|
<languages/>
{{Wikimedia}}
<translate>
:'''Warning''': Do not rely on any information on this page being up-to-date or correct.
<!--T:2-->
:The information at https://wikitech.leuksman.com/ is likely to be more accurate.
[[<tvar name="1">Special:MyLanguage/Wikipedia</tvar>|Wikipedia]] and the other [[<tvar name="2">Special:MyLanguage/Wikimedia projects</tvar>|Wikimedia projects]] are run from server racks located in several data centres.
</translate>
__TOC__
Other languages: [[de:Wikipedia:Server|de]], [[eo:Vikipedio:Statuso de la serviloj|eo]], [[es:Wikipedia:Servidores|es]], [[fr:Wikipédia:Statut des serveurs|fr]], [[it:Wikipedia:Server|it]], [[ウィキメディアのサーバ|ja]], [[위키미디아 서버|ko]], [[zh:Wikipedia:服务器|zh]]
<translate>
[[Wikipedia]] and the other [[Wikimedia]] projects are run from several racks full of servers.
== System architecture == <!--T:3-->
The names of the Florida servers are primarily based on famous historical [[en:encyclopedist|encyclopedist]]s, while the Kennisnet and Yahoo! machines have been named after various types of [[en:plant|plants]]. (See [[Talk:Wikimedia servers|the discussion page]] for name suggestions.) Recently there have been so many new servers that many have just been given numerical names.
</translate>
[[File:Wikipedia webrequest 2022.png|thumb|400px|<translate><!--T:7--> Simplified overview of the MediaWiki software as deployed by Wikimedia Foundation (as of August 2022). Essentially a complex version of a [[:en:LAMP (software bundle)|LAMP "stack"]].</translate>]]
'''See also''':
*[http://wikitech.leuksman.com/view/Server_admin_log Server admin log]
*[https://wikitech.leuksman.com/view/Server_roles Server roles]
*[[Wikimedia servers/hardware orders|Recent hardware orders]]
*[[Wikimedia partners and hosts]]
<translate>
{|
=== Network topology === <!--T:4-->
| __TOC__ || [[Image:Floridaserversfront1.jpg|thumb|200px|Servers installed in Tampa, Florida as of September 2005]]
|}
<br style="clear:both" />
<!--T:5-->
The Wikimedia Foundation's [[<tvar name=nt>{{lwp|Network topology}}</tvar>|network topology]] is described on the [[<tvar name="1">wikitech:Network design</tvar>|"Network design" page on Wikitech]].
=== Software === <!--T:6-->
== Overall system architecture ==
</translate>
''Note: much of the following is out of date as configurations change quickly and frequently. [https://wikitech.leuksman.com/view/Server_roles Server roles] may be more up-to-date.''
* <translate><!--T:8--> Our DNS servers run [[w:en:Comparison of DNS server software#gdnsd|gdnsd]]. We use geographical DNS to distribute requests between our seven data centers (3x US, 2x Europe, 1x Asia, 1x South America) depending on the ___location of the client.</translate>
[[Image:Wikimedia-servers-2006-05-09.svg|thumb|300px|overview, may 2006 (see also: [[server layout diagrams]])]]
* <translate><!--T:9--> We use [[w:en:Linux Virtual Server|Linux Virtual Server]] (LVS) on commodity servers to load balance incoming requests. LVS is also used as an internal load balancer to distribute MediaWiki requests. For back end monitoring and failover, we have our own system called [[<tvar name="1">wikitech:PyBal</tvar>|PyBal]].</translate>
* 89 machines in [[w:Florida|Florida]], 3 near [[w:Paris|Paris]], 11 in [[w:Amsterdam|Amsterdam]], 23 in [[w:Yahoo!|Yahoo!]]'s Korean hosting facility.
* <translate><!--T:10--> For regular MediaWiki web requests (articles/API) we use [[w:en:Varnish (software)|Varnish]] and [[w:en:Apache Traffic Server|Apache Traffic Server]] caching proxy servers in front of [[w:en:Apache HTTP Server|Apache HTTP Server]].</translate>
* <translate><!--T:11--> All our servers run [[w:en:Debian|Debian GNU/Linux]].</translate>
* <translate><!--T:12--> For distributed object storage we use [[w:en:Swift (distributed storage)|Swift]].</translate>
* <translate><!--T:13--> Our main web application is [[w:en:MediaWiki|MediaWiki]], which is written in [[w:en:PHP|PHP]] (~70 %) and [[w:en:JavaScript|JavaScript]] (~30 %).<ref>See [<tvar name="1">https://www.openhub.net/p/mediawiki/analyses/latest/languages_summary</tvar> MediaWiki analysis], [<tvar name="2">https://www.openhub.net/p/mediawiki-extensions-wmf/analyses/latest/languages_summary</tvar> MediaWiki WMF-supported extensions analysis].</ref></translate>
* <translate><!--T:14--> Our structured data is stored in [[w:en:MariaDB|MariaDB]] since 2013.<tvar name=ref-start><ref>{{cite web | url=https://blog.wikimedia.org/2013/04/22/wikipedia-adopts-mariadb/ | title=Wikipedia Adopts MariaDB | trans-title=</tvar> Wikipedia Adopts MariaDB <tvar name=ref-end>| accessdate=2014-07-20 | authorlink=https://blog.wikimedia.org/author/asher/ | date=2013-04-22 | format=text/html | work=blog.wikimedia.org | publisher=Wikimedia Foundation, Inc. | language=english }}</ref></tvar> We group wikis into clusters, and each cluster is served by several MariaDB servers, replicated in a single-master configuration.</translate>
* <translate><!--T:15--> We use [[w:en:Memcached|Memcached]] for caching of database query and computation results.</translate>
* <translate><!--T:16--> For full-text search we use [[w:en:Elasticsearch|Elasticsearch]] ([[<tvar name="1">mw:Special:MyLanguage/Extension:CirrusSearch</tvar>|Extension:CirrusSearch]]).</translate>
* <translate><!--T:17--> <tvar name=url>https://noc.wikimedia.org/</tvar> – Wikimedia configuration files.</translate>
[[File:Wikimedia Foundation Servers 2015-90.jpg|thumb|350px|<translate><!--T:19--> Wikimedia server racks at CyrusOne</translate>]]
* The [[en:Database server|master database]] server runs [[en:MySQL|MySQL]] and stores all articles.
