climateprediction.net home page
Posts by wateroakley

Posts by wateroakley

1) Message boards : Number crunching : Tasks by application = hoarding (Message 64529)
Posted 22 days ago by wateroakley
Post:
Science United has attached 846 computers to this project since 16 Feb 2019. Somebody else can check how many tasks they have in progress at the moment.
With a few minutes of spare time ... rough orders of magnitude:

For the last 30 days ... 847 devices have made contact.
The first 100 computers with a recent credit number, about 75% are M$ and 25% are Linux.
Scanning down the list for credits, a few WAH returns from M$ devices and six Linux devices with more than 1,000 recent credit.
These six Linux devices are trickling 13 tasks to credit with 11 other tasks in the wings.
More than 90% of the 847 do not appear to be doing anything useful, unsurprisingly for the 75% M$ cohort.

Since 2019 ... the user appears to have 4470 devices attached.
Scanning down the sorted list for credits, somewhere around 90% have not successfully returned task completions.
2) Message boards : Number crunching : Tasks by application = hoarding (Message 64428)
Posted 6 Sep 2021 by wateroakley
Post:
If you can't actually complete them in the next 2 weeks you should Abort them and let others run them.
Hmm. This i7-based PC is running four tasks in an ubuntu VM. The timescales range from about 7 days for a hadam4 to 31 days for a hadam4h. In the early days of CPDN, tasks took around 3 months on an Intel P4. I'm quite used to nursing the long deadlines, even though M$ up-chucked today, crashing all bar one of the VM tasks. I can't afford the luxury of a 10th generation AMD or Intel CPU machine. Not sure how many of us would do anything if you want us to abort the long hadam4h's or set the deadline at 2 weeks?

I'd rather not waste my time looking for ET.
3) Message boards : climateprediction.net Science : Climate change in the News (Message 64417)
Posted 30 Aug 2021 by wateroakley
Post:
'Washout' summer actually one of the hottest

The comments of Faceache demonstrate the public's lack of understanding of science with something like 70% of them citing their personal experience as disproving the data.
Mean 2020/21 temperatures in the CET observation record have been within a whisker of the blazing 1976 temperatures in England that anyone over the age of 50 remembers. The mean CET temperature across June and July 2021 has been hotter than 41 of the 45 years since 1976. The jury is out on August. Scarborough in early August was very variable, does that prove that anecdotes can trump the data?
4) Questions and Answers : Unix/Linux : Will start crunching with linux.. Advice starting up? (Message 64337)
Posted 11 Aug 2021 by wateroakley
Post:
Win10, Oracle VirtualBox 6.1 with ubuntu 20.04 VM.
Some tried and tested settings from our i7-3770 with 4 cores (8 threads) and 32GB memory,

Memory. Allow 3GB memory per cpu for CPDN and the VM OS. It may be more than some model types require, but others have suggested 3GB. I have no idea what future needs may be.
32GB physical memory lets you run a VM memory of 20GB and leaves enough room for Windoze to be happy.
We found that 16GB physical memory for four ubuntu VM cpus to run CPDN was too small. CPDN tasks were throttled by about 25%. It became unusable for heavy office applications (Word and Excel).
We tested a Q9650 cpu and 8GB memory with an ubuntu VM running 2 cpus, Yes it ran cpdn, but insufficient memory. Not worth the effort.

CPU. 4 physical cores, 8 threads. We run the ubuntu VM with 4 cpus.
Task manager reports that VirtualBox (running the VM & CPDN) takes about 55% of CPU.
We can run the rare Windows CPDN tasks on 2 or 3 cores in the Win10 host. Otherwise they run the Africa Rainfall tasks.
Trying to put all i7 cores into use (4 VM ubuntu threads with CPDN and 4 Win10 threads for CPDN) is not a good idea. CPDN tasks slowed down by circa 20% The throttle was probably from running out of L3 cache.

