[World Community Grid] Nieuws updates

[DrDrax]

Oh nice er komen wat wu`s binnen druppelen hier, zijn we weer up ?!

[wildhagen]

DrDrax schreef op dinsdag 2 augustus 2022 @ 16:04:
Oh nice er komen wat wu`s binnen druppelen hier, zijn we weer up ?!

Nope, helaas niet. Ze zitten nog in een testfase, dit is al weken het geval. Af en toe releasen ze weer een batch WU's.

Tot nu toe alleen OPN, MCM en ARP gezien, nog geen HSTB en SCC. En van OPN alleen CPU-werk, nog geen wu's voor GPU's.

[wildhagen]

Een volledige restart volgende week lijkt dichtbij te zijn:

Hi everyone, many volunteers have been asking about the continued drought of workunits. We have been working hard to resolve related problems, so we can restart the full scale of WCG, but, here are some challenges that we are still facing.

We have been working on several issues, and finally resolved some of them:

- fixed the issue whereby OPNG workunits were transitioning into an incorrect state due to repeated pre-emption by higher priority processes when no CPU resources were available to the scheduler.

- provisioned additional servers to resolve multiple issues stemming from insufficient vCPU and oversubscribed services pre-empting processes that would otherwise have completed normally.

- fixed all OPNG batches that had transitioned to a frozen/idle state due to the above going unnoticed for too long.

While technically, we can now push out new workunits, we ran into some network problems at our data center.

Unfortunately, vacation schedules of network engineers in the data center dictate that the earliest we might expect a resolution is August 8. Due to recent personnel changes the center does not have anyone on call to assist any earlier.

As a result of these circumstances and the current status of the WCG infrastructure, we should be able to start sending an appreciable number of new WUs for you to crunch on or shortly after August 8th, 2022.

We will update you early next week on the status of these issues, as we also hope that once we complete this work we will be able to fully restart.

Zie https://www.worldcommunit...g/viewthread_thread,44241

Laten we hopen dat het doorgaat!

[Tao]

wildhagen schreef op zaterdag 6 augustus 2022 @ 04:47:
Een volledige restart volgende week lijkt dichtbij te zijn:


[...]

Zie https://www.worldcommunit...g/viewthread_thread,44241

Laten we hopen dat het doorgaat!

Met augustus als populaire vakantiemaand denk ik dat september realistischer is...
Maar we gaan het zien.

[wildhagen]

Een volledige restart is het nog niet, maar er worden nu wel meer WUs uitgegeven, ik heb weer volle bak:



Wel veel download-errors, maar na een aantal (automatische) retries komen ze wel binnen. Mijn PC is weer aan het crunchen iig.

Er schijnt ook weer OPNG werk te zijn, maar ik heb die zelf nog niet gehad, alleen OPN WU's hier.

[fractus]

Ik heb ook de eerste WUs weer binnen, alleen MCM en de helft zit nog vast in de downloadqueue. Er lijkt weer iets te gebeuren, ben benieuwd

[wildhagen]

Ook OPNG lijkt nu, zij het mondjesmaat, binnen te komen.



Mijn buffers blijven nu best vol, wel met de nodige retries voor de downloads, maar toch.

[Tao]

Wow zeg, WCG is wakker geworden.
Bij PC opstart was ik gewend BOiNC weer uit te schakelen, maar ik zag - zonder op project update te hoeven drukken - 32 CPU taken en 1 GPU taak verwerkt worden.
Inderdaad OPN + MCM. Na die ene OPNG nog geen andere gehad.

Nu maar kijken of ze het vol gaan houden...

/Edit: ik krijg max 32 WU's (cpu). Geen WU's in de ingestelde cache.
/Edit 2: nu draait er slechts 1 WU. Zo schiet dat niet op WCG!!!

[ Voor 19% gewijzigd door Tao op 21-08-2022 19:34 ]

[KuuKe]

Ik zie nog geen punten is de stats van Jis verschijnen. Kan zijn dat ze nog aan het testen zijn en nog een stats vrijgeven. Kan ook zijn dat daar met de verhuizing eea aan is gewijzigd. Misschien dat @Jis er eens een blik op kan werpen?

[DaMoUsYs]

URL om de stats op te halen lijkt nog niet veranderd

[Jis]

Zo te zien wordt de Teams pagina nog niet bijgewerkt bij WCG:
https://www.worldcommunit...tsByTeamAT.do?sort=points
https://www.worldcommunit...T.do?sort=points&xml=true

De members pagina wel:
https://www.worldcommunit...points&teamId=025RSMBR9N1
https://www.worldcommunit...ordsPerPage=5000&xml=true

Het script blijft aantal uur proberen om de teams op te halen. Na verlopen van de retries wordt de rest alsnog uitgevoerd. Ik zie echter nog geen punten toegekend dusver.

[wildhagen]

Nieuwste update: https://www.worldcommunit...g/viewthread_thread,44273

Ik moet zeggen dat ik nu weer plenty WU's krijg. Wel veel transfer errors nog, maar na wat (automatische) retries komt alles wel binnen.

Heb ruim voldoende werk voor al mijn cores op dit moment. Grotendeels OPN, maar ook wat ARP en MCM.

[Tao]

wildhagen schreef op zaterdag 27 augustus 2022 @ 14:42:
................

Ik moet zeggen dat ik nu weer plenty WU's krijg. Wel veel transfer errors nog, maar na wat (automatische) retries komt alles wel binnen.

Heb ruim voldoende werk voor al mijn cores op dit moment. Grotendeels OPN, maar ook wat ARP en MCM.

Ik krijg alleen "download is stalled" berichten in m'n logboek.
Vanochtend mochten 14 WU's gedownload en verwerkt worden, waarna opnieuw "stalling".
Het gaat nog (lang) niet lekker daaro.

[wildhagen]

Tao schreef op zondag 28 augustus 2022 @ 12:26:
[...]


Ik krijg alleen "download is stalled" berichten in m'n logboek.
Vanochtend mochten 14 WU's gedownload en verwerkt worden, waarna opnieuw "stalling".
Het gaat nog (lang) niet lekker daaro.

Onder het kopje Transfer in de BOINC client kan je op Retry klikken. Als je niet op de automatische retry wil wachten.

In de meeste gevallen komen ze dan als nog binnen.

Ik heb nu 96 cores volop draaien, met een werkvoorraad voor 1.5 tot 2 dagen.

75 procent is OPN, 20 procent MCM en 5 procent ARP. HSTB heb ik nog niet gezien, maar dat werk was voor de migratie ook al schaars, dus dat zegt niet zoveel.

[m_p_g]

Tao schreef op zondag 28 augustus 2022 @ 12:26:
[...]
Ik krijg alleen "download is stalled" berichten in m'n logboek.

Hier ook een bestand van maar 436 bytes dat pas na xx retry acties een keer gedownload kon worden.

Het begin voor WCG is er, maar nog verre van stabiel bij relatief weinig server load.

[Tao]

wildhagen schreef op zondag 28 augustus 2022 @ 14:01:
[...]


Onder het kopje Transfer in de BOINC client kan je op Retry klikken. Als je niet op de automatische retry wil wachten.