<translate>
* The Apaches are running identically-configured [[en:Apache HTTP Server|Apache]] web servers. The Apache servers accept requests from users, get data from the database if necessary, and format the requests back to the users, by running the [[MediaWiki]] software implemented in [[en:PHP programming language|PHP]] with Turck [[en:MMCache|MMCache]] PHP cache ([[PHP caching and optimization|our experience]]). They share their work directories by NFS, so uploads etc. should remain quite in sync.
== Hosting == <!--T:20-->
</translate>
{{See also|<translate nowrap><!--T:21--> w:Wikimedia Foundation#Technology</translate>|wikitech:Data centers}}
<translate>
* The [[en:Squid cache|Squid]] systems maintain large caches of pages, so that common or repeated requests don't need to touch the Apache or database servers. They serve most page requests made by visitors who aren't logged in. They are currently running at a hit-rate of approximately 75%, effectively quadrupling the capacity of the Apache servers behind them. This is particularly noticeable when a large surge of traffic arrives directed to a particular page via a web link from another site, as the caching efficiency for that page will be nearly 100%. They also [[en:Load balancing (computing)|load balance]] the Apaches. [[en:Round robin DNS|Round-robin DNS]] is balancing the load among the Squids. See [[cache strategy]] for more details.
<!--T:22-->
As of <tvar name="1">{{DateT|2024|4}}</tvar>, we have the following colocation facilities (each name except for Magru is derived from an acronym of the facility’s company and an acronym of a nearby airport):
</translate>
; eqiad :<translate><!--T:23--> Application services (primary) at [[w:en:Equinix|Equinix]] in [[w:en:Ashburn, Virginia|Ashburn, Virginia]] (Washington, DC area).</translate>
; codfw :<translate><!--T:24--> Application services (secondary) at [[w:en:CyrusOne|CyrusOne]] in [[w:en:Carrollton, Texas|Carrollton, Texas]] (Dallas-Fort Worth area).</translate>
; esams
: <translate><!--T:25--> Caching at [[w:en:EvoSwitch|EvoSwitch]] in [[w:en:Amsterdam|Amsterdam]], the [[w:en:Netherlands|Netherlands]].<ref name="Sep06">Suffered a major DoS attack on September 6/7, 2019. See [[<tvar name="1">foundationSite:news/2019/09/07/malicious-attack-on-wikipedia-what-we-know-and-what-were-doing/</tvar>|dedicated article on WMF website]].</ref></translate>
; ulsfo :<translate><!--T:26--> Caching at [<tvar name="1">https://web.archive.org/web/20190524154655/https://unitedlayer.com/our-infrastructure</tvar> United Layer] in [[w:en:San Francisco|San Francisco]].</translate>
; eqsin :<translate><!--T:27--> Caching at Equinix in [[w:en:Singapore|Singapore]].</translate>
; drmrs :<translate><!--T:111--> Caching at [[w:en:Digital Realty|Digital Realty]] in [[w:en:Marseille|Marseille]], France.</translate>
; magru: <translate><!--T:115--> Caching in [[w:São Paulo, Brazil|São Paulo, Brazil]].</translate>
<translate>
The system is designed to [[w:failover|failover]] to backup configurations at both the squid and Apache levels, and backup database support is in place, but not in an automatic failover.
=== History === <!--T:28-->
<!--T:29-->
(Details on database replication in MySQL: [http://dev.mysql.com/doc/internals/en/replication.html])
The backend web and database servers are in Ashburn, with Carrollton to handle emergency fallback in the future. Carrollton was chosen for this as a result of the [[<tvar name="1">wmf:RFP/2013 Datacenter</tvar>|2013 Datacenter RfC]]. At EvoSwitch, we have a Varnish cache cluster and several miscellaneous servers. The Kennisnet ___location is now used only for network access and routing.
<!--T:30-->
== Hosting ==
Ashburn ('''eqiad''') [[<tvar name="1">wmfblog:2013/01/19/wikimedia-sites-move-to-primary-data-center-in-ashburn-virginia/</tvar>|became the primary data center]] in January 2013, taking over from Tampa ('''pmtpa''' and '''sdtpa''') which had been the main data centre since 2004. Around April 2014, sdtpa (Equinix – formerly Switch and Data – in [[w:en:Tampa, Florida|Tampa, Florida]], provided networking for pmtpa) was shut down, followed by pmtpa (Hostway – formerly PowerMedium – in Tampa, Florida) in October 2014.
