Posts by Perle

61) Message boards : Number crunching : Boinc farms. (Message 11601) Posted 7 Dec 2005 by Perle Post: Thanks for the responses. Now I dont feel so bad. Not counting all the friends and family that run Boinc for me !! My personal dual core 4400 2 Barton 3200's 1 Hp Kayak dual slot 1 933mhz 1 Micron Netframe quad slot 2 xeon 550mhz I have gone thru countless system upgrades. Then started picking up older dual and enterprise class quad processor server machines. These things big heavy and most run on 208v power, so I upgraded my breaker panel to run these extra 220 circuits. Remember...there is difference between voltage and wattage. The garage had 4 of these big Proliant servers up and running. They were quad proc 550 mhz servers 1 or 2 gig of ram. Also a Micron quad server. I ran these for a while, it was obvious that basically one of my Barton core 3200's were still more productive than a single server. I had 3 other dual proc P3 800mhz servers also and a HP Kayak xu800 933mhz dual proc workstation, and a couple of other Athlon systems. Then the electrical bill came in and the wife dropped the hammer. 8] We agreed....no more than five systems. So I sold all the servers and older athlon boxes, and bought my dual core 4400. I kept the Micron and the Kayak.....they run on standard 120v, host my music server and are fun to play with. Still trying to figure out if the micron can be upgraded to 900mhz xeons. Still am thinking of getting just one more older server, like a Proliant 8500 or a PowerEdge 8450 in 900mhz. These are 8 processor xeon servers !!!!!!!!!!!
62) Message boards : Number crunching : Boinc farms. (Message 11578) Posted 6 Dec 2005 by Perle Post: How bad is your boinc farm??? I remember years ago seti farms were cropping up. So I started collecting older systems and mounting them on plywood boards and built a rack system from 2x4 lumber, cabling and power strips and extension cords strung everywhere...the garage was heated thru the winter. 8] I had nearly 27 systems running. All of them P2/P3 slot 1, K6/K7 systems, a few athlon 1 gig systems. These things did 2 maybe 3 S@H wu a day. Those are all gone now. So how bad is your farm ??
63) Message boards : Number crunching : View Computers page format (Message 11093) Posted 28 Oct 2005 by Perle Post: I noticed with the new format that the list is now by what computer connected last. Is there a way to switch the list back to the previous format? Most productive system at the top, then downwards in descending order. Thanks.
64) Message boards : Number crunching : Host corruption (Message 10891) Posted 25 Oct 2005 by Perle Post: Same here.....reset host location. Then updated and it cleared it. Problem is many of my machines are across town or completely out of state. Is it possible to do a remote update ?
65) Message boards : Number crunching : Host corruption (Message 10887) Posted 25 Oct 2005 by Perle Post: I have several that are corrupted. All are running the 4.45 boinc. All corrupted have contacted 10/25 Here is the list of corrupted hosts. 49433 44886 56397 35234 21147 35226 22192 35221 56399
66) Message boards : Number crunching : only 10 k WU's left (Message 9208) Posted 7 Aug 2005 by Perle Post: I realise this is prolly a flawed point of arguement and I did read most of the thread...but ! My resources are set at 50.25.25 / 3 days. LHC / SETI / Einstein at a 3 day cache respectively. Why must BOINC and these other ass-ociatied parameters be involved? Why cant I just crunch at the straight forward 50.25.25 / 3 ??? It just seems it should just be that simple. I understand people need flexibility to match their systems.
67) Message boards : Number crunching : Quad P3 Xeon....any suggestions ? (Message 8740) Posted 20 Jul 2005 by Perle Post: Ya, but you have modern Xeons. Mine are the older Slot 1 style. Single threaded is a bit of an overstatement......single stranded is prolly more appropriate ! To build a new dualie Xeon or Opteron box, not so bad, under $2000. Wait...um these would have to be water cooled, of course....okay $2500. I love my Koolance systems. 8]
68) Message boards : Number crunching : Quad P3 Xeon....any suggestions ? (Message 8720) Posted 20 Jul 2005 by Perle Post: Most of the systems I was looking at had plenty of ram 2-4 gig and several HD's if not a functioning OS. The geeky allure is the 8 xeon procs with nothing to do but crush numbers. That task is something that these proc's do well. But to balance the cost of the unit+shipping+electricity to operate 220v Old Iron Does not equate to the cost of a modern system, for example a AMD 64 X2 4400 or 4800 divided by WU credit production. I like the coolness of a 8 xeon box crushing away...but a modern Dual Core AMD is way more productive than 8 P3 xeon 700's. right? The quad 550 xeon is about a week away. I will share the numbers that are generated as a perspective.