In de meeste gevallen komen ze dan als nog binnen.

Ik heb nu 96 cores volop draaien, met een werkvoorraad voor 1.5 tot 2 dagen.

75 procent is OPN, 20 procent MCM en 5 procent ARP. HSTB heb ik nog niet gezien, maar dat werk was voor de migratie ook al schaars, dus dat zegt niet zoveel.

Ik retry me rot, maar alleen stalled WU's...
Nouja, dan maar wachten tot het wel gaat lukken. Ik laat BOiNC gewoon aan staan, maar reken er maar niet op dat e.e.a. verwerkt wordt. Dan valt het altijd mee.

[wildhagen]

De download problemen zouden nu opgelost moeten zijn. Heb al 36 uur geen enkele backoff of retry meer gehad.

[Tao]

wildhagen schreef op donderdag 1 september 2022 @ 08:53:
De download problemen zouden nu opgelost moeten zijn. Heb al 36 uur geen enkele backoff of retry meer gehad.

Ik nog regelmatig.
Ook krijg ik niet meer WU's dan ik realtime kan verwerken (dus geen cache gevuld krijgen).
Wel is het zo dat ik nu net zoveel WU's krijg als ik qua aantal threads kan processen.
Ze hebben nog steeds wat te fixen daaro.

[ Voor 10% gewijzigd door Tao op 02-09-2022 09:43 ]

[lrsv]

Downloads geven geen problemen meer hier.
Enkel spijtig dat 'runtime' niet meer wordt bijgehouden.

[wildhagen]

Er lijkt nog wel een probleem met validatie te zijn, met name bij ARP en OPNG.

Dan zijn door alle partijen de WUs ingeleverd, maar blijft hij in de Pending Validation Jail hangen.

Gebeurt alleen niet bij alle WU's van de genoemde projecten, sommige valideert hij wél gewoon goed.

Is door diverse mensen gemeld op het forum, zal nog wel naar gekeken worden.

En uiteraard de stats die nog niet werken maar dat was al bekend.

Heb al dagen geen enkel download probleem meer, dat loopt soepel nu. Ook de hoeveelheid werk is nu zodanig dat de BOINC client aangeeft geen nieuw werk meer op te halen: "new work not needed - cache full".

[wildhagen]

En hup, die downloadproblemen met 'transient http errors' zijn weer terug. Door meerdere mensen gemeld in het forum.

Ik begin een beetje klaar te raken met die onprofessionwle Krembil-toestand die kennelijk écht niet weten waar ze mee bezig zijn. Een totaal gebrek aan updates helpt hierbij ook niet mee.

[HenAadNel]

Zo is het niet leuk
1 downloadbestand blijft hangen op 4 uur

[wildhagen]

HenAadNel schreef op vrijdag 9 september 2022 @ 13:13:
Zo is het niet leuk
1 downloadbestand blijft hangen op 4 uur

Ja, ik denk dat dit er iets mee te maken heeft:



Hele site én forum is dus onbereikbaar nu (want gebruikt HSTS).

Alle downloads en uploads zijn ook onmogelijk geworden nu.

Amateuristische prutsorganisatie zeg.... begin er echt klaar mee te raken.

[ Voor 22% gewijzigd door wildhagen op 09-09-2022 14:17 ]

[HenAadNel]

Zo besloten om even alles te laten wat het is

Pending Validation even naar 0

[Tao]

Ik heb BOiNC maar weer voor onbepaalde tijd uitgezet.
Zolang er op deze wijze geklungeldhandeld wordt, is er voor mij de lol af.

[wildhagen]

Some good news to share before the weekend. Our data centre was able to fix some of the networking issues that should help us to stabilize the flow of workunits from the data centre to your devices. We ultimately aim to prevent the transient issues that have been the most commonly reported bug from volunteers during the testing phase and prevent further interruptions to the supply of workunits available to our volunteers.

Working with personnel at the data centre, our engineers were able to find a temporary workaround for the networking issue whereby migrating VMs to dedicated physical hosts with a previously configured virtualized interface to the VLAN; thus, the switch that provides access to the storage layer of the WCG was successful in connecting additional VMs to the storage layer. Oversubscribing some of these new VMs that will join the production Aurora/Mesos cluster to the physical resources of the bare metal hosts allowed us to fit the VMs, and will increase the overall capacity of the workunit processing backend by roughly ~45%. This will provide an additional upload/download server to help mediate the issues reported by volunteers in awaiting download/upload of workunits (and reduce the manual workarounds we had to do to make workunits available again).

Currently, we are running diagnostics on these servers, and will put them into service as part of the production Aurora/Mesos cluster as soon as we are finished. The oversubscription to the physical resources of the host should not affect the performance of the VMs, as the newly migrated VMs we are putting into service were co-located specifically with VMs that use those resources intermittently or sparingly (Build Servers, DevOps, internally hosted apps, etc.)

It may seem like a small advance - but besides directly helping with the workunit management - this will also free up some time for our tech team to finish other pressing challenges, and brings us closer to a more stable WCG.

Thank you for your support, patience and understanding.
- WCG team

Zie https://www.worldcommunit...g/viewthread_thread,44311

[Suicyder]

Het is niet veel maar wel iets, maar ik zie resultaten terug @ Results scherm (let op inloggen)

Zouden langzaamaan er dan nog weer stats komen

[wildhagen]

Suicyder schreef op woensdag 28 september 2022 @ 15:33:
Het is niet veel maar wel iets, maar ik zie resultaten terug @ Results scherm (let op inloggen)

Zouden langzaamaan er dan nog weer stats komen

Die pagina werkt hier bij mij al weken gewoon prima? Ik zie eigenlijk al mijn results daar, of ze nu In Progress, Valid of Pending Validation zijn,.

Alleen de "My Contribution"-pagina (https://www.worldcommunitygrid.org/contribution/overview) en alle history-gerelateerde pagina's werken nog altijd niet, niet meer bijgewerkt sinds 21 februari....

[Suicyder]

Bij mij waren dit de eerste resultaten welke aldaar kwamen te staan sinds februari. Zodoende mijn post

[HenAadNel]

er staat nu runtime enz enz voor 1 dag in
https://www.worldcommunitygrid.org/contribution/overview

[DaMoUsYs]