''see [[Wikimedia partners and hosts]] for more''
<!--T:31-->
<div style="float: right; background-color: #f9f9f9; border-style:solid; border-width:1px; border-color:#cccccc; padding:5px; ">
In the past we've had other caching locations like Seoul ('''yaseo''', Yahoo!) and Paris ('''lopar''', Lost Oasis); the [[<tvar name="1">strategy:Wikimedia Movement Strategic Plan Summary/Increase Reach</tvar>|WMF 2010–2015 strategic plan reach target]] states: "additional caching centers in key locations to manage increased traffic from Latin America, Asia and the Middle East, as well as to ensure reasonable and consistent load times no matter where a reader is located."
[[Image:Paris servers DSC00190.jpg|150px|]]
[[Image:Wikimedia_Paris_servers_back_1218_144129X.jpg|133px|]]
<!--T:32-->
The cluster near Paris.<br>
EvoSwitch and Kennisnet are recognised as benefactors for their [[W:en:Gifts in kind|in-kind donations]]. See the [[<tvar name="1">foundation:Benefactors</tvar>|current list of benefactors]].
Our servers are the three in the middle.<br>
<small>(From top to bottom: bleuenn, chloe, ennael.)</small>
</div>
<!--T:33-->
At present all DBs, and most Apaches and Squids, are hosted at the [[en:Florida|Florida]] Power Medium data center. From the start of the project until September 2004 Bomis.com paid for all bandwidth.
A list of servers and their functions used to be available at the [<tvar name="1">https://web.archive.org/web/20130119062922/http://wikitech.wikimedia.org/view/Server_roles</tvar> server roles] page; no such list is currently maintained publicly (perhaps the private [<tvar name="2">https://racktables.wikimedia.org/</tvar> racktables tool] has one). It used to be possible to see a [<tvar name="3">https://icinga.wikimedia.org/cgi-bin/icinga/status.cgi?servicegroup=all&style=summary&nostatusheader</tvar> compact table of all servers grouped by type] on icinga, but this is [[<tvar name="4">phab:T62112</tvar>|no longer publicly available]]. However, the [[<tvar name=5>phab:diffusion/OPUP/browse/production/manifests/site.pp</tvar>|puppet configuration]] provides a pretty good reference for the software that each server runs.
</translate>
[[File:B-roll of Wikimedia Foundation servers at CyrusOne in Carrollton, Texas - photographed in 2015.webm|thumb|<translate><!--T:104--> B-roll of servers in Texas in 2015</translate>]]
<translate>
== Status and monitoring == <!--T:34-->
<!--T:35-->
[http://www.lost-oasis.fr/ Lost Oasis] (bandwidth) and [http://www.jexiste.org Jexiste] (rack space) in [[:en:France|France]] are providing hosting and bandwidth for the European Squid cache service using ''chloe'', ''bleuenn'' and ''ennael''. The machines are installed in the [[:en:Telecity|Telecity]] data center in [[:en:Aubervilliers|Aubervilliers]] near [[:en:Paris|Paris]] and serve France, Luxembourg and Switzerland.
You can check one of the following sites if you want to know if the Wikimedia servers are overloaded, or if you just want to see how they are doing.
</translate>
* [https://www.wikimediastatus.net/ https://www.wikimediastatus.net/]
* <translate><!--T:36--> [<tvar name="1">https://grafana.wikimedia.org/</tvar> Grafana]</translate>
** <translate><!--T:105--> <s>[<tvar name="1">https://grafana.wikimedia.org/dashboard/db/prometheus-dc-overview</tvar> Data center overview] with total bandwidth, non-idle CPU and load per group of servers; versions also exist with a [<tvar name="2">https://grafana.wikimedia.org/dashboard/db/prometheus-cluster-breakdown</tvar> per-subcluster dropdown] and [<tvar name="3">https://grafana.wikimedia.org/dashboard/db/prometheus-global-overview</tvar> total numbers without graphs]</s> ''(down in July 2019)''</translate>
*** <translate><!--T:106--> <tvar name="1">https://grafana.wikimedia.org/</tvar> should get you started</translate>
** <translate><!--T:108--> [<tvar name="1">https://grafana.wikimedia.org/d/CbNAwAXnk/filippo-blackbox-smoke-icmp</tvar> Networking latency]</translate>
* <translate><!--T:107--> [<tvar name="1">https://icinga.wikimedia.org/</tvar> Icinga] ([[<tvar name="2">phab:T62112</tvar>|private]])</translate>
<translate>
[[Kennisnet/en|Kennisnet]] in [[:en:Netherlands|the Netherlands]] is providing hosting and bandwidth for eleven servers since June 2005. They are installed at [http://www.sara.nl/aboutsara/aboutsara_pr_eng.html SARA] in [[en:Amsterdam|Amsterdam]] and provide service for the Netherlands, Belgium, Germany, UK, Austria, Poland and Scandinavia. Also the [[Toolserver]] Zedler is installed here.
<!--T:37-->
If you are seeing errors in real time, visit <tvar name=chan>{{channel|wikimedia-tech}}</tvar> on [[IRC|irc.libera.chat]]. Check the topic to see if someone is already looking into the problem you are having. If not, please report your problem to the channel. It would be helpful if you could report specific symptoms, including the exact text of any error messages, what you were doing right before the error, and what server(s) are generating the error, if you can tell.
