes care of delivering it from PAIX to SSL. The university-owned connection isn't gigabit all the way from PAIX to SSL.

Bandwidth is available second-by-second, 24 hours/day. Turning off uploads and downloads, or limiting them to certain days, is just wasting every second during those days.
ID: 917077 · Report as offensive
Chelski
Avatar

Send message
Joined: 3 Jan 00
Posts: 121
Credit: 8,979,050
RAC: 0
Malaysia
Message 917279 - Posted: 13 Jul 2009, 3:44:12 UTC

Thanks for the updates. It looks like it will be months rather than weeks before a permanent corrective action can be in place for the bandwidth issue.

Meanwhile, is it possible for an interim solution? Even a 50% uptime wireless will help a lot with the frequent backlogs that take days to clear these days (since "normal" days the line is at least 80-90% utilised, and during recovery maybe 20% is available to clear backlogs). Or for that matter, since the "non-SETI" line up the hill is only using like 10mbits/sec or so, to have a throttled limit of SETI traffic to flow through it, like ~50mbits/sec will be of great help to the project.
ID: 917279 · Report as offensive
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 30608
Credit: 53,134,872
RAC: 32
United States
Message 917296 - Posted: 13 Jul 2009, 5:06:45 UTC - in response to Message 917077.  

Thanks to the team that keeps our crunchers fed with work. Here's my thought on the comment about turning downloads off. 1st Turn data recorders off, split the tapes that have just been filled with fresh data to save trashing data collected. Is the gigabit Internet connection a flat rate or is it a set rate per petabyte? from the Current Budget fall 2008 it says that to get a 1Gbit connection to the lab it will be $80000. I'm not sure if this is the connection that the lab can only use 100Mbit of.

SETI buys their bandwidth delivered to one of the internet exchanges (PAIX) so it is very inexpensive.

Campus takes care of delivering it from PAIX to SSL. The university-owned connection isn't gigabit all the way from PAIX to SSL.

Bandwidth is available second-by-second, 24 hours/day. Turning off uploads and downloads, or limiting them to certain days, is just wasting every second during those days.

$50,000+ and $2375/month
http://ist.berkeley.edu/services/is/net/fiberoptic

Much better buy
http://ist.berkeley.edu/services/is/dc/colo

And just in case they didn't see it:
http://ist.berkeley.edu/services/catalog/datacenter/serverdonation

ID: 917296 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 917299 - Posted: 13 Jul 2009, 5:48:43 UTC - in response to Message 917296.  


SETI buys their bandwidth delivered to one of the internet exchanges (PAIX) so it is very inexpensive.

Campus takes care of delivering it from PAIX to SSL. The university-owned connection isn't gigabit all the way from PAIX to SSL.

Bandwidth is available second-by-second, 24 hours/day. Turning off uploads and downloads, or limiting them to certain days, is just wasting every second during those days.

$50,000+ and $2375/month
http://ist.berkeley.edu/services/is/net/fiberoptic

Much better buy
http://ist.berkeley.edu/services/is/dc/colo

And just in case they didn't see it:
http://ist.berkeley.edu/services/catalog/datacenter/serverdonation

This is worth reading:

Some things have changed, like replacing Cogent with Hurricane Electric (at a lower cost), but the concept is still there.

Note that your first link is for off-campus service, provided by Comcast cable, and the Space Sciences Lab is on-campus.
ID: 917299 · Report as offensive
Profile Gundolf Jahn

Send message
Joined: 19 Sep 00
Posts: 3184
Credit: 446,358
RAC: 0
Germany
Message 917311 - Posted: 13 Jul 2009, 7:38:15 UTC - in response to Message 917299.  
Last modified: 13 Jul 2009, 7:40:02 UTC

This is worth reading: http://www.mail-archive.com/setiathome@klx.com/msg01049.html.

I changed [img] to [url] tags ;-)

Gruß,
Gundolf
ID: 917311 · Report as offensive
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 30608
Credit: 53,134,872
RAC: 32
United States
Message 917313 - Posted: 13 Jul 2009, 7:44:34 UTC - in response to Message 917299.  


SETI buys their bandwidth delivered to one of the internet exchanges (PAIX) so it is very inexpensive.

Campus takes care of delivering it from PAIX to SSL. The university-owned connection isn't gigabit all the way from PAIX to SSL.

Bandwidth is available second-by-second, 24 hours/day. Turning off uploads and downloads, or limiting them to certain days, is just wasting every second during those days.