Stats doen het hier ook weer

[World Community Grid] hitparade van 29 september

Top 30 members daily
rank		points	results	name	points	results	rank
1.	()	2.666.365	691	vsral	161.193.256	211697	25	()
2.	()	1.602.143	1063	wildhagen	396.757.421	495824	10	()
3.	()	1.247.201	189	SpekAal	502.305.914	359590	8	()
4.	()	1.075.144	292	mklcln	71.145.803	69917	51	()
5.	()	960.107	230	Division Brabant	946.693.951	1630812	4	()
6.	()	825.923	337	M_P_G	2.045.984.003	1631321	1	()
7.	()	266.926	343	Team_Kwakende_kip	183.991.411	395962	23	()
8.	()	263.122	109	Rather21	40.118.001	53814	70	()
9.	()	202.601	61	Herman J	18.642.062	22759	110	( 1)
10.	()	175.060	234	DrDrax	195.921.808	288932	18	()
11.	()	161.411	93	geejay	22.350.487	36666	96	( 1)
12.	()	127.148	26	Asterix78	25.960.703	16341	90	()
13.	()	103.618	16	sander2774	19.296.322	26870	105	( 1)
14.	()	87.457	23	RedTape	9.488.407	4128	156	( 1)
15.	()	85.066	68	Wikkel	127.445.400	260832	34	()
16.	()	81.747	125	sjakie@ouweouwe	115.733.312	183678	38	()
17.	()	75.915	98	akiep	82.484.983	127570	44	()
18.	()	62.468	67	funz	29.145.091	41355	84	( 1)
19.	()	58.888	41	svol	80.492.908	141012	46	()
20.	()	56.895	67	Theadalus	212.209.479	621608	17	()
21.	()	52.672	36	Gizmo74	300.017.866	601955	11	()
22.	()	34.241	55	GJP80	48.728.486	73014	62	()
23.	()	32.509	39	SirJWH	33.239.666	23164	78	()
24.	()	30.530	40	Schizm	61.414.419	90186	55	()
25.	()	30.149	51	PcWes	15.789.472	26509	118	( 1)
26.	()	25.426	38	Prodix@gmail	25.426	38	464	( 17)
27.	()	23.905	28	Kenoobie	35.468.969	57491	75	()
28.	()	23.581	52	eridanos	167.413.516	279581	24	()
29.	()	22.761	41	Freecourt	55.801.581	97115	59	()
30.	()	20.749	25	rubendewolf	44.300.632	76074	66	()

Total flushers: 84
More...

Top 30 members overall
rank		points	results	name	points	results	rank
1.	()	2.045.984.003	1631321	M_P_G	825.923	337	6
2.	()	1.946.812.667	1365974	BlueTooth76 [DPC]	0	0	0
3.	()	1.528.228.367	2467534	WCW	0	0	0
4.	()	946.693.951	1630812	Division Brabant	960.107	230	5
5.	()	654.741.608	268233	[DPC] hansR	0	0	0
6.	()	532.659.490	411127	M-spec	0	0	0
7.	()	527.136.902	817573	DE KOMETEN	0	0	0
8.	()	502.305.914	359590	SpekAal	1.247.201	189	3
9.	()	404.545.186	643618	barreljan	0	0	0
10.	()	396.757.421	495824	wildhagen	1.602.143	1063	2
11.	()	300.017.866	601955	Gizmo74	52.672	36	21
12.	()	295.778.259	455247	DukeBox	0	0	0
13.	()	293.841.564	155127	HenkM	0	0	0
14.	()	255.411.120	417150	DPC_Oi!	0	0	0
15.	()	246.269.399	449244	Los Alcoholicos	3.285	6	54
16.	()	215.632.640	396260	BurnHard	0	0	0
17.	()	212.209.479	621608	Theadalus	56.895	67	20
18.	()	195.921.808	288932	DrDrax	175.060	234	10
19.	()	194.461.602	455629	depriens	15.730	33	33
20.	()	191.029.769	205831	hog555	0	0	0
21.	()	189.172.332	164697	Digiloog	5.773	11	44
22.	()	185.465.456	315994	dennishoog	0	0	0
23.	()	183.991.411	395962	Team_Kwakende_kip	266.926	343	7
24.	()	167.413.516	279581	eridanos	23.581	52	28
25.	()	161.193.256	211697	vsral	2.666.365	691	1
26.	()	152.897.535	324780	Glurp	0	0	0
27.	()	145.506.634	221478	Mechaniker	1.061	2	74
28.	()	144.310.493	203388	Racing_Reporter	0	0	0
29.	()	141.648.674	277041	evildutchman	0	0	0
30.	()	140.795.972	293097	ERBrouwer	0	0	0

More...

Teams daily
rank		points	name	points	rank
13.	()	14.085.476	Canada	16.286.016.477	26
14.	()	13.895.207	BOINC.Italy	10.645.119.012	43
15.	()	12.867.453	Rechenkraft.net	19.269.934.756	22
16.	()	12.186.473	IBM	252.355.776.682	1
17.	()	12.033.152	LinusTechTips_Team	7.545.554.204	53
18.	()	11.634.800	SETI.Germany	25.297.408.472	19
19.	()	11.061.879	Team China	40.993.410.625	11
20.	()	10.802.030	Dutch Power Cows	23.123.498.798	21
21.	()	10.496.867	L'Alliance Francophone	27.417.383.766	17
22.	()	10.200.630	Planet 3DNow!	19.194.387.514	23
23.	()	8.722.722	BOINC@Taiwan	27.924.554.781	15
24.	()	8.465.761	SETIKAH@KOREA	4.250.074.683	81
25.	()	8.315.779	Minnesota Crunchers	8.654.536.227	46
26.	()	7.507.926	SNURK and friends	10.015.842.428	44
27.	()	7.494.812	Decrypthon	27.421.234.001	16

More...

Teams overall
rank		points	name	points	rank
14.	()	31.680.832.983	HardOCP	2.244.839	78
15.	()	27.924.554.781	BOINC@Taiwan	8.722.722	23
16.	()	27.421.234.001	Decrypthon	7.494.812	27
17.	()	27.417.383.766	L'Alliance Francophone	10.496.867	21
18.	()	25.328.200.722	MIESNER Enterprises	0	0
19.	()	25.297.408.472	SETI.Germany	11.634.800	18
20.	()	23.183.690.138	Sony VAIO Team	71.216	883
21.	()	23.123.498.798	Dutch Power Cows	10.802.030	20
22.	()	19.269.934.756	Rechenkraft.net	12.867.453	15
23.	()	19.194.387.514	Planet 3DNow!	10.200.630	22
24.	()	18.647.643.843	Czech National Team	17.137.702	12
25.	()	17.034.919.990	UserFriendly.Org	6.465.451	32
26.	()	16.286.016.477	Canada	14.085.476	13
27.	()	16.015.692.106	Christians	5.111.980	42
28.	()	15.339.466.880	Rochester Institute of Technology	461.004	293

More...