== Energy use == <!--T:38-->
[http://www.yahoo.com/ Yahoo!] is providing servers, hosting, and bandwidth for twenty-three servers in [[en:South Korea|South Korea]].
</translate>
<translate><!--T:112--> In 2017, the WMF board of trustees adopted a [[<tvar name=1>foundation:Resolution:Environmental Impact</tvar>|resolution]] containing a commitment to minimize the Foundation's overall environmental impact, especially around data centres through using green energy.</translate> <translate><!--T:39--> The community-led [[<tvar name="1">Special:MyLanguage/Sustainability Initiative</tvar>|Sustainability Initiative]], created in 2015, aims at reducing the environmental impact of the servers by calling for renewable energy to power them.</translate>
Zedler was donated by [http://www.sun.com/ Sun Microsystems] to [[Wikimedia Deutschland]] to provide hosting for miscellaneous tools written by users, and is not part of the main cluster.
<translate>
Cluster Nomenclature:
<!--T:40-->
* PM TPA - Powermedium / [[:en:Tampa, Florida|Tampa, Florida, US]]
The Wikimedia Foundation's servers are spread out in five colocation data centers in Virginia, Texas and San Francisco in the United States, Amsterdam and Marseille in Europe, and Singapore in Asia.
* LO PAR - Lost Oasis / [[:en:Paris, France|Paris, France]]
* KN AMS - KennisNet / [[:en:Amsterdam, Netherlands|Amsterdam, Netherlands]]
* YA SEO - Yahoo! / [[:en:Seoul, South Korea|Seoul, South Korea]]
<!--T:113-->
More equipment, hosting and bandwidth offers for squid cache clusters are welcome, see [http://wp.wikidev.net/Volunteer_Squid_Sites this page for requirements].
<u>In 2021, the servers used '''358.8 kW''' (kilowatts), summing up to about '''3.143 GW h''' (gigawatt hours) of electrical energy '''per year'''. The total carbon footprint of the servers was '''1,073 metric tons CO2-eq''' in 2021.</u><ref>[https://meta.wikimedia.org/wiki/File:Wikimedia_Foundation_Environmental_Sustainability_(Carbon_Footprint)_Report_for_2021.pdf Wikimedia Foundation Environmental Sustainability (Carbon Footprint) Report for 2021]</ref>
<!--T:41-->
[[Image:KN-Servers2.JPG|thumb|right|220px|Kennisnet Amsterdam servercluster in its own rack]]
Only the few servers in Amsterdam and in Marseille run on renewable energy, the other use different conventional energy mixes. In 2016, just '''9%''' of Wikimedia Foundation data centers' energy came from renewable sources, with the rest split evenly between coal, gas and nuclear power (34%, 28%, and 28%, respectively). The bulk of the Wikimedia Foundation's electricity demand is in Virginia and Texas, which both have fossil fuel heavy grids.
</translate>
{| class="wikitable"
!<translate><!--T:42--> Server name</translate>
!<translate><!--T:43--> Data center ___location</translate>
!<translate><!--T:44--> Provider</translate>
!<translate><!--T:45--> Date opened</translate>
!<translate><!--T:46--> Average energy consumption (kW)</translate>
!<translate><!--T:47--> Energy sources</translate>
!<translate><!--T:48--> Carbon footprint (CO<sub>2</sub>/year)</translate>
!<translate><!--T:49--> Renewable option and cost</translate>
|-
|eqiad
|Ashburn, VA
20146-20149
USA
|Equinix ([https://www.equinix.com/ <translate><!--T:50--> Website</translate>])
<translate>
<!--T:51-->
|February 2011
|May 2016: 130
May 2015: 152
|2016: <br/> 32% coal
<!--T:52-->
==Orders and detailed hardware descriptions==
20% natural gas
''See also: [[/hardware orders]]''
<!--T:53-->
=== 2006 ===
25% nuclear
*[http://wikimediafoundation.org/wiki/Resolution_Hardware_sept_2006 Hardware purchase Sep-06] (part II not bought mid december) (~ $300,000)
*[http://wikimediafoundation.org/wiki/Resolution:Hardware_Purchase_Jul-06 Hardware Purchase Jul-06] (~ $61,440)
*[http://wikimediafoundation.org/wiki/Resolution_Hardware_Jun06 Hardware Purchase June-06] (~ $60,000)
*[http://wikimediafoundation.org/wiki/Resolution_Hardware_Purchase_Feb-06 Hardware ordered February 06, 2006] (~ $138,000)
<!--T:54-->
=== 2005 - 129 new servers ===
17% renewable
*[[Hardware ordered November 15, 2005]]
</translate>
*[[Hardware ordered October 18, 2005]]
*[[Hardware ordered October 6, 2005]]
*[[Hardware ordered September 14, 2005]]
*[[Hardware ordered August 30, 2005]]
* Donated: 23 multi-purpose servers in South Korea
*[[Hardware ordered May 2005]]: 2 db servers.
*[[Hardware deployed May 9, 2005]]: 20 apache servers (10 with 1GB, 10 with 3GB for memcached use).
*[[Hardware ordered January 2005]]: 10 new servers.
* Donated: 2 squid servers in Paris (florence, sophie)
* Donated: 8 squid servers and three multipurpose servers in Amsterdam
|<translate><!--T:55-->
=== 2004 - 39 new servers ===
1,040,000 lb = 520 short tons = 470 metric tons
*[[Hardware ordered December 2004]]: 5 new servers: all 3GB apache/memcached/squid type.
