Message boards :
Number crunching :
New work discussion - 2
Message board moderation
Previous · 1 . . . 31 · 32 · 33 · 34 · 35 · 36 · 37 . . . 42 · Next
Author | Message |
---|---|
Send message Joined: 5 Aug 04 Posts: 1120 Credit: 17,195,460 RAC: 2,697 |
I got a UPS. I was then surprised to find it was constantly adjusting the voltage, as it kept getting too high. It was outside the legal specs, and I reported it, but they said if they lowered my voltage (I'm next to the transformer), the far end of the street would be too low. Long ago, I used to get a big voltage dip every weekday morning around 9AM. Like from 117VAC down ro 105 VAC. My UPS could handle that. I used to joke that the people across the street were riunning an aluminum refonery in their garage even though the neighborhood is zoned residential.. The entire neighborthood was built in the 1950s although there were a very few houses built before that. Down the streets was a single power line at 2400 volts with a transformer that dropped it to 220 volts with a ground in between. Some people called that split phase. Somewhere around the year 2000 they rewired the entire neighborhood with 4800 (or something like that) volts three phase. Each of the little streets got one of those phases. We all got new transformers of course. Our power got much better after that. There was a time when one of the phases of my 220 volt power got too high. I called the power company and they replaced the meter. They said my old meter developed high resistance in the ground lead, and it was easier to replace the meter than to fix it. It did fix the problem. |
Send message Joined: 9 Oct 20 Posts: 690 Credit: 4,391,754 RAC: 6,918 |
Long ago, I used to get a big voltage dip every weekday morning around 9AM. Like from 117VAC down ro 105 VAC. My UPS could handle that. I used to joke that the people across the street were riunning an aluminum refonery in their garage even though the neighborhood is zoned residential..This area is zoned residential, but there's a tradesman, a gardening company, a mobile tyre fitting service, and a taxi company, all running out of their own homes. technically they're not allowed, but they're not disturbing anyone, so they get away with it (for decades). Perhaps I have a more sensible (or lazy!) council than most. In Hull (north of England), someone dared to make his own driveway so he could park his electric car next to his house to charge it. The council said you can't do that (even though many other houses had driveways, so no logical thought went into that), when he continued using it they (illegally!) put in bollards to stop him driving his car out of his driveway! He should have taken them to court for theft of his car. The entire neighborthood was built in the 1950s although there were a very few houses built before that. Down the streets was a single power line at 2400 volts with a transformer that dropped it to 220 volts with a ground in between. Some people called that split phase.Sounds low. We have 11kV to 240V transformers, although we have one transformer for 100 houses. Not sure which is most efficient. You're using a higher voltage along the street, but you have more transformers. If I designed it, I'd probably use 11kV like the UK, but individual transformers like the US. Less voltage dips, less heating loss in lines. I guess it depends on how much wire, transformers, etc cost compared to how much heating loss you get. There was a time when one of the phases of my 220 volt power got too high. I called the power company and they replaced the meter. They said my old meter developed high resistance in the ground lead, and it was easier to replace the meter than to fix it. It did fix the problem.Odd, meters here couldn't do that. The neutral passes straight through it along a thick copper bar. It simply tacks onto it to power itself and to work out the power factor, which residential customers are not charged for, so it needs to charge us less if we draw current out of synch with volts. The live goes through a thick shunt to measure the tiny voltage drop across it to read the current. If that bit went wrong it would either melt the meter or charge you stupidly high amounts and you'd notice. |
Send message Joined: 5 Aug 04 Posts: 1120 Credit: 17,195,460 RAC: 2,697 |
Sounds low. We have 11kV to 240V transformers, although we have one transformer for 100 houses It is so long ago that this was done. Perhaps I lost the decimal point. In my neighborhood, we have one transformer for 3 or 4 houses. |
Send message Joined: 9 Oct 20 Posts: 690 Credit: 4,391,754 RAC: 6,918 |
It is so long ago that this was done. Perhaps I lost the decimal point.Poor thing, locate it immediately, it's lonely out there. In my neighborhood, we have one transformer for 3 or 4 houses.With less load per transformer, don't you get a bigger % drop for something big switching on in one of the three houses? |
Send message Joined: 1 Jan 07 Posts: 1058 Credit: 36,603,015 RAC: 15,658 |
Started to look at the "interrupted upload" theory for the uploads which won't complete. I simulated a network fault by the simple expedient of pulling out a network cable half way through an upload. The first upload retry after that gave me a "file locked by file_upload_handler", which is normal. So I've backed off the file for a couple of hours by triggering multiple failures, and I'll give it another try after the lock timeout has expired. Not sure how long that is, but It'll be worth trying again after another hour. |