Milestones
name	points	results	milestone
Division Brabant	946.693.951	1.630.812	946.000.000
Gizmo74	300.017.866	601.955	300.000.000
Theadalus	212.209.479	621.608	212.200.000
depriens	194.461.602	455.629	194.450.000
Digiloog	189.172.332	164.697	189.170.000
eridanos	167.413.516	279.581	167.400.000
vsral	161.193.256	211.697	160.000.000
jeroen_r	128.582.970	241.305	128.580.000
Wikkel	127.445.400	260.832	127.400.000
sjakie@ouweouwe	115.733.312	183.678	115.700.000
kuiphuis	69.370.157	116.937	69.370.000
Schizm	61.414.419	90.186	61.400.000
[DPC]Final Frontier	60.921.435	114.964	60.920.000
Freecourt	55.801.581	97.115	55.800.000
GJP80	48.728.486	73.014	48.700.000
rubendewolf	44.300.632	76.074	44.300.000
Rather21	40.118.001	53.814	40.000.000
Kenoobie	35.468.969	57.491	35.450.000
funz	29.145.091	41.355	29.100.000
rodikal	27.407.468	37.706	27.400.000
geejay	22.350.487	36.666	22.250.000
sander2774	19.296.322	26.870	19.250.000
Herman J	18.642.062	22.759	18.500.000
Karen1	13.244.325	20.652	13.240.000
willempm	12.200.152	19.462	12.200.000
Qlii256	7.845.053	13.829	7.840.000
Jasper Pol	7.480.517	6.008	7.480.000
vleesjus88	6.600.110	9.420	6.600.000
timterbeest	6.550.703	9.727	6.550.000
joeyboon	5.759.228	8.031	5.750.000
williamr1979	4.014.299	6.102	4.010.000
Sharaghnu	3.341.057	5.205	3.340.000
medivh_	3.211.867	6.287	3.210.000
fijnedagvan	1.802.033	3.000	1.800.000
SantaMuerte	1.518.667	2.491	1.510.000
Didier_Salembier	841.847	1.375	840.000

Links
World Community Grid homepage
World Community Grid stats van alle Teams
World Community Grid stats van DPC
World Community Grid stats van DPC voor BOINC clients
Bron DPCH

[wildhagen]

Mooi, dan kunnen we de wekelijkse stats-topics weer herstarten denk ik?

Helaas idd nog maar één dag history, vanaf 29 september. Alles van 21 februari tot en met 28 september lijkt leeg....

[KuuKe]

Als het blijft draaien (je weet het maar nooit ...) vanaf aanstaande maandag weer weektopics lijkt me.

[wildhagen]

Ik krijg net een mailtje dat WCG formeel herstart is, met onmiddellijke ingang.

[thunder7]

En prompt liggen alle servers er het hele weekend uit. Wat een amateurs daar.

[Frelg]

thunder7 schreef op maandag 3 oktober 2022 @ 08:31:
En prompt liggen alle servers er het hele weekend uit. Wat een amateurs daar.

Dit weekend ben ik ook weer begonnen. Nog wel moeilijke downloads (in behandeling), maar er komt wel wat door.
Inloggen echter ook niet, al een paar dagen. Zucht.

[thunder7]

Het ellendige is, dat hoe krachtiger de computer, hoe slechter het downloaden gaat. Ik heb hier een dual Xeon V4 staan met 80 threads in totaal, die komt aan werken niet toe omdat het downloaden steeds mis gaat. Een veel oudere machine met slechts 24 threads komt er nog wel af en toe doorheen, dus die heeft wel werk. Maar ze zijn inmiddels een jaar bezig, en dan nog je forum niet in de lucht kunnen krijgen/houden - ongelofelijk.

[m_p_g]

Het lijkt er sterk op dat IBM helaas niet de nodige support heeft geleverd aan de Krembil organisatie om zelfs maar van een een enigszins haperende transitie overgang te kunnen spreken.

De 1ste oktober was eindelijk, na maanden werk, vol goede moed het startschot gegeven:

Finally, with a functioning infrastructure and critical issues resolved we are ready to reboot the World Community Grid!

Inmiddels zijn de servers schijnbaar al in de reboot gecrashed....

[wildhagen]

Gisteren was alles weer UP qua site, forum etc, en uiteraard is alles weer DOWN nu, zelfde error als dit weekend.

Amateurisme ten top daar. Gisteren werd nog gemeld dat deze storing niet meer zou kunnen optreden. Nog geen 24 uur verder is die dus wéér opgetreden.

[Suicyder]

Ja het is nog huilen met de pet op, krijg een enorm bak aan tasks toegewezen, maar moet eerst hopen dat ze binnen gehengeld zijn voor de dealine

[Frelg]

Website worldcommunitygrid.org

System error
World Community Grid is currently experiencing an unexpected error. Please check Facebook or Twitter for more information.

En klikkerdeklik naar Twitter:

We posted an explanation for the earlier system error in our forum for you to read: https://worldcommunitygri...g/viewthread_thread,44362

Here is a short update for those confused about the recent Thank You emails and its implications on the WCG restart: https://worldcommunitygri...g/viewthread_thread,44381


En weer klikkerdeklik op de link naar worldcommunitygrid.org:

Error 500: javax.servlet.ServletException: net.myvietnam.mvncore.exception.DatabaseException: Error executing SQL in MVNForumPermissionWebHelper.getPermissionsForGroupGuest.

You had one job...

Uitleggen wat er mis is (gegaan) met de site via een link op Twitter naar de site zelf.

Het is bijna onmogelijk om voor te stellen hoe het daar werkt. Oh wacht... t werkt niet.

[wildhagen]

https://www.facebook.com/worldcommunitygrid.org/



No shit, Sherlock?

Minstens (!) 16 uur downtime is niet "brief"!

Als ik mijn baan zou uitvoeren zoals die lui bij Krembil, had ik allang mijn ontslag gehad, en was dat nog volkomen terecht ook.

[wildhagen]

En wederom lijkt de website down.... zucht....

[m_p_g]

En dat met een volgens Krembil functionerende infrastructuur.

Dan vraag ik mij wel met een groot vraagteken af, wanneer zij een infrastructuur niet vinden functioneren

[wildhagen]

Yeah, right. De website is al 20 uur lang down. Brief maintenance check, my ass!

[wildhagen]

En jawel, de website is wéér down. En aangezien Krembil in het weekend niet werkt, wordt dit pas maandagavond (onze tijd) opgelost....

[wildhagen]

Eindelijk weer eens een update, o.a. over de regelmatige downtime:

We are happy to announce that we have been able to increase the number of WUs available to volunteers. Global Stats updates are running normally and My Contributions page dashboard has been updated daily since the Thank You emails were sent. They are available for most users now, and we are resolving the last issues that will bring this to all volunteers. We continue working on updating all the stats and displaying forum streaks on the website, but they are stored in the database and reflected in the results tab of the My Contribution page.

Once again, a huge thank you to everyone for supporting WCG at this stage, submitting bugs and helping other volunteers in the forums. It is great to see an increased flow of results back to scientific partners. It is exciting to see that run time days for the preceding four weeks reached 1,160,151, and almost half of it (528,805) was achieved in the last week alone. Thank you.

Unfortunately, as the workload increased, we have encountered several system errors over the past two weeks. While we thought we knew the root cause and how to prevent the error in the future, we did not uncover the full complexity of what caused the error. Although we are closer to understanding the main cause, we continue to collect metrics during these events so that we can resolve it fully. Once we find a permanent fix for this issue we will provide more details in an update.

Aldus https://www.worldcommunit...g/viewthread_thread,44502

[Frelg]

(deel van de) Melding van 21 oktober

Access to the WCG website is restricted for a short time while we work to expand the storage capacity of the WCG. This should not prevent workunits from being sent out.

Op dit moment bijna geen WU meer (6). De machines staan idle dus. Hebben jullie hier ook last van?

[HenAadNel]

Frelg schreef op donderdag 27 oktober 2022 @ 12:39:
(deel van de) Melding van 21 oktober

Access to the WCG website is restricted for a short time while we work to expand the storage capacity of the WCG. This should not prevent workunits from being sent out.