Disc. Put the ubuntu VM onto an SSD.
Allow at least 10GB of .vdi virtual disc per VM cpu. Use an expandable .vdi format. Minimum 40GB for 4 cpus.
Beware the generic VM youtubers, they wildly underestimate the disc size in their videos. My new friend is gparted.
If you want to avoid all the extra cpdn processing from continually hitting the data TBW on a large SSD, a 128GB SSD is an inexpensive option.

Good luck.
5) Message boards : Number crunching : Computation Errors (Message 64335)
Posted 11 Aug 2021 by wateroakley
Post:
Is this me or CP???
https://www.cpdn.org/result.php?resultid=22107426
This task was running on computer ID 1521341. i9-10980XE, Linux Mint with 36 cores in just 15GB of memory? 1521341 downloaded 36 tasks on 31/7 and 25 have aborted or crashed fairly quickly with a range of Signal(n) or SIGSEGV errors too.

3GB of memory per cpu is a figure discussed by others. 15GB of memory is way too small for CPDN to be happy with 36 tasks. With insufficient memory, the tasks will be 'swapped' all the time and may be exposed to soft errors in the disc I/O. (Discs will hide these soft errors from users until it's too late!). Task 22131978 (a CM3S) is running at less than half the sec/TS of our older i7, suggesting that tasks are being throttled. The 'Invalid Theta' error may simply be another emanation of the task aborts and crashes.

The Intel specs suggest that i9-10980XE (Core-X) maximum memory size is 256GB? Is 1521341 a Linux virtual machine and you could increase the virtual configuration of memory and reduce the number of cores? Good luck.
6) Questions and Answers : Unix/Linux : ARM application (Message 64306)
Posted 9 Aug 2021 by wateroakley
Post:
The program uses one of the SSE instruction sets.
I forget now whether it's SSE2 or SSE4.


I remember when this came in and older CPU's that didn't support SSE2 became obsolete as far as CPDN is concerned. Not sure whether SSE4 is a requirement now or not. Linux or other OS having SSE2/4 may be needed but the CPU needs to support the instruction set as well.


SSE2 was first supported around 2000 on Pentium devices as part of INTEL proprietary IA-32 architecture, a step beyond the late-1990's INTEL MMX instruction set. AMD provided an emulation of IA-32 instructions on Athlon devices. SSE2 was (IIRC) was the limit to use INTEL P4 and AMD Athlon devices for early CPDN contributors. IIRC the SSE2 instruction set (or emulation by the Hardware Abstraction Layer) of the target CPU is essential for CPDN.
7) Message boards : climateprediction.net Science : Observation-based early-warning signals for a collapse of the Atlantic Meridional Overturning Circulation (Message 64305)
Posted 9 Aug 2021 by wateroakley
Post:
Is anyone modelling the AMOC?
About the RAPID-AMOC Programme https://rapid.ac.uk/background.php
RAPID-AMOC (2014-2018) builds on the RAPID (2001-2007) and RAPID-WATCH (2008-2014) programmes and will extend the time series of the strength and structure of the Atlantic Meridional Overturning Circulation (AMOC) to 16 years.

Atlantic overturning: new observations and challenges 8-9 March 2021, Royal Society, London, UK
Scientific discussion meeting organised by Professor Meric Srokosz, Professor Penny Holliday and Professor Harry Bryden.
For more information see the meeting page on the Royal Society's website at https://royalsociety.org/science-events-and-lectures/2021/03/atlantic-overturning/.

