March 2001 - Ag-tech - mailman3.cels.anl.gov

RE: [AG-TECH] Hardware news - towards a single machine node
by Jason Leigh 31 Mar '01

31 Mar '01

Ivan attached are specs on how to put together a passive stereo display driven by a PC and a dual channel GeForce2 (AGAVE- Access Grid Augmented Virtual Environment). So far the sites that are building them are USGS, Access, NWU, U Michigan Geology. Using the CAVElib you can drive it in either Windows or Linux. We are also bundling the software as part of the AG CDROM. Jason ============================================================= Jason Leigh, PhD spiff(a)evl.uic.edu Electronic Visualization Lab (M/C 154) EVL Phone (312) 996-3002 University of Illinois at Chicago EVL FAX (312) 413-7585 851 S. Morgan St. Room 1120 SEO http://www.evl.uic.edu/spiff Chicago, IL 60607-7053 > -----Original Message----- > From: owner-ag-tech(a)mcs.anl.gov [mailto:[email protected]]On > Behalf Of Ivan R. Judson > Sent: Saturday, March 31, 2001 8:19 AM > To: Ag-Tech@Mcs. Gov > Cc: Jeff Ramacher; Todd Needham > Subject: RE: [AG-TECH] Hardware news - towards a single machine node > > > > Hey all, > > I'll be working in the next few weeks on a written definition of AG Node > Requirements. This will mostly be formalizing alot of the information we > already have. This will shake out what it is we expect the current 4 > machine node to do. Defining what this functionality is, and building the > API's for each component will let us explore this kind of idea without > impacting the current operating schedule of the AG (ie, we're way too > successful to have people randomly breaking nodes for experiments). > > One part of this is to define "Core Node Functionality" and > "Auxillary Node > Functionality". This is the line between what a minimum node is (the Core > Functionality) and what a cool far out node might include (microscopes, > milling machines, mood indicators, whatever). If you have ideas for what > you might augment a node with, please send those to me so I can > include them > as examples... > > Any thoughts? Ideas? Are you going to toss me off the ship yet? > > --Ivan > > PS -- Oh my, we might actually start version numbering! > > > -----Original Message----- > > From: owner-ag-tech(a)mcs.anl.gov [mailto:[email protected]]On > > Behalf Of Jeff Ramacher > > Sent: Friday, March 30, 2001 7:15 PM > > To: Todd Needham > > Cc: disz(a)mcs.anl.gov; Ag-Tech@Mcs. Gov > > Subject: Re: [AG-TECH] Hardware news - towards a single machine node > > > > > > I think it would be really cool (but probably way too expensive to > > develop) to move some of the AV processing off of the system CPU to a > > special purpose AG pci board or something like that. Does anyone know > > if there's a "vic in a chip" kind of thing already? > > > > Just my thoughts on a Friday afternoon... > > -- > > jeffr > > > > Todd Needham wrote: > > > > > > Given discussions regarding CPU requirements for a single > machine node, > > > thought you might like to know I saw a pre-production dual P4 (1.7GHz > > > Foster) from Dell today: 64 & 32 bit PCI buses, built-in 1394, onboard > > > fast Ethernet. The real news is that they expect to start shipping > > > these in May with a reasonable starting configuration under $4K. > > > Needless to say, 4 cameras and all those video streams are > not going to > > > be an issue. > > > > > > > > > > > > We also had discussions with Viewcast/Osprey about getting > P4-optimized > > > drivers for their capture cards and we're following up separately with > > > their development team next week. It would still be nice to > get all of > > > this running on a single proc P4 to keep the cost down even further. > > > > > > > > > > > > Todd Needham > > > > > > Manager, Research Programs Group > > > > > > Microsoft Research > > > > > > http://research.microsoft.com > > > > > > > > > > > > > > > > -- > > Jeff Ramacher (jeffr(a)mhpcc.edu) | (808) 879-5077, x264 > > Maui High Performance Computing Center (MHPCC) | fax (808) 879-5018 > > | http://www.mhpcc.edu > > > >

1 0

RE: [AG-TECH] Hardware news - towards a single machine node
by Gurcharan.S.Khanna＠Dartmouth.EDU 31 Mar '01

31 Mar '01

hi, you wrote---"Any thoughts? Ideas? Are you going to toss me off the ship yet?" we're behind you 100%. i've been trying to do this with limited support for some time now. and i know personally there are half a dozen others with similar aspirations/needs. it's made the difference between being able to participate at some level vs. being left out of the game entirely. and overall it's been a positive experience. so, formalizing something for this setup would be great. i think your idea of a core node and auxilliary node is great. we've been working on a similar model here at Dartmouth, as we experiment with different size and format events, and try to streamline our equipment and room configurations. i (and all the rest of us) look forward to seeing what you come up with. i'd be happy to contribute our (ongoing) experiences if that helps. -gurcharan ========== Gurcharan S. Khanna Associate Director Research Computing Dartmouth College http://www.dartmouth.edu/~rc/accessgrid --- "Ivan R. Judson" wrote: Hey all, I'll be working in the next few weeks on a written definition of AG Node Requirements. This will mostly be formalizing alot of the information we already have. This will shake out what it is we expect the current 4 machine node to do. Defining what this functionality is, and building the API's for each component will let us explore this kind of idea without impacting the current operating schedule of the AG (ie, we're way too successful to have people randomly breaking nodes for experiments). One part of this is to define "Core Node Functionality" and "Auxillary Node Functionality". This is the line between what a minimum node is (the Core Functionality) and what a cool far out node might include (microscopes, milling machines, mood indicators, whatever). If you have ideas for what you might augment a node with, please send those to me so I can include them as examples... Any thoughts? Ideas? Are you going to toss me off the ship yet? --Ivan --- end of quote ---