Op dit moment bijna geen WU meer (6). De machines staan idle dus. Hebben jullie hier ook last van?

Dit is van TPCBF
Veteraan Cruncher
USA
MCM1 is nu uitgedroogd, met slechts een paar keer opnieuw en daar, en nu ook een paar uur geen nieuwe OPN1...

ander persoon
Al een paar uur geen OPN1-, OPNG- of MCM1-taak. Mijn computer is inactief.
Ik denk dat het team alle WU's in de WU-kluis heeft opgesloten, voordat ze naar huis gingen.

[m_p_g]

Frelg schreef op donderdag 27 oktober 2022 @ 12:39:
(deel van de) Melding van 21 oktober

.... This should not prevent workunits from being sent out.

This should not SHALL prevent workunits from being sent out.

Zo is het wat meer overeenkomstig de werkelijkheid.

[wildhagen]

Research update from the SCC team (October 2022)

Toch mooi om te zien dat WCG dus een concrete bijdrage levert aan het bestrijden van (in dit geval) kinderkanker.

[Tao]

wildhagen schreef op zaterdag 29 oktober 2022 @ 06:36:
Research update from the SCC team (October 2022)

Toch mooi om te zien dat WCG dus een concrete bijdrage levert aan het bestrijden van (in dit geval) kinderkanker.

Dat is mooi nieuws!
Jammer dat WCG er momenteel een t.....zooitje van maakt, anders zouden we meer kunnen bijdragen.
Ik ga met deze continue sores nog niet herstarten.

[wildhagen]

Tao schreef op zaterdag 29 oktober 2022 @ 11:41:
[...]


Dat is mooi nieuws!
Jammer dat WCG er momenteel een t.....zooitje van maakt, anders zouden we meer kunnen bijdragen.
Ik ga met deze continue sores nog niet herstarten.

Moet eerlijk zeggen dat het de laatste week wel aanzienlijk stabieler lijkt te zijn, sinds ze hun resources (db-server en up/downloadserver) hebben uitgebreid.

Veel minder transient HTTP errors, en áls ze er zijn, zijn ze meestal na één of 2 (automatische) retries ook weer opgelost.

Wel is er vaak alleen OPN1 werk, MCM1 en ARP1 is er niet altijd, en OPNG ook niet, maar heel erg vind ik dat nu ook weer niet. Als er maar iets te doen is.

Wat wel nog een minpuntje is dat de daily teamstats er niet meer lijken te zijn in de dagelijkse overzichten, maar dat vind ik een minor dingetje, het gaat mij vooral om het wetenschappelijke gedeelte.

[Tao]

wildhagen schreef op zaterdag 29 oktober 2022 @ 14:05:
[...]

.....

...., het gaat mij vooral om het wetenschappelijke gedeelte.

Same here.

[wildhagen]

OPNG workunits
Over the past 2 weeks many volunteers noticed the lack of OPNG workunits. The shortage is not a technical issue, but rather it has a “science-based” reason. Here is a short message the OPN team shared with us:

“The scientists are analyzing the available data to choose a set of protein conformations for the next rounds of molecular docking in GPUs. The focus is on PLPro, one of the SARS-CoV-2 proteases. The dockings will include covalent and non-covalent compounds.”

When new OPNG tasks will be prepared, we will make a small announcement informing everyone.

SSD delay
Unfortunately, the new storage upgrade that we mentioned in this previous update has been delayed by a late shipment of the SSDs, with no guaranteed ETA for implementation after arrival at this time. We are in constant communication with the SHARCNET team as to the status of the new drives, and will update you when significant progress has been made.

Missing devices in Contributions page
Several volunteers experienced missing devices and results in the Contributions page. This issue has been problematic for some time and we are now putting more effort into solving it. For the time being, we have scheduled a batch update from the BOINC database directly, bypassing the message passing via IBM MQ that notifies the website database when new devices have been inserted into the BOINC database. Insertion into the BOINC database occurs at time of first contact with our scheduler. The new record is then placed on a service-to-service queue, to which the webserver and website database server are subscribed. When processed, this message should result in visibility of the device in both databases and to users on the website immediately.

The fact that only some devices do not show up as a result of this process and remain “hidden” is what we must understand before we can fully resolve the issue. For now, re-running the batch update procedure regularly will help to recover visibility of most devices newly added to WCG. We highly encourage those having this issue to send an email to [e-mail] with screenshots and extra details you can provide.

Aldus https://www.worldcommunit...g/viewthread_thread,44843

Vooralsnog al tijden geen structurele down-/upload problemen, dat lijkt nu definitief verleden tijd gelukkig. Alles draait nu eigenlijk al weken behoorlijk stabiel.

[thunder7]

Ben ik de enige voor wie nu de hele site www.worldcommunitygrid.org down is?

Tussen 4:10 en 06:30 was het down, volgens mij.

[ Voor 27% gewijzigd door thunder7 op 15-12-2022 06:48 ]

[wildhagen]

Voorlopig geen OPN1 werk meer, dus alleen MCM1 met heel af en toe een ARP1 wu ertussen:

OPN1 is entering a planned pause while researchers prepare the next round of molecular docking experiments for OPN1 and OPNG projects

As mentioned in a previous update, OPN team is finalizing protein conformations for the viral protease PLPro, which will be the target for the next phase of OPN1 and OPNG molecular docking experiments sent out to volunteers. Once the new workunits are prepared and ready, we will resume sending both OPN1 and OPNG workunits to volunteers. Similarly to SCC - this is a planned pause. We will provide an update as soon as we have an ETA for the new workunits.

Zie https://www.worldcommunit...g/viewthread_thread,44862

[wildhagen]

WCG ligt weer plat, maar dit keer ligt het niet aan WCG (of Krembil) zelf: https://www.facebook.com/worldcommunitygrid.org/

Er is een grote systeemstoring bij University Health Network (UHN) Toronto, waar WCG wordt gehost: https://globalnews.ca/new...bvWiJnFK7UXD731UcybclbI5s

The University Health Network says it is experiencing an outage with its network.

In a tweet Monday just before 1 p.m., the UHN said it was working “to resolve the issue.”

By 2 p.m., the network’s website was still unavailable.

In a statement emailed to Global News, Gillian Howard, a spokesperson for the UHN said the network is “experiencing outages in our digital systems across our networks.”

“Our digital team is investigating the issues and will update periodically,” the statement read. “Clinical areas are using down time procedures at this time.”

It was not immediately clear what caused the outage.

The University Health Network includes Toronto General Hospital, Toronto Western Hospital, the Princess Margaret Cancer Centre, the Toronto Rehabilitation Institute and the Michener Institute.

In de log van de BOINC-client levert het iets in deze trant op:

10-1-2023 06:57:36 | World Community Grid | Requesting new tasks for CPU and AMD/ATI GPU
10-1-2023 06:57:45 | | Project communication failed: attempting access to reference site
10-1-2023 06:57:45 | World Community Grid | Scheduler request failed: Couldn't resolve host name
10-1-2023 06:57:46 | | Internet access OK - project servers may be temporarily down.