AMOC & RAPID on climateprediction.net:

RAPID-RAPIT
The aim of this experiment is to assess the risk of the Atlantic meridional overturning circulation (AMOC) collapse in the coming century. The models issued to volunteers for this experiment are identified by the inclusion of HADCM3N in the model name. Motivation The Atlantic meridional overturning circulation (AMOC), often referred to as the thermohaline circulation, […]

RAPID-CHAAOS
Aim The aim of this project is to find a set of model versions that can shadow recent surface and subsurface ocean changes. Once this model set is identified the range of uncertainty in the Atlantic Meridional Overturning Circulation (AMOC) for these model versions will be assessed. Experiment design This experiment uses HadCM3, a coupled […]
8) Questions and Answers : Unix/Linux : ARM application (Message 64297)
Posted 9 Aug 2021 by wateroakley
Post:
Interesting though. My ARM device is running some flavour of Linux. Linux is Linux, what does the architecture of the CPU have to do with the matter?
There are far too many design, implementation, programming and compiling variables to assume that the CPU architecture won't matter. The programs we are running were aimed at super-computers. Support for ARMv8-A in Linux is from around 2012. It's a long while since I used the first generation 6502 assembler. The original ARM was hard-wired, no Microcode (the Hardware Abstraction Layer) that translates the executable program into the chip's reduced instruction set. Even it ran an executable in Linux, these RISC chips are aimed at low power devices and will have (IIRC) an external L3 cache typically from 256KB to 4MB. Without architecture targeted programming/compilation, plus deep cache, don't expect data and cpu intensive programs (like CPDN) to behave nicely on a RISC device.
9) Questions and Answers : Unix/Linux : Run Linux work units with Windows 10 WSL (Message 64213)
Posted 25 Jul 2021 by wateroakley
Post:
Virtual machines. If you are considering running a Linux virtual machine for CPDN using Windows 10 with Oracle VirtualBox 6.1 and ubuntu 20.04 VM, after four months ...

Memory. 16GB physical memory is the bare minimum. The usable VM memory will max out at circa 11GB, leaving 5GB for the M$ OS and bloatware. The Win10 i7-3770 16GB memory PC example, with ubuntu VM and four CPDN tasks, was unworkable for heavy office application users (excel/word/pdf) - the screens would regularly go blank. Resource Manager reported all physical memory was in use. With three CPDN tasks and 10.5GB VM memory it was just about workable, slightly better with two tasks. QED: more physical memory needed.

With 32 GB of physical memory this Win10 PC now runs VitrtualBox with the ubuntu VM and four CPDN tasks; plus heavy office applications and three of the very rare Windows CPDN tasks. The ubuntu VM config is 4 cpu (4 of 8 threads) and 20GB virtual memory. Resource manager reports 28GB physical memory in use, 55% cpu load for VirtualBox and 45% cpu for everything else.

I recall 3GB memory per CPDN task is required? If you don't do anything else on a 4 cpu (8 threads) PC, 24GB physical memory may be enough for running Win10 and VirtualBox with an ubuntu VM and 4 CPDN tasks. If you have a newer multi-processer PC then more memory will be needed,

Virtual Disc. The ubuntu VM will need about 10GB of virtual disc per virtual cpu. That's a virtual disc size of 40GB to run the VM and four CPDN tasks. Youtube videos with tutorials for creating an ubuntu VM in VirtualBox wildly understate the virtual disc storage that is required for CPDN. Schoolboy disc size mistakes in Linux can be fixed with Gparted. If you're worried about Total Bytes Written (TBW) on your expensive SSD, then a small (128GB) SSD is very cheap option.

Old machines, are they worth the effort? We created an ubuntu VM with CPDN on an older headless PC with Win10 and VirtualBox (Q9650 quad cpu, 8GB memory, 128GB SSD). Yes, CPDN worked with two tasks, but there's insufficient memory to run more than 2 tasks on a 4-cpu machine. Not much else would run happily either. QED: Don't bother.
10) Questions and Answers : Unix/Linux : Run Linux work units with Windows 10 WSL (Message 64146)
Posted 8 Jul 2021 by wateroakley
Post:
Running CPDN tasks with ubuntu VM under Virtualbox appears to be reasonably resilient to the unannounced Windows10 updates. Certainly better with the VM than the loss rate of CPDN tasks on native windoze from automatic updates. Since I lost one task on 14/4, five more updates have passed without loss. ubuntu VM has complained a couple of times when I restart the VM, with 'file error' issues, necessitating a manual reboot and restart.