<small>= 0.32 * 130 kW * 8765.76 hr/yr * 2.1 lb CO<sub>2</sub>/kWh for coal</small>
*[[Hardware ordered October 2004]]: 7 new servers: 2 database, 5 apache
*[[Hardware ordered August 2004]]: 10 new servers. Search database server (bacon), NFS storage server (albert), 8 3.0GHz P4 web servers (diderot, goeje, avicenna, dalembert, tingxi, alrazi, friedrich, harris), gigabit ethernet switch, 146GB SCSI drive for Suda.
*[[Hardware ordered May 2004]]: 5 new servers. Replacement for Geoffrin database server (ariel), four 2.8GHz P4 general purpose machines (maurus, rabanus, yongle) and a pair of 250GB ATA drives. Based on [[upgrade discussion April 2004]].
*[[Hardware ordered January 2004]]: 9 new servers: 8 multipurpose machines (bart, bayle, browne, coronelli, isidore, moreri, vincent, zwinger) and 1 database server (suda).
*Donated: three for Paris squids, extra RAM for them purchased :) :)
<!--T:56-->
== Server list ==
<small>+ 0.20 * 130 kW * 8765.76 hr/yr * 1.22lb CO<sub>2</sub>/kWh for nat gas</small>
The popularity of the Wikimedia projects necessitates the use of many servers, all but two run the [[en:Linux|GNU/Linux]] operating system. Below may be the currentconfiguration, [https://wikitech.leuksman.com/view/Server_roles Server roles] is likely more up-to-date.
<!--T:57-->
For IP addresses, see [http://wikitech.leuksman.com/view/IP_addresses here].
<small>+ 0.25 * 130 kW * 8765.76 hr/yr * 0 lb CO<sub>2</sub>/kWh for nuclear</small>
{|+ border="2" cellpadding="4" cellspacing="0" style="margin: 1em 1em 1em 0; background: #f9f9f9; border: 1px #aaa solid; border-collapse: collapse;"
| Name || Date || Tasks || OS || Hardware || Named after
<!--T:58-->
|- style="background-color:#e1e1e1;"
<small>+ 0.17 * 130 kW * 8765.76 hr/yr * 0 lb CO<sub>2</sub>/kWh for renewable</small>
| 12 || colspan=6 | [http://zwinger.wikimedia.org/ganglia/?c=MySQL&m=&r=hour&s=descending&hc=4 MySQL Database]
</translate>
|<translate><!--T:59--> In 2015, Equinix [<tvar name=url1>https://obamawhitehouse.archives.gov/the-press-office/2015/11/30/white-house-announces-additional-commitments-american-business-act</tvar> made] "a long-term commitment to use 100 percent clean and renewable energy". In 2017, Equinix [<tvar name=url2>https://www.datacenterknowledge.com/archives/2017/04/04/equinix-to-press-on-with-clean-energy-goals-after-trumps-climate-order/</tvar> renewed] this pledge.</translate>
|-
|codfw
|[[samuel]] || [[Hardware_ordered_May_2005|05/2005]] || [[w:Database master|Database master]] || <small>[[w:Fedora Core|Fedora Core]]</small> || <small>2U, 2 x [[w:Opteron|Opteron]] 250 2.4[[w:GHz|GHz]], 16 [[w:GB|GB]] (8x2G) PC3200/[[w:DDR400|DDR]] Registered ECC - [[w:Interleaved|Interleaved]] [[w:RAM|RAM]], 6x [[w:Seagate|Seagate]] Cheetah 15K.4 73GB U320 15KRPM SCA [[w:SCSI|SCSI]] [[w:RAID|RAID]] 0 </small>|| [[en:Samuel Johnson|Samuel Johnson]]
|Carrollton, TX
|-
75007
|[[ariel]] || [[Hardware ordered May_2004|05/2004]] || [[w:Database slave|Database slave]]|| <small>Fedora Core </small> || <small>2U, 2 x Opteron 248 2.2GHz, 8 GB (8x1G) PC2700 registered ECC RAM, 6x73GB 15K SCA SCSI drives (4 RAID 1+0 (146GB), 2 RAID 1 (72GB))</small>|| [[en:Ariel Durant|Ariel Durant]]
USA
|-
|CyrusOne ([https://www.cyrusone.com/data-centers/north-america/texas/dfw1-dallas/ <translate><!--T:60--> Website</translate>])
|[[holbach]]|| [[Hardware ordered October 2004|10/2004]] ||Database slave || <small>Fedora Core</small> || <small>2U, 2xOpteron 242 1.6GHz, 4 GB (4x1G) PC2700 registered ECC RAM, 6x74GB WD Raptor 10K SATA RAID 0</small> || [[en:Baron d'Holbach|Baron d'Holbach]]
<translate>
|-
<!--T:61-->
|[[ixia]] || [[Hardware ordered September 14, 2005|09/2005]] || Database slave|| <small>Fedora Core</small> || <small>2U, 2 x Opteron 275 (dual-core) 2.2GHz, 16 GB RAM, 6x Seagate Cheetah 15K.4 73GB U320 15KRPM SCA SCSI RAID 0</small>|| [[en:Ixia|Ixia]]
|May 2014
|-
|May 2016: 77