We kunnen dus alleen maar afwachten en hopen dat het snel is opgelost....

Edit: alles lijkt weer te werken nu qua upload.

[wildhagen]

Er is weer een update over de status van de projecten:

Mapping Cancer Markers: We have increased the number of threads assigned to creation of MCM workunits by about 60% on each workunit management server. As a result, the 7 day average of completed batches sent back to MCM1 research servers (10,000 workunits to a batch) has risen to 57 from 45, roughly a 25% increase. As we increased the number of threads in a two-step process and conducted maintenance on the storage server twice during the 7 day period over which the average is calculated, we do expect this number to increase further even without additional adjustment. We will continue to assess whether we can optimize it more (without interfering with the work currently underway at SHARCNET to resolve our network congestion ticket and move to the new, faster storage server).

Africa Rainfall Project: The ARP team is finalizing storage issues, and plans to resume downloading results from our servers, which will enable more workunits to be sent to volunteers. Workunits are being sent out at a slower rate due to the backlog of completed results. Further, errors in specific generations have caused some generations to lag behind. Together with the ARP team we have investigated the cause of these errors, and we determined that re-sending them after adjusting the granularity/time-step should resolve the problem.

OpenPandemics: As mentioned in a previous update, OPNG workunits are paused as the team is preparing a new set of protein targets. We will provide an update once the new workunits are ready to be distributed.

Help Stop TB: The team has been analyzing previous results and devising new strategies for the search. We are waiting for the Help Stop TB team to provide us with new workunits.

Smash Childhood Cancer: Once the team finalizes new targets, they will be able to prepare workunits for the next phase of the SCC project.

Zie https://www.worldcommunit...g/viewthread_thread,44874

[wildhagen]

We are excited to share an update from the Smash Childhood Cancer team about the start of the new phase in their quest for curing cancer. The new target protein is MyoD1 (Myogenic Differentiation 1 protein) and one of its mutations, L122R. MyoD1 is a transcription factor that activates muscle-specific genes, and is involved in the switch between cell growth and cell differentiation. The loss of control over this switch can cause the formation of rhabdomyosarcomas.

L122R is a mutation frequently found in rhabdomyosarcoma, and it is linked to a worse prognosis. Considering the crystal structure deposited in Protein Data Bank, the goal is to find compounds that act on the mutant-type but not on the wild-type.

New SCC work units have been prepared to help with the search, and volunteers who follow this project will be able to download them once we restart BOINC after the storage failure.

Aldus https://www.worldcommunit...s/article.s?articleId=783

Rhabdomyosarcoma (RMS) is a highly aggressive form of cancer that develops from mesenchymal cells that have failed to fully differentiate into myocytes of skeletal muscle. Cells of the tumor are identified as rhabdomyoblasts.

There are four subtypes – embryonal rhabdomyosarcoma, alveolar rhabdomyosarcoma, pleomorphic rhabdomyosarcoma, and spindle cell/sclerosing rhabdomyosarcoma. Embryonal, and alveolar are the main groups, and these types are the most common soft tissue sarcomas of childhood and adolescence. The pleomorphic type is usually found in adults.

Zie Wikipedia: Rhabdomyosarcoma

Ook van MCM is een project-update verschenen: https://www.worldcommunit...s/article.s?articleId=782

[m_p_g]

Hieronder een update van WCG admin Jurisica.
Dit geeft ook gelijk een beetje antwoord op de status bij Krembil (wat menigeen al wist).

In deed, we use VMs, and we do have multiple blades - but we do not have capacity (yet) for redundancy or sufficient capacity for growth.

This is all older equipment - but despite multiple attempts we do not have yet generous IT vendor or other partner that would give us much needed refresh and redundancy. (suffice to note - there was a planned refresh across academic HPC in Canada last year - but it did not happen yet).

However, two possible leads - *if* they work - we would be moved several years ahead.

on a quick update, finally, /science filesystem is on the move to the new storage from the recovery storage unit. As of last night, after 3 hours, the new storage /science filesystem shows 1.4TB used. Assuming such average rate of file transfer, it will take about 74 hours. Hopefully, we will be able to restart BOINC from the new storage and finally put the failure behind us. We will keep you posted.

Er wordt nog steeds aan gewerkt om misschien aan het einde van deze maand weer online te kunnen gaan met WCG. Het is te hopen dat er niet weer andere problemen opduiken bij de herstart.

[wildhagen]

Weer nieuws over SCC (Smash Childhood Cancer):

Childhood cancer cases rise every year, yet only a few drugs to combat them have been approved for medical use. Finding drug candidates that can target cancer cells is a lengthy process that takes research teams years to complete with the limited equipment they have access to. The Smash Childhood Cancer project teamed up with the World Community Grid to quicken the rate of clinical testing by using computational donations from thousands of volunteers across the world. To date, WCG volunteers have donated over 48,000 CPU years to childhood cancer research and have been instrumental in researching various types of tumors and sarcomas.

The SCC team has selected a new target protein – FLI1 – a member of the ETS transcription factor family that controls cell proliferation, differentiation and survival. Fusion of FL1 gene and EWSR1 is frequently present in cases of Ewing Sarcoma and other cancers1. EWSR1 encodes for an RNA-binding protein (EWS), and when EWS is fused to FLI1 the latter becomes constantly activated (instead of finely regulated) creating the molecular environment for tumor formation. The EWS-FLI1 fusion protein has been thought to inhibit p53 and/or activate NOTCH signaling, accelerating sarcoma progression. Useful FLI1 inhibitors need to specifically target this fusion protein but not other related ones to avoid inadvertent side effects2.

Zie https://www.worldcommunit...s/article.s?articleId=785

Eerder was al aangekondigd dat er binnenkort weer SCC-workunits gaan komen. Dit zou voor het eerst in lange tijd zijn, want de laatste workunits stammen alweer van ver vóór de start van de migratie naar Krembil.

[wildhagen]

De migratie is, na 14 maanden, nu formeel afgerond:

As of April 13, 2023, the World Community Grid has finished the alpha testing and has restarted. We greatly appreciate the community’s help and patience during the past 14 months. We are excited to expand this powerful computing platform and give more people the ability to engage in citizen science and increase research breakthroughs by using scalable yet sustainable computing.

We have faced many challenges since we started to run WCG, and it took us substantially longer than we had anticipated. For volunteers who chose to pause their contributions during the restart phase, we highly encourage you to begin crunching again with our new and improved system. The impressive growth of WCG’s computing capacity and volunteer involvement over the years has naturally slowed down over the last period – but together, we can ensure that this platform will grow and enable seemingly impossible scientific research to come to life.

This does not mean that our back-end system is perfect, or that we fixed all known issues (see our Comprehensive Bug List). We will fix the remaining issues over time. We are also working on expanding the activities – and will post updates over the next few weeks.

Zie https://www.worldcommunit...s/article.s?articleId=786



Verdere projectupdates staan in het gelinkte artikel.