I recently increased the physical memory to 32GB (from 16GB) and allocated 20GB memory to the ubuntu VM (from 10.5GB). Yesterday was the first unannounced windows update since the memory upgrade. No losses and the ubuntu VM did not complain after restarting. It's also got enough physical memory to stop windoze swapping out office applications/data or blanking the video to the screens when I'm trying to work on documents.
11) Message boards : climateprediction.net Science : Misconfigured Machine? (Message 64086)
Posted 26 Jun 2021 by wateroakley
Post:
24 processor computer with 1361 errors :(
https://www.cpdn.org/show_host_detail.php?hostid=1517679
https://www.cpdn.org/results.php?hostid=1517679
12) Questions and Answers : Unix/Linux : Incorrect Disk Space Available to BOINC (Message 64049)
Posted 12 Jun 2021 by wateroakley
Post:
I used the VirtualBox Virtual Media Manager to increase the initial 10GB limit to 20GB. VirtualBox recognizes that the limit is now 20GB. Neither Ubuntu or BOINC sees the change.


That is as expected. There are three steps:
1. The VM media manager will increase the disc partition size to 20GB, nothing else. More than 30GB virtual disc will be needed. After two tries, I ended up at 40GB.
2. To increase the ubuntu disc into the expanded VM partition, I use the Gparted utility.
3. Check and adjust the BOINC local preferences (disc and memory) to allow BOINC/CPDN to use the increased ubuntu disc space.

Note 1. These three steps worked for me with a dynamically allocated VM disc using Oracle VM Virtualbox 6.1. I've not tried it with a fixed disc size.
Note 2. There is a lot of stopping ubuntu and rebooting (starting) the ubuntu VM throughout the process steps. It's quite logical when you think about the steps.
Note 3. Always suspend BOINC activity before playing with the partitions and discs.
Note 4: With WIN10, Oracle Virtualbox,6.1 and ubuntu 20.04 VM, 16GB RAM is barely enough to run four CPDN tasks. 3GB for each CPDN task is the minimum recommended elsewhere.
Note 5. Allocating more than 11GB memory for the VM (physical RAM 16GB) ran into the buffers of WIN10.
Note 6. With a 40GB virtual disc, BOINC and four concurrent CPDN tasks take up 15GB, ubuntu (et al) takes up 12GB, about 10GB free to BOINC, and 1GB free but not available to BOINC.
Note 7. For heavy users of other software the VM ubuntu and BOINC/CPDN software can be a less 'friendly' combination than running BOINC/CPDN under a native OS setup. I routinely run into blank screen issues whilst chrome excel and word are heavily used. The answer (for me) is to stop the ubuntu VM and close Virtualbox.
13) Questions and Answers : Unix/Linux : Shutting down for re-boot. (Message 63934)
Posted 4 May 2021 by wateroakley
Post:
Hi.

I've recently started running CPDN tasks on my Linux box again. Am I correct in remembering that HadAM4/N216 crash if I reboot the system? It would be nice to know.

Thanks.
Richard
Running ubuntu 20.04 under Oracle VM VirtualBox on a Windoze10 machine. Before shutting down the ubuntu VM or Windoze, I always suspend the CPDN/BOINC activity.
When Windoze decided on it's last unexpected update and reboot, I lost one Hadam4h (RIP) of the four CPDN tasks running on the ubuntu VM.
14) Questions and Answers : Windows : Lack of WU's (Message 63891)
Posted 19 Apr 2021 by wateroakley
Post:
Why are there not more WU's that can be run on Windows machines? Relatively few people have Linux, yet the vast majority of WU's are for Linux only. Surely you are reducing the possible processing of data by restricting the use of Windows machines. I personally have not been able to process data for this project for about four months now due to the lack of WU's.