1 0

RE: [AG-TECH] Hardware news - towards a single machine node
by Ivan R. Judson 31 Mar '01

31 Mar '01

Hey all, I'll be working in the next few weeks on a written definition of AG Node Requirements. This will mostly be formalizing alot of the information we already have. This will shake out what it is we expect the current 4 machine node to do. Defining what this functionality is, and building the API's for each component will let us explore this kind of idea without impacting the current operating schedule of the AG (ie, we're way too successful to have people randomly breaking nodes for experiments). One part of this is to define "Core Node Functionality" and "Auxillary Node Functionality". This is the line between what a minimum node is (the Core Functionality) and what a cool far out node might include (microscopes, milling machines, mood indicators, whatever). If you have ideas for what you might augment a node with, please send those to me so I can include them as examples... Any thoughts? Ideas? Are you going to toss me off the ship yet? --Ivan PS -- Oh my, we might actually start version numbering! > -----Original Message----- > From: owner-ag-tech(a)mcs.anl.gov [mailto:[email protected]]On > Behalf Of Jeff Ramacher > Sent: Friday, March 30, 2001 7:15 PM > To: Todd Needham > Cc: disz(a)mcs.anl.gov; Ag-Tech@Mcs. Gov > Subject: Re: [AG-TECH] Hardware news - towards a single machine node > > > I think it would be really cool (but probably way too expensive to > develop) to move some of the AV processing off of the system CPU to a > special purpose AG pci board or something like that. Does anyone know > if there's a "vic in a chip" kind of thing already? > > Just my thoughts on a Friday afternoon... > -- > jeffr > > Todd Needham wrote: > > > > Given discussions regarding CPU requirements for a single machine node, > > thought you might like to know I saw a pre-production dual P4 (1.7GHz > > Foster) from Dell today: 64 & 32 bit PCI buses, built-in 1394, onboard > > fast Ethernet. The real news is that they expect to start shipping > > these in May with a reasonable starting configuration under $4K. > > Needless to say, 4 cameras and all those video streams are not going to > > be an issue. > > > > > > > > We also had discussions with Viewcast/Osprey about getting P4-optimized > > drivers for their capture cards and we're following up separately with > > their development team next week. It would still be nice to get all of > > this running on a single proc P4 to keep the cost down even further. > > > > > > > > Todd Needham > > > > Manager, Research Programs Group > > > > Microsoft Research > > > > http://research.microsoft.com > > > > > > > > > > -- > Jeff Ramacher (jeffr(a)mhpcc.edu) | (808) 879-5077, x264 > Maui High Performance Computing Center (MHPCC) | fax (808) 879-5018 > | http://www.mhpcc.edu >

1 0

Re: [AG-TECH] Hardware news - towards a single machine node
by Jeff Ramacher 30 Mar '01

30 Mar '01

I think it would be really cool (but probably way too expensive to develop) to move some of the AV processing off of the system CPU to a special purpose AG pci board or something like that. Does anyone know if there's a "vic in a chip" kind of thing already? Just my thoughts on a Friday afternoon... -- jeffr Todd Needham wrote: > > Given discussions regarding CPU requirements for a single machine node, > thought you might like to know I saw a pre-production dual P4 (1.7GHz > Foster) from Dell today: 64 & 32 bit PCI buses, built-in 1394, onboard > fast Ethernet. The real news is that they expect to start shipping > these in May with a reasonable starting configuration under $4K. > Needless to say, 4 cameras and all those video streams are not going to > be an issue. > > > > We also had discussions with Viewcast/Osprey about getting P4-optimized > drivers for their capture cards and we're following up separately with > their development team next week. It would still be nice to get all of > this running on a single proc P4 to keep the cost down even further. > > > > Todd Needham > > Manager, Research Programs Group > > Microsoft Research > > http://research.microsoft.com > > > > -- Jeff Ramacher (jeffr(a)mhpcc.edu) | (808) 879-5077, x264 Maui High Performance Computing Center (MHPCC) | fax (808) 879-5018 | http://www.mhpcc.edu

1 0

Hardware news - towards a single machine node
by Todd Needham 30 Mar '01

30 Mar '01

Given discussions regarding CPU requirements for a single machine node, thought you might like to know I saw a pre-production dual P4 (1.7GHz Foster) from Dell today: 64 & 32 bit PCI buses, built-in 1394, onboard fast Ethernet. The real news is that they expect to start shipping these in May with a reasonable starting configuration under $4K. Needless to say, 4 cameras and all those video streams are not going to be an issue. We also had discussions with Viewcast/Osprey about getting P4-optimized drivers for their capture cards and we're following up separately with their development team next week. It would still be nice to get all of this running on a single proc P4 to keep the cost down even further. Todd Needham Manager, Research Programs Group Microsoft Research http://research.microsoft.com