[wildhagen]

Project update van MCM1:

In our previous MCM project update, we started to introduce putative lung cancer biomarkers that are present with top scores across all the signature sizes considered. The first update focused on VAMP1, a gene linked to patient survival and differentially expressed in normal lung compared to lung cancer. This update highlights a potential biomarker gene called FARP1, which is linked to metastatic development.

The protein FARP1 encodes is a guanine nucleotide exchange factor that plays a role in dendritic formation and growth. FARP1 can be found in cell membranes, cytoplasm, synapses and dendrites (Uniprot). Since both VAMP1 and FARP1 are involved with synapses, we investigated whether the two proteins interact together. While they do not interact directly (IID; Integrated Interaction Database) they share two interacting partners, RAB6A and APBA1. RAB6A plays a role in neuron projection development, while APBA1 regulates APP (Alzheimer's disease amyloid precursor protein), and is thought to have a role in the secretion of synaptic vesicles.

Neither RAB6A nor APBA1 are present in the top 99th percentile, but they suggest that there is indeed a neuronal signal associated with our two proteins of interest – VAMP1 and FARP1. The importance of this finding is related to the difficult-to-study connection between lung cancer and one of its most frequent sites of metastasis: the brain[1]. Moreover, it has been shown that FARP1 is one of the key drivers of migration and invasion in lung cancer[2].

We thus decided to investigate further the importance of this gene using external datasets. FARP1 is a potential diagnostic biomarker, similar to VAMP1, as highlighted in Figure 1. Furthermore, FARP1 is differentially expressed in metastatic samples, strengthening the connection between the molecular function of this protein and the phenotype it can help develop.

cBioPortal shows that FARP1 is significantly linked to the patients’ smoking status, and according to mirDIP, it is targeted by hsa-mir-105 (a microRNA specific to former smokers, similarly to VAMP1) and hsa-mir-150 (a microRNA specific to never smokers)[3]. The connection between smoking and lung cancer development is well established, but the molecular landscape behind such development is still not completely understood. Our signatures, derived from a dataset of patients with a history of smoking, can shed light on the molecular mechanisms and support researchers investigating the connection.

While we keep validating our results, we are excited about the impact our results can have on patient lives and research, computed by thousands of volunteers. Thank you for the contribution to this research.

Zie https://www.worldcommunit...s/article.s?articleId=794

[wildhagen]

Ter info, voor de crunchers van WCG: op 25 juli ligt het datacenter waar WCG draait plat vanwege onderhoud:

On July 25th, we will experience a system outage due to maintenance of the datacenter where the WCG servers reside. As a result of this outage, all systems will be down, including our website/forums and BOINC. Site director estimates they will need 2 days.

Zie https://www.worldcommunit...s/article.s?articleId=796

Vermoedelijk voor 2 dagen dus, dus zorg dat je buffers aan workunits groot genoeg zijn (ik ga hem op 4 dagwn zetten), zodat je toch door kunt crunchen!

[wildhagen]

Update over MCM:

In our March update, we outlined our investigation of lung cancer biomarkers conducted by testing multiple signatures in a dataset of tissues belonging to patients who have a history of lung cancer to find any groups of probes that could indicate the patient has early stage lung cancer. As the work on lung cancer signatures continues, we have started to interpret data from ovarian cancer, which we share in this update.

Similarly to our approach with lung cancer, to investigate ovarian cancer biomarkers, we used a collection of multiple datasets with a total of 711 samples. Of the samples, 410 had short survival of under 3 years, and 301 had long survival of over 4 years. Our analyses focused on differentiating patients with long survival from those with a short survival.

Thanks to the help of the WCG volunteers, we were able to test over 160 billion (1.626353x1011) candidate ovarian cancer signatures across several signature sizes

Meer info: https://www.worldcommunit...s/article.s?articleId=802

Outage:

On Tuesday, September 26th at 10:00 AM EST, there will be a system outage as the data centre where our servers reside needs to update the firmware on the equipment providing the science filesystem. They estimate that this will take approximately 1 hour. During this time, users will be unable to get new work or upload completed work, but the website and forums will still be available.

Thanks for your patience and understanding.

Zie https://www.worldcommunit...s/article.s?articleId=801

[wildhagen]

Er waren deze week weer 2 project updates, van MCM en OPN:

OPN:

The Open Pandemics team has released new work units, which are targeting the DNA polymerase of cytomegalovirus.

Scientists at Scripps Research are doing molecular modeling simulations to look for possible candidates for the development of treatments for viruses, including COVID-19. This research requires massive computing power to carry out millions of simulated laboratory experiments. The molecules identified as promising candidates are then tested in laboratories by collaborators of the OPN team.

New work units are targeting the DNA polymerase of cytomegalovirus, a common virus which is particularly harmful to pregnant or immunocompromised individuals. The goal is to identify small molecule binders to one of the polymerase subunits, preventing its interaction with another polymerase subunit, and thereby disrupting the assembly of a functional polymerase. About 40 million molecules from Enamine will be simulated to identify candidates for purchase and evaluation by our collaborators.

Zie https://www.worldcommunit...s/article.s?articleId=806 en Cytomegalovirus.

MCM:

We continue our work on characterizing lung cancer biomarkers identified in the MCM1 project. This update focuses on PCSK5, a gene associated with lung cancer survival and which shows differential expression across various cancer types.

Precision medicine is enabled by using molecular markers (signatures) that identify disease early, and stratify patients into subgroups with different disease progression patterns, leading to potentially different treatment strategies. The Mapping Cancer Markers project analyzes data sets with millions of data points collected from patients with cancers and sarcomas to find such diagnostic, prognostic and predictive signatures.

Since November 2013, World Community Grid volunteers have donated over 875,900 CPU years to the project, helping analyse data on lung and ovarian cancer and sarcoma, much more thoroughly than otherwise possible. We are immensely grateful for the volunteers who continue to donate to this project.

Further characterising the 26 top-scoring genes in lung cancer, we have already discussed VAMP1, FARP1, GSDMB, ADH6, and IL13RA1 (in our March, April, July, September, and November 2023 updates). Here, we outline information on PCSK5.

PCSK5 encodes proprotein convertase subtilisin/kexin type 5, a serine endoproteinase that processes different proproteins (Uniprot). PCSK5 has been implicated in diverse biological processes. Studies in mice have revealed that it is involved in early heart development[1] and ovarian follicle development[2]. Another study suggested that it may facilitate SARS-CoV-2 entry into myocardial tissue[3]. Genetic variation in PCSK5 has been demonstrated to modulate high-density lipoprotein (HDL) cholesterol levels[4]. In addition to its association with cancer, links have been identified between PCSK5 and other diseases. For example, mutations in PCSK5 have been identified in patients with the VACTERL association[5] and Herlyn-Werner-Wunderlich syndrome[6]. PCSK5 has also been identified as a diagnostic biomarker for endometriosis[7] and inflammatory skin disease[8].

Meer info: https://www.worldcommunit...s/article.s?articleId=805 en PCSK5.

[wildhagen]

En wederom een projectupdate van Mapping Cancer Markers:

We continue to characterise lung cancer biomarkers identified in the MCM1 project. This update focuses on HSD17B11, a gene associated with lung cancer survival. HSD17B11 is a protein coding gene, relatively ubiquitously expressed across organs and tissues. It is a short-chain alcohol dehydrogenase that metabolises secondary alcohols and ketones.