In a similar M$ situation here, the alternative is to run a virtual machine to host LINUX. It's a bit of a faff for first time Windows users, but I've finally got Oracle VM VirtualBox running and Ubuntu. Good luck.
15) Message boards : Number crunching : Little work, yet the most "important" thing in the world? (Message 63852)
Posted 12 Apr 2021 by wateroakley
Post:
45 minutes to install and set up a VM, Ubuntu and BOINC? Phew!

After fighting too many computing OS and layers for too many years, I had settled on an easy life using M$. With time on my hands in this pandemic year, the plan to run Linux/BOINC/CPDN on a Windows VM 'host' was testing my patience. Maybe I'll take a look at ubuntu mate next time?
16) Message boards : climateprediction.net Science : Alpine Ice, Lead Pollution AD 1170 -1216 (Message 62273)
Posted 1 Apr 2020 by wateroakley
Post:
Cambridge University Press, Antiquity:
Alpine ice and the annual political economy of the Angevin Empire, from the death of Thomas Becket to Magna Carta, c. AD 1170–1216
High-resolution analysis of the ice core from Colle Gnifetti, Switzerland, allows yearly and sub-annual measurement of pollution for the period of highest lead production in the European Middle Ages, c. AD 1170–1220. Here, the authors use atmospheric circulation analysis and other geoarchaeological records to establish that Britain was the principal source of that lead pollution. The comparison of annual lead deposition at Colle Gnifetti displays a strong similarity to trends in lead production documented in the English historical accounts. This research provides unique new insight into the yearly political economy and environmental impact of the Angevin Empire of Kings Henry II, Richard the Lionheart and John.

Link: https://www.cambridge.org/core/journals/antiquity/article/alpine-ice-and-the-annual-political-economy-of-the-angevin-empire-from-the-death-of-thomas-becket-to-magna-carta-c-ad-11701216/C01D8EA75B0D2A7DFC8FAFFD7E3BB7C8#fndtn-information

The BBC (and other media) report the research in very different way for the general public:
Thomas Becket: Alpine ice sheds light on medieval murder

Link: https://www.bbc.co.uk/news/science-environment-52095694
17) Message boards : climateprediction.net Science : Climate change in the News (Message 62232)
Posted 14 Mar 2020 by wateroakley
Post:
THE ROYAL SOCIETY

Public newsletter

Is there a point at which more CO2 will not cause further warming?

Climate change: evidence and causes
View our updated content Climate Change answers to key questions including:
How do scientists know that recent climate change is largely caused by human activities? How fast is sea level rising?
And is there a point at which adding more CO2 will not cause further warming?

http://newsletters.royalsociety.org/c/11Oc9PvXHwWIlmE3TLeChmKJ
18) Message boards : Number crunching : Upload failures (Message 60902)
Posted 10 Sep 2019 by wateroakley
Post:
Blimey, the last wah2_cam25 zip upload, that has been lurking for weeks, has finished.

08-Sep-19 9:05:20 PM | climateprediction.net | Finished upload of wah2_cam25_a07q_200405_18_691_011369827_0_r1430190651_6.zip
19) Message boards : Number crunching : Project going off line (Message 60248)
Posted 6 Jun 2019 by wateroakley
Post:
Hello, just taken a short break from a diy destruction day to order some materials on-line … Four completed tasks reported and one new task downloaded. :)
20) Message boards : climateprediction.net Science : Climate change in the News (Message 58765)
Posted 16 Sep 2018 by wateroakley
Post:
I am not sure but I suspect someone has done a correlation between (ocean) temperature and frequency and strength of hurricanes - but I guess it would be put down as fake news.

The GDFL provides statistical information on the relationships between SSTs and Atlantic hurricanes:
https://www.gfdl.noaa.gov/global-warming-and-hurricanes/

The US EPA Climate Change Indicators: Tropical Cyclone Activity. Figure 3 represents the power dissipation index (PDI):
https://www.epa.gov/climate-indicators/climate-change-indicators-tropical-cyclone-activity


Next 20

©2021 climateprediction.net