Send message Joined: 9 Oct 20 Posts: 690 Credit: 4,391,754 RAC: 6,918 |
Ask in LHC what they do, they have continuable transfers (I think both ways). |
Send message Joined: 15 May 09 Posts: 4531 Credit: 18,703,882 RAC: 16,510 |
With respect to the uploads, I got his from Andy. Hi, Which is I guess why all uploads to that server are currently not working. |
Send message Joined: 1 Jan 07 Posts: 1058 Credit: 36,603,015 RAC: 15,658 |
Which is I guess why all uploads to that server are currently not working.But they are working. The previous one from the test machine was: 13/07/2023 04:31:26 | climateprediction.net | Started upload of wah2_nz25_21ho_209505_25_995_012222330_0_r1885463224_5.zip 13/07/2023 04:32:24 | climateprediction.net | Finished upload of wah2_nz25_21ho_209505_25_995_012222330_0_r1885463224_5.zip |
Send message Joined: 15 May 09 Posts: 4531 Credit: 18,703,882 RAC: 16,510 |
But they are working. Not right now for me they aren't 13/07/2023 10:40:59 | climateprediction.net | Started upload of wah2_eas25_a0yo_199311_25_994_012216686_0_r1176625585_13.zip 13/07/2023 10:41:01 | climateprediction.net | Temporarily failed upload of wah2_eas25_a0yo_199311_25_994_012216686_0_r1176625585_13.zip: connect() failed 13/07/2023 10:41:01 | climateprediction.net | Backing off 00:38:31 on upload of wah2_eas25_a0yo_199311_25_994_012216686_0_r1176625585_13.zip 13/07/2023 10:41:02 | | Project communication failed: attempting access to reference site 13/07/2023 10:41:04 | | Internet access OK - project servers may be temporarily down. Hmm I can ping upload7. |
Send message Joined: 1 Jan 07 Posts: 1058 Credit: 36,603,015 RAC: 15,658 |
But on the other hand... 13/07/2023 11:16:46 | climateprediction.net | Started upload of wah2_nz25_21ho_209505_25_995_012222330_0_r1885463224_6.zipBoth of those two file sizes together add up to the indicated file size of 89 MB So both theories seems to be blown out of the water! But we are talking about different things - mine was an nz25, where yours was an eas25. |
Send message Joined: 22 Feb 11 Posts: 32 Credit: 226,546 RAC: 4,080 |
And it is upload11 |
Send message Joined: 15 May 09 Posts: 4531 Credit: 18,703,882 RAC: 16,510 |
And it is upload11Upload11 for the NZ tasks, 7 for the EAS ones. Edit: And I can confirm my NZ uploads are working normally. |
Send message Joined: 29 Oct 17 Posts: 1044 Credit: 16,202,110 RAC: 11,091 |
With respect to the uploads, I got his from Andy.With respect to Andy, it can't be the operating system version that's causing some uploads to fail. If that was the case we'd see far more upload problems. It seems to be specific to this host/user/file unless there are far more examples we're not aware of.Hi, |
Send message Joined: 15 May 09 Posts: 4531 Credit: 18,703,882 RAC: 16,510 |
I thought that was odd too but I am guessing the OS upgrade is why no uploads for the EAS tasks are even starting at the moment. |
Send message Joined: 15 May 09 Posts: 4531 Credit: 18,703,882 RAC: 16,510 |
As it is now getting close to 48 hours since Andy's email, I have sent one to check if a script needs to be restarted after the OS upgrade on upload7. (None of my EAS uploads are going anywhere whereas the NZ ones are all going through normally.) |
Send message Joined: 22 Feb 06 Posts: 490 Credit: 30,787,652 RAC: 11,328 |
interesting that I have a resend from the NZ batch which when looked it had failed after the first zip file with negative theta error. This has now got past the fourth zip on my machine!! |
Send message Joined: 29 Oct 17 Posts: 1044 Credit: 16,202,110 RAC: 11,091 |
interesting that I have a resend from the NZ batch which when looked it had failed after the first zip file with negative theta error. This has now got past the fourth zip on my machine!!Weather & climate models are very non-linear. Small differences in numerical calculations can quickly cause big differences in runs of identical code on different hardware (many, many, years ago this was a topic of my PhD). There are places in the code where just a single bit difference is enough. For example, for a cloud to form the air must be saturated, so the code computes the saturation at each grid-point and compares it to the value needed for a cloud to form. A single bit difference in that comparison is all you need to have, or, not have a cloud form. A cloud makes significant changes to its local environment. Differences in the numerics can come from different rounding in the processor, differences in numerical libraries the code might be linked to. Code errors that might be reading random memory locations can also cause small differences (maybe not enough to crash the model). There have been studies to look at this in the very early days of CPDN on the long-running climate models. |
Send message Joined: 22 Feb 06 Posts: 490 Credit: 30,787,652 RAC: 11,328 |
On checking there would seem to be little difference in the machines. The failure machine is an i! 3770 (3.4GHz) whereas mine is an i7 4790K (4.0GHz). Both have similar amounts of RAM and are running the same version on WIN10. |
Send message Joined: 9 Oct 20 Posts: 690 Credit: 4,391,754 RAC: 6,918 |
Looks like all the WAHs are now running smoothly. Can we have some more please? Or a date for those Linux tasks? |
Send message Joined: 15 May 09 Posts: 4531 Credit: 18,703,882 RAC: 16,510 |
Looks like all the WAHs are now running smoothly.I currently have 19 EAS uploads of around 118MB each queued waiting for upload7 to start working again. At least my NZ uploads are working. |
©2024 cpdn.org