Identifying molecular markers and their combination (signatures) enables us to detect disease earlier (diagnostic signatures) and stratify patients into subgroups based on disease progression patterns (prognostic signatures), potentially leading to identifying which patients may benefit from different treatment options (predictive signatures). The Mapping Cancer Markers project analyzes data sets with millions of data points collected from patients with cancers and sarcomas to find such diagnostic, prognostic and predictive signatures.

Since November 2013, World Community Grid volunteers have donated over 894,000 CPU years to the project, helping analyse data on lung and ovarian cancer and sarcoma, much more thoroughly than otherwise possible. We are immensely grateful for this continued support.

Further characterising the 26 top-scoring genes in lung cancer, we have already discussed VAMP1, FARP1, GSDMB, AHD6, IL13RA1, PCSK5, and TLE3 in previous MCM updates. Here, we outline information on HSD17B11. Importantly, there is a strong relationship across all these proteins so far, as highlighted in FIgure 1. HSD17B11 is the 4th most connected protein from our list, with FARP1, TLE3, PCSK5 being more connected.

HSD17B11 is a gene that encodes a protein called hydroxysteroid 17-beta-dehydrogenase 11. Hydroxysteroid 17-beta-dehydrogenase 11 can convert androstan-3-alpha,17-beta-diol to androsterone in vitro, which suggests that it may participate in androgen metabolism during steroidogenesis (Uniprot). Considering its structure (Figure 2), HSD17B11 has four known ligands that can bind to it, including androsterone, glycerol, sulfate ion, and chloride ion.

Meer info: https://www.worldcommunit...s/article.s?articleId=809

Mij gaat dit als niet-medicus ietwat boven de pet, maar het klinkt als goed nieuws iig

[wildhagen]

WCG is nu down tot en met 3 januari 2025 ivm werkzaamheden in het datacenter.

Laatste update:

- We are making final preparations to power down. Before powering off all VMs and the storage servers, we will kill all traffic at the load balancer so we can allow the filesystems and databases to settle before taking a final backup.

- When we kill traffic at the load balancer overnight on Dec. 6, there will be no access to the website, forums, BOINC scheduler, BOINC uploads, or BOINC downloads.

- We have transferred database backups for BOINC and the website/forums, as well as all source code offsite. We will be transferring the final backups overnight as well.

- Today (Dec 6), we have been taking snapshots of all production VMs and sometime this evening we should have all instances that make up our backend infrastructure backed up as snapshots, ready to deploy from resilient storage if there are any issues powering up our instances when we are back online in January.

- We will be working on the new project (Mapping Arthritis Markers), long-standing issues with device synchronization between the website and BOINC databases, server status page, real-time stats API, and a new look for the forums as well as new badges during the downtime. We look forward to sharing the results of the data center refresh with you in the New Year - thank you for your support and happy holidays!

De actuele status kan je volgen via https://www.cs.toronto.edu/~juris/jlab/wcg.html en dan rechtsboven even het tabje "Operational Status" aanklikken.

En er wordt dus ook gerefereerd aan een nieuw komend project: Mapping Arthritis Markers!

Dat zou het eerste nieuwe project zijn na de migratie naar Krembil. En het zou wel welkom zijn, met alleen MCM1 en ARP1 als lopende projecten (en OPNG als Intermittend Project) werd de spoeling wel heel dun....

[wildhagen]

Interessant nieuws. Men gaat kijken of men MCM1 (en het nieuwe project MAM1 dat in januari start) ook op GPU's kan laten draaen:

- We are working with the backups of the BOINC and website/forums databases offsite to prepare the Mapping Arthritis Markers project for the launch in January.

- While we have the opportunity, we will look into optimizing the MCM1 and MAM1 applications. One example, we will create an encoding for the fields in the MCM1 configuration file and further compress the already tiny configuration file for MCM1, rather than send it in plain text as before. This update will also be used to streamline the configuration file across individual projects.

- We are also exploring the potential for a GPU version of MCM1 and MAM1 projects, and future applications for searching for biomarker signatures.

-We are working on tiered caching across the download servers where workunits that have been loaded into BOINC recently will trigger a write of the input files to a distributed in-memory cache, backed up by the local disk of the download servers.

Tevens gaat men de downloads van ARP1 die nogal heftig en intensief zijn optimaliseren:

- For ARP1 project, we will be merging all file downloads into a single archive. Previously, even the individual files were getting transient HTTP errors/interruptions frequently and had to wait in queue. We did fix that bug before we went down. This should resolve the bug that was interrupting larger downloads and uploads multiple times, and we are also exploring other optimizations that should improve workunit distribution.

Hopefully, these changes will result in a more efficient file transfers, less I/O pressure on NFS and the storage server, and therefore more throughput under load, which we will need as we look to GPU-enable these and future applications on the grid.

Zie https://www.cs.toronto.edu/~juris/jlab/wcg.html (klik op Operational Status)

Klinkt als goede acties waar men nu bezig is.

[wildhagen]

Wederom een nieuws update:

- We have begun testing the Mapping Arthritis Markers (MAM1) prototype based on the current MCM1 code. We expect to launch the beta version of the application shortly after we are back online on the 3rd of January. We will provide a firm launch date once we are back online. The GPU version of the project may take some additional time to develop and test.

- We have been reviewing the MCM1 application code and considering alternative upgrade paths for MCM1 and related projects to enable GPU compute for gene signature search in general. We have decided to work towards replacing the SvmLightLib dependency and other related sections of the code with more modern, but still self-contained dependencies that will provide the opportunity to select CPU vs. GPU backend depending on user preferences and device capability. This should also enable us to take advantage of modern instruction sets and possibly other architectures in the future.

- We will be able to publish a preliminary benchmark of the old vs. new application code, comparing across a selection of hardware that WCG staff have at home. We expect performance improvements and better memory utilization for the new version of the MCM1 and by extension initial implementation of the MAM1 project.

- In light of the merger of this pull request (introducing "BUDA" to the latest BOINC server releases), we find a strong motivation to upgrade our BOINC server version to track BOINC upstream as nearly every bioinformatics application we have experience running in an HPC environment we could run on the grid, if only we could run containers. We have considered multiple different strategies to accomplish this migration in the past, and now that we have occasion to test existing and new applications with the newest BOINC server version during this downtime offsite, we will put together a roadmap after the launch of MAM1 and upgrade BOINC server to use BUDA/containers going forward.

[wildhagen]

January 10, 2025
- All our servers are back online. We are downloding processed work units and sending out new ones, and they have the right deadlines (after the initial glitch with the time).

- We have noticed 92,194 results that ended up in an error state. We saved all these in a file so we can repair this issue. We'll be able to rescue these, especially since there is no file deleter or db purge daemon running for now, so all data related to these workunits is still on the filesystem.

- We will continue to monitor the system to make sure it is stable.

- We are also working on getting MAM project into beta.

Thank you all for continuing to support research. Happy new year.