|[[lomaria]] || [[Hardware ordered September 14, 2005|09/2005]] || Database slave|| <small>Fedora Core</small> || <small>2U, 2 x Opteron 275 (dual-core) 2.2GHz, 16 GB RAM, 6x Seagate Cheetah 15K.4 73GB U320 15KRPM SCA SCSI RAID 0</small>|| [[en:Lomaria|Lomaria]]
May 2015: 70
|-
|2016: <br/> 23% coal
|[[adler]] || [[Hardware_ordered_May_2005|05/2005]] || Database slave|| <small>Fedora x86_64</small> || <small>2U, 2 x Opteron 250 2.4GHz, 16 GB (8x2G) PC3200/DDR400 Registered ECC - Interleaved RAM, 6x Seagate Cheetah 15K.4 73GB U320 15KRPM SCA SCSI RAID 0 </small>|| [[en:Mortimer Adler|Mortimer Adler]]
|-
<!--T:62-->
|[[thistle]] || [[Hardware ordered September 14, 2005|09/2005]] || Database slave|| <small>Fedora Core</small> || <small>2U, 2 x Opteron 275 (dual-core) 2.2GHz, 16 GB RAM, 6x Seagate Cheetah 15K.4 73GB U320 15KRPM SCA SCSI RAID 0</small>|| [[en:Thistle|Thistle]]
56% natural gas
|-
|[[webster]]|| [[Hardware ordered October 2004|10/2004]] ||Database slave || <small>Fedora Core</small> || <small>2U, 2xOpteron 242 1.6GHz, 4 GB (4x1G) PC2700 registered ECC RAM, 6x73GB Seagate Cheetah 10K.6 U320 SCSI drives RAID 0</small> || [[en:Noah Webster|Noah Webster]]
|-
|db1 .. db4|| [http://wikimediafoundation.org/wiki/Resolution_Hardware_Purchase_Feb-06 02/2006] || 4x Database slave || <small>Fedora Core</small> || <small>2U, 2 x Opteron 248 (dual-core) 2.2GHz, 16 GB RAM, 6x Seagate Cheetah 15K.4 73GB U320 15KRPM SCA SCSI RAID 0</small> || -
|- style="background-color:#e1e1e1;"
| 14 || colspan=6 | [http://zwinger.wikimedia.org/ganglia/?c=Miscellaneous&m=&r=hour&s=descending&hc=4 Miscellaneous]
|-
|[[albert]]|| [[Hardware ordered August_2004|08/2004]] ||NFS storage server. || <small>SUSE Lin. 9.1</small> || <small>2U, 2xOpteron 242 1.6GHz, 1 GB (2x512M) PC2700 registered ECC RAM, 6x250GB SATA (RAID 5) ([[Albert|details]])</small> || [[en:Albertus Magnus|Albertus Magnus]]
|-
|[[amane]]|| 11/2005||NFS storage server || <small> Fedora Core</small> || <small>3U, 2x Xeon 3.4GHz, 8GB RAM, 16 x 400GB SATA disk</small> || [[en:Nishi Amane|Nishi Amane]]
|-
| [[zwinger]] ||[[Hardware ordered January_2004|01/2004]] || NFS, mail, DNS, coda (experimental), misc stuff || <small>Red Hat Lin. 9</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 1 GB (cap?) unbuf DDR400 non-ECC RAM, 250 GB SATA ([[Zwinger|details]])</small>|| [[en:Theodor Zwinger|Theodor Zwinger]]
|-
| [[larousse]] || <small>Prehistoric</small> || IRC, [[servmon]], cricket, [[Realtime recent changes|rcdumper]], udpmcast, VPN routing, DNS || <small>Fedora Core</small> || <small>1U, 1 x Pentium III 866MHz, 2 GB RAM, 32GB+18GB SCSI disk</small>|| [[en:Pierre Larousse|Pierre Larousse]]
|-
|[[bacon]]|| [[Hardware ordered August_2004|08/2004]] ||Former database slave (being repurposed)|| <small>Fedora Core</small> || <small>2U, 2xOpteron 242 1.6GHz, 4 GB (4x1G) PC2700 registered ECC RAM, 6x200GB SATA RAID 10</small> || [[en:Francis Bacon|Francis Bacon]]
|-
|[[suda]] ||[[Hardware ordered January_2004|01/2004]] || Former database slave (being repurposed) || <small>Fedora Core</small> || <small>2U, 2 x Opteron 246 2.0GHz, 4 GB (4x1G) PC2700 registered ECC RAM, 4x146GB SCSI RAID 1+0 (292GB) and 2x36GB SCA SCSI RAID 1 (36GB)</small> || [[en:Suda|Suda]]
|-
|[[vincent]] ||[[Hardware ordered January_2004|01/2004]] || Search server || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 1 GB (cap?) unbuf DDR400 non-ECC RAM, 80 GB SATA </small> || [[en:Vincent of Beauvais|Vincent of Beauvais]]
|-
|[[coronelli]] ||[[Hardware ordered January 2004|01/2004]] || Search server|| <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 3 GB (2x1G? +?) unbuf non-ECC DDR400 RAM, 80 GB SATA</small> || [[en:Vincenzo Coronelli|Vincenzo Coronelli]]
|-
|[[maurus]] ||[[Hardware ordered May 2004|05/2004]]|| Search server|| <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.8 GHz, 4 GB (4x1G) unbuf DDR 400 ECC RAM, 80 GB SATA</small> || [[en:Rabanus Maurus|Rabanus Maurus]]
|-