$50,000+ and $2375/month
http://ist.berkeley.edu/services/is/net/fiberoptic

Much better buy
http://ist.berkeley.edu/services/is/dc/colo

And just in case they didn't see it:
http://ist.berkeley.edu/services/catalog/datacenter/serverdonation

This is worth reading: http://www.mail-archive.com/setiathome@klx.com/msg01049.html

Some things have changed, like replacing Cogent with Hurricane Electric (at a lower cost), but the concept is still there.

Note that your first link is for off-campus service, provided by Comcast cable, and the Space Sciences Lab is on-campus.

Please read it more carefully
Service to provide fiber connectivity to a department building that is outside of the main campus perimeter.

SSL is outside the main campus perimeter. Check the campus map.
http://www.berkeley.edu/map/maps/large_map.html
This makes the $80,000 figure in the budget make sense.
ID: 917313 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 917398 - Posted: 13 Jul 2009, 17:58:59 UTC - in response to Message 917313.  


Service to provide fiber connectivity to a department building that is outside of the main campus perimeter.

SSL is outside the main campus perimeter. Check the campus map.
http://www.berkeley.edu/map/maps/large_map.html
This makes the $80,000 figure in the budget make sense.

I'm really not trying to call you out, or question you, nor do I work for IST or at the SSL.

The page says:
This is provided by Comcast on a case-by-case basis, depending on their ability to provide fiber in the vicinity of the building. If fiber is available, the cost will vary depending on the amount of construction (e.g. trenching, provisioning for conduit) required to get the fiber into the building.

I'm not sure, but I'm guessing based on Matt's posts, that Comcast does not have fiber near SSL.

Either way, Matt says that IST is doing the analysis to see what's available, and what is possible.

I also think that in general, Campus has treated SETI@Home pretty well. They've not only let them buy bandwidth off-campus (instead of forcing them to pay campus rates), but they've provided transport from PAIX to SSL.
ID: 917398 · Report as offensive
deac

Send message
Joined: 28 Jan 00
Posts: 9
Credit: 1,675,239
RAC: 0
United States
Message 917781 - Posted: 15 Jul 2009, 3:45:56 UTC - in response to Message 915855.  

You sound stressed, busy, and overly occupied. I have been in such positions, and yes, sometimes you need to pull the plug, reorganize, and come back on line with a refreshed vision, attitude, and better connection.

The constant demand that you push out work units can't possibly be healthy at all. Stop, reorganize, start stronger, begin anew with a better IT infrastructure, if that's what it takes, and you will be able to spend your days studying stellar clusters, instead of responding to complaints related to why 8500GT CUDA video cards overheat when they run version 36.09.09 vs version 36.09.10 graphic drivers...

Your dilemma is not a dilemma: you are trying to balance project participant's egos and complaints against real science; you are trying to please people that update that score daily, competitively so, against the real meaning of science behind the entire undertaking...You are in a sense engaging in a failing proposition, for the people crunching numbers on here could care less, if we discover a signal, they care much more about whether JoeBlo123's CUDA processor finished a workunit 5 minutes quicker than their SLI setup...

I don't envy you, it is a full time marketing job, or an HR job...However, while none of these number cruncher characters understand the statistics or the real science behind the enterprise, I can guarantee you, that they will be frustrated if their PC systems have nothing to do. And they will want to do something...They like SETI, it is part of them...

Shutdown for a month or two to upgrade your systems, do it. The people will return, for what you have going for you, is a most noble, exquisite, and promising enterprise, relating ONLY to what is the most fundamental and invigorating and puzzling question in our existence: " Are We alone"

P.S. I know the number crunchers aren't alone, or they think they aren't...I feel sorry for them...

Bye now, I must discover new primes...
ID: 917781 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 918003 - Posted: 15 Jul 2009, 13:19:28 UTC - in response to Message 917781.  

By golly, I think he gets it!
ID: 918003 · Report as offensive
Nicolas
Avatar

Send message
Joined: 30 Mar 05
Posts: 161
Credit: 12,985
RAC: 0
Argentina
Message 918310 - Posted: 16 Jul 2009, 1:07:14 UTC - in response to Message 916021.  

If they were to shutdown for a month to work on things. Wouldn't they need to plan that in advance for the wu's that would expire while out in the field?

No, I don't think you're getting the idea.

They'd shutdown for a month and maybe leave us with no work to crunch for almost a month. They wouldn't necessarily give us a month worth of work first. But during that time they'd do useful work instead of constantly fighting fires, and the project as a whole will be better when it comes back.

Contribute to the Wiki!
ID: 918310 · Report as offensive
Previous · 1 · 2 · 3 · 4

Message boards : Technical News : On Bandwidth (Jul 08 2009)


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.