69) Message boards : Number crunching : Fortran Errors ? (Message 8716) Posted 20 Jul 2005 by Perle Post: check dat.....I do have the files, where do I send them? Here is a sample 2005-07-19 18:42:01 [LHC@home] Unrecoverable error for result wjun1B_v6s4hvnom_mqx__19__64.277_59.287__4_6__6__35_1_sixvf_boinc26188_0 (The session was canceled. (0xf0) - exit code 240 (0xf0)) *UNHANDLED EXCEPTION** Reason: Access Violation (0xc0000005) at address 0x0042F342 read attempt to address 0x00000010 gurgle !
70) Message boards : Number crunching : Quad P3 Xeon....any suggestions ? (Message 8703) Posted 19 Jul 2005 by Perle Post: I agree Power consumption vs. WU ratio is a serious consideration. Picking up a new X2 4400 with a meg of L2 for each proc in relation to the power consumed by the big slow 4 or 8 way box.....would probably pay for itself. Especially with the 8 Xeon box requiring a 220v outlet.....thats like running your clothes dryer 24/7. But just for uber geekiness.......8 processors.....that rules almost as much as tater tots do !!! The 4 xeon box was cheap and its a 110v box. If it doesnt work out...there is always Ebay....somebody else would prolly want it. Purchasing it kinda helps out the other guy who has fallen on some hard times.
71) Message boards : Number crunching : Quad P3 Xeon....any suggestions ? (Message 8701) Posted 19 Jul 2005 by Perle Post: I just purchased for $150 a Micron NetFRAME Quad P3 550 xeon server. I got this mostly just to play with and for BOINC. I know that Micro$oft 2003 will work but would Linux be a faster OS for BOINC? I have never worked with Linux....but this seems the perfect machine for it. I have also seen 8 processor P3 xeon servers for under $500.......good for boinc? or just a waste of electricity (208 volt power consumption) versus a new AMD 64 X2 4400 system for about $850. Old age and treachery VS. Youth and vigor
72) Message boards : Number crunching : Fortran Errors ? (Message 8699) Posted 19 Jul 2005 by Perle Post: New install of windows....I went with the Borg edition this time. Boinc 4.45 running. No errors so far. Sorry Paul, I didnt review the forums prior to the install so I dont have any data relating to the problem. Just whats posted in the forums in regards to the fortran error.
73) Message boards : Number crunching : Fortran Errors ? (Message 8688) Posted 19 Jul 2005 by Perle Post: Boinc 4.25 ....I dont like how 4.45 allows Einstein dominance due to the report deadline issue. I did restart and re - install Boinc and the problem persisted. This is the only machine that is having this problem....so clearly a fresh install of WinDoze is in order. This is the only machine I have that is having these issues.
74) Message boards : Number crunching : Fortran Errors ? (Message 8673) Posted 18 Jul 2005 by Perle Post: I keep getting these fortran errors. jwe0019i-u The program was terminated abnormally with eception code EXCETION_ACCESS_VIOLATION. Fortran error summary jwe0019i u 1 7/18/2005 12:11:46 PM\|LHC@home\|Unrecoverable error for result wjun1B_v6s4hvnom_mqx__17__64.277_59.287__8_10__6__35_1_sixvf_boinc23366_1 (The session was canceled. (0xf0) - exit code 240 (0xf0)) Any ideas ?
75) Message boards : Number crunching : v64lhc.D1-D2-MQonly-inj-no-skew-1s6_8588.2693_1_sixvf_51.zip...??? (Message 7145) Posted 22 Apr 2005 by Perle Post: One thing I have noticed is how LHC is being very selective about which machines actually get work units. Lately, we can all agree the that WU's have been sparse. But usually my AMD64 3200 and my AMD 3200 Barton have been consistently getting work at a much higher rate than the rest of my machines. So my point is ..... only my fastest machines have been getting work. Untill day....everything got several new LHC WU's in the Q' ... woots for me.
76) Message boards : Number crunching : 4.64 and 4.66 units what should i do with them ? (Message 6981) Posted 13 Apr 2005 by Perle Post: Should I run them wait for 4.67 ? Or should I "reset" the project and discard the few units I do have ?
77) Message boards : Number crunching : WU for only on of my machines ??? (Message 6856) Posted 6 Apr 2005 by Perle Post: I have noticed that only 2 of my systems are getting WU from LHC. Naturally it is the 2 fastest machines in my inventory. I am just curious about this anomaly. One is a 3200 barton and the other is 64x3200.
78) Message boards : Number crunching : Hello - Anybody back here from christmas holidays? (Message 5576) Posted 14 Jan 2005 by Perle Post: It would be great for an update from CERN. My CPU's are getting bored just working on SETI. 8]
79) Message boards : Number crunching : comparision of computational power (Message 4708) Posted 31 Oct 2004 by Perle Post: Just as a point of comparison. My old dog PIII 650 will complete a wu in about 2hr 40 min My AMD atlon 2.0 will complete a wu in about 28 min woots
80) Questions and Answers : Preferences : how do I set up different cache for different machines (Message 4707) Posted 31 Oct 2004 by Perle Post: Thanks.....I put my slower machines on a "work" schedule of 3 days, and left my faster machines on a "home" 5 day schedule.

Previous 20 · Next 20

LHC@home