|[http://wp.wikidev.net/Avicenna avicenna]||[[Hardware ordered August_2004|08/2004]]|| LVS load balancer ||<small>Fedora Core</small> || <small>1U, 1 x Pentium 4 3 GHz, 0.5 GB (2x256M) unbuf DDR400 RAM, 200 GB SATA</small> || [[en:Avicenna|Avicenna]]
|-
|[[alrazi]]||[[Hardware ordered August_2004|08/2004]]|| LVS load balancer||<small>Fedora Core</small> ||<small>1U, 1 x Pentium 4 3 GHz, 0.5 GB (2x256M) unbuf DDR400 RAM, 200 GB SATA </small> ||[[en:Al-Razi|Al-Razi]]
|-
|[[isidore]] ||[[Hardware ordered January_2004|01/2004]] || Database for fundraising, misc. tasks || <small>FreeBSD5.3-REL</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 1 GB (cap?) unbuf DDR400 non-ECC RAM, 80 GB SATA</small> || [[en:Isidore of Seville|Isidore of Seville]]
|-
|[http://wp.wikidev.net/Khaldun khaldun] || [[Hardware ordered October 2004|10/2004]] || Install server || <small>Ubuntu 6.06 LTS</small>|| <small>1U, 1x Pentium 4 3.0 GHz, 1 GB RAM, 2x250 GB SATA RAID 0</small> || [[en:Ibn Khaldun|Ibn Khaldun]]
|-
|[http://wp.wikidev.net/Browne browne] ||[[Hardware ordered January 2004|01/2004]] || Squid (inactive) || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 4 GB (4x1G) unbuf non-ECC DDR400 RAM, 80 GB SATA</small> || [[en:Thomas Browne|Thomas Browne]]
|- style="background-color:#e1e1e1;"
| 18 || colspan=6 | [http://zwinger.wikimedia.org/ganglia/?c=Squids Squid]
|-
|srv5 .. srv10|| [[Hardware_ordered_January_2005|01/2005]] || 6x Squid || <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.2GHz, 3GB RAM, 2x 36GB SATA</small> || -
|-
|sq1 .. sq10 || [[Hardware_ordered_November_15%2C_2005|11/2005]] || 10x Upload/Image Squids || <small>Fedora Core</small> || <small>1U, 2xOpteron 248 (2.2Ghz), 4x1GB DDR ECC RAM, 4x 36GB 10k SATA</small> || -
|-
| [[bayle]] ||[[Hardware ordered January 2004|01/2004]] || Apache, 1120MB memcached, switched between Squid and Apache as needed || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 2 GB (2x1G) unbuf DDR400 non-ECC RAM, 80 GB SATA</small> || [[en:Pierre Bayle|Pierre Bayle]]
|-
|[[will]] ||[[Hardware ordered May 2004|05/2004]] || Squid || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.8 GHz, 4 GB RAM, 2 x 200GB SATA in RAID 1 ([[Hardware order May 2004|details]])</small> || [[en:Will Durant|Will Durant]]
|-
|- style="background-color:#e1e1e1;"
| 158 || colspan=6 | [http://zwinger.wikimedia.org/ganglia/?c=Internal%20apaches&m=&r=hour&s=descending&hc=4 Apache]
|-
| srv0 || [[Hardware ordered October 2004|10/2004]] || Apache, DB backup slave|| <small>Fedora Core</small>|| <small>1U, 1x Pentium 4 3.0 GHz, 1 GB RAM, 2x 250GB SATA RAID 0</small> || [[en:Averroes|Averroes]]
|-
|srv1 .. srv4 || [[Hardware_ordered_January_2005|01/2005]] || 4x Apache|| <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.2GHz, 1GB RAM, 1x 200GB SATA</small> || -
|-
|srv11 .. srv20 || [[Hardware_deployed_May 9,_2005|05/2005]] || 10x Apache|| <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.4GHz, 1GB RAM, 1x 80GB SATA</small> || -
|-
|srv21 .. srv27 || [[Hardware_deployed_May 9,_2005|05/2005]] || 7x Apache|| <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.4GHz, 3GB RAM, 1x 80GB SATA</small> || -
|-
|srv28 .. srv30 || [[Hardware_deployed_May 9,_2005|05/2005]] || 4x Apache, external storage || <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.4GHz, 3GB RAM, 1x 80GB SATA</small> || -
|-
|srv31 .. srv50 || [[Hardware ordered August 30, 2005|09/2005]] || 10x Apache || <small>Fedora Core</small> || <small>1U, 2x Opteron 248 (2.2GHz), 2x1GB DDR ECC RAM, 1x 250GB 7.2k SATA</small> || -
|-
|srv51 .. srv70 || [[Hardware_ordered_October_6%2C_2005|10/2005]] || 20x Apache || <small>Fedora Core</small> || <small>1U, 2xOpteron 248 (2.2Ghz), 4x1GB DDR ECC RAM, 1x 7.2k 250GB SATA</small> || -
|-
|srv71 .. srv80 || [[Hardware_ordered_November_15%2C_2005|11/2005]] || 10x Apache || <small>Fedora Core</small> || <small>1U, 2xOpteron 248 (2.2Ghz), 4x1GB DDR ECC RAM, 1x 250GB 7.2k SATA</small> || -
|-