1 0

RE: [AG-TECH] Display Experiment: Different Video Cards
by Dave Pape 30 Mar '01

30 Mar '01

Ivan- As I said in the mail Jason forwarded, when I tried a PCI & AGP card together it just locked up the system. If you have any better success, let me know. I can send you the XF86Config files I used, if it'll help. Also FYI, my other notes & benchmark charts are all one level up from the URL in my earlier mail, at http://www.evl.uic.edu/pape/CAVE/linux/nVidia/. I'm also keeping a list of known multi-display options at http://www.evl.uic.edu/pape/CAVE/linux/multi-head.html. -Dave --------------------------------------------------------------------------- Dave Pape Electronic Visualization Laboratory, UIC pape(a)evl.uic.edu http://www.evl.uic.edu/pape/

1 0

UKy Wednesday Seminar: Abstract.
by Satish K. Vanimisetti 30 Mar '01

30 Mar '01

Hi, The abstract for nest week's seminar is up at: http://www.ccs.uky.edu/ccs/profiles/Spring01/wed/apr4.html Thanks Satish

1 0

RE: [AG-TECH] Display Experiment: Different Video Cards
by Bob Olson 30 Mar '01

30 Mar '01

> benchmarks or statistical info, but OpenGL games and screen savers look really > good, as well as the video channels from the AG lobby. What sort of losses (vic-reported %'s) do you see when you have a wallful of images up? (I'm chasing the performance problems with vic on PCI adapters again, after seeing very poor performance on a 1.4GHz P4 with a Permedia-based Appian quad-display card, which worked like a champ, btw). The same machine, with about the same number of streams up on a 1920x1200 or whatever the highest res was on the AGP card, just rocked. --bob

1 0

RE: [AG-TECH] Display Experiment: Different Video Cards
by Michael Gates 30 Mar '01

30 Mar '01

I've actually got two of the Inno3D Tornado GeForce2 MX dual PCI cards in a new Display machine that I'm testing with and it works fantastic. I'm running them in a Dell OptiPlex Gx110 w/ 256mb RAM, a single 833MHZ CPU running win2k. It also has an integrated AGP Intel video card that I can use as well, giving me 5 heads total. One thing that I noticed is the way that the Tornado card does the dual video. It double's it's display size instead of creating a separate video pane. IE, if you're running two 1024x768 monitors on one card, it will appear to Windows that you have a single 2048x1536 monitor. Which is fine, except that it will center things on the 'big' monitor causing splits like we used to see in multi-headed NT4. As long as you use them for the projected displays and move all your windows into them manually, it's not a big deal. I don't have any benchmarks or statistical info, but OpenGL games and screen savers look really good, as well as the video channels from the AG lobby. -Mike ====================== Michael Gates Information Technology Support and User Services mgates(a)sdsc.edu 858.534.8369 mgates(a)pager.sdsc.edu Room 265 ====================== -----Original Message----- From: owner-ag-tech(a)mcs.anl.gov [mailto:[email protected]]On Behalf Of Ivan R. Judson Sent: Thursday, March 29, 2001 8:45 PM To: Ag-Tech@Mcs. Gov Subject: [AG-TECH] Display Experiment: Different Video Cards Hey everyone, Just wanted to let you know I put out an order for the following cards: Inno3D Tornado GeForce2 MX CRT 32MB Dual Video PCI $145.00 Asus AGP-V7100 GeForce2 MX 32MB Dual Video AGP $196.57 The idea here is to have 4 heads, 2 on AGP, 2 on a single PCI slot. These are both reasonably modern chipsets. At least the performance on the nVidia based cards seem to have been the best of the best lately. If this works, and tests go well, I'll report such to you all so you can consider swapping graphics cards. The incremental cost would be ~$500.00, which I think is a reasonable cost to upgrade the graphics on the display box. If you strongly disagree or agree, please chime in -- it'd be a good data point. I will verify these cards work both under Windows 2000 and Linux. Stay tuned, more details to follow. --Ivan --- http://www.mcs.anl.gov/~judson Futures Laboratory Argonne National Laboratory

1 0

Re: [AG-TECH] Net Split?
by Robert Olson 30 Mar '01

30 Mar '01

At 10:21 PM 3/29/2001 -0700, Don Morton wrote: >So - is this somewhat "normal" behaviour? Well, unfortunately, yes, when multicast is broken in a weird way. >Is there anything we could >have done to rectify the situation? Snagging the network guy was the right thing to do. Snagging Bill Nickless would have been good too, since he may have been able to see some weirdness in the router state. The periodicity of the dropouts may be a clue; problems in mcast between network boundaries can show up that way as protocol timers go off and te like. > In some cases it was quite >humourous - Ah, you've got a good attitude for this :-). --bob

1 0