|srv81 .. srv120 || [http://wikimediafoundation.org/wiki/Resolution_Hardware_Purchase_Feb-06 02/2006] || 40x Apache || <small>Fedora Core</small> || <small>1U, 2xOpteron 265 (1.8Ghz), 4x1GB DDR ECC RAM, 2x 250GB 7.2k SATA</small> || -
|-
| [[anthony]] ||[[Hardware ordered December 2004|12/2004]] || Apache/ Memcached || <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.0 GHz, 2x1 GB + 2x512 MB DDRII-533 RAM, 1x200 GB SATA RAID 0</small> || [[en:Anthony Winkler Prins|Anthony Winkler Prins]]
|-
|[[bart]] ||[[Hardware ordered January 2004|01/2004]] || Apache || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 4 GB (4x1G) unbuf DDR400 non-ECC RAM, 80 GB SATA</small> || [[en:Bartholomeus de Glanvilla|Bartholomeus de Glanvilla]]
|-
| [[benet]] ||[[Hardware ordered December 2004|12/2004]] || Apache/ Memcached/ Squid. Database slave, test production use. || <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.0 GHz, 2x1 GB + 2x512 MB DDRII-533 RAM, 2x250 GB SATA RAID 0 || [[en:William Rose Benet|William Rose Benet]]
|-
| [[biruni]] ||[[Hardware ordered December 2004|12/2004]] || Apache/ Memcached/ Squid. || <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.0 GHz, 2x1 GB + 2x512 MB DDRII-533 RAM, 1x200 GB SATA</small> || [[en:Al-Biruni|Al-Biruni]]
|-
| [[dalembert]]||[[Hardware ordered August_2004|08/2004]]|| LVS ||<small>Fedora Core</small> || <small> 1U, 1 x Pentium 4 3 GHz, 0.5 GB (2x256M) unbuf DDR400 RAM, 200 GB SATA</small> || [[en:Jean le Rond d'Alembert|Jean le Rond d'Alembert]]
|-
|<small>[http://wp.wikidev.net/Diderot diderot], friedrich, goeje, harris</small>||[[Hardware ordered August 2004|08/2004]]|| 4x Apache. ||<small>Fedora Core</small> || <small>1U, 1 x Pentium 4 3 GHz, 0.5 GB (2x256M) unbuf DDR400 ECC RAM, 200 GB SATA</small> || <small> [[en:Denis Diderot|Denis Diderot]], [[en:Friedrich Arnold Brockhaus|Friedrich Arnold Brockhaus]], [[en:Michael Jan de Goeje|Michael Jan de Goeje]], [[en:John Harris|John Harris]] </small>
|-
|[http://wp.wikidev.net/Hypatia hypatia] || [[Hardware ordered October 2004|10/2004]] || Apache, IRC bots || <small>Fedora Core</small>|| <small>1U, 1x Pentium 4 3.0 GHz, 1 GB RAM, 1x 200GB SATA</small> || [[en:Hypatia of Alexandria|Hypatia of Alexandria]]
|-
|<small>[[humboldt]], [http://wp.wikidev.net/Kluge kluge]</small> || [[Hardware ordered October 2004|10/2004]] || 2x Apache || <small>Fedora Core</small>|| <small>1U, 1x Pentium 4 3.0 GHz, 1 GB RAM, 1x 200GB SATA</small> || <small> [[en:Alexander von Humboldt|Alexander von Humboldt]], [[en:Friedrich Kluge|Friedrich Kluge]] </small>
|-
|[[moreri]] ||[[Hardware ordered January_2004|01/2004]] || Apache, 280MB memcached || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.6 GHz, 1 GB (cap?) unbuf DDR400 non-ECC RAM, 80 GB SATA</small> || [[en:Louis Moréri|Louis Moréri]]
|-
|[http://wp.wikidev.net/Rabanus rabanus] ||[[Hardware ordered May 2004|05/2004]]|| Apache, DHCP boot server || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.8 GHz, 4 GB (4x1G) unbuf DDR 400 ECC RAM, 80 GB SATA</small> || [[en:Rabanus Maurus|Rabanus Maurus]]
|-
|<small>[http://wp.wikidev.net/Rose rose], smellie</small>||[[Hardware ordered December 2004|12/2004]] || 2x Apache / Memcached. || <small>Fedora Core</small> || <small>1U, 1x Pentium 4 3.0 GHz, 2x1 GB + 2x512 MB DDRII-533 RAM, 1x200 GB SATA</small> || <small>[[en:William Rose Benet|William Rose Benet]], [[en:William Smellie|William Smellie]] </small>
|-
|[http://wp.wikidev.net/Tingx tingxi]||[[Hardware ordered August_2004|08/2004]]|| Apache, IRC bots. Currently [http://wp.wikidev.net/Tingxi malfunctioning] and in need of developer attention||<small>Fedora Core</small> || <small>1U, 1 x Pentium 4 3 GHz, 0.5 GB (2x256M) unbuf DDR400 RAM, 200 GB SATA</small> || [[en:Jiang Tingxi|Jiang Tingxi]]
|-
|[[yongle]] ||[[Hardware ordered May_2004|05/2004]]|| Apache, 1960MB memcached || <small>Fedora Core</small> || <small>1U, 1 x Pentium 4 2.8 GHz, 4 GB (4x1G) unbuf DDR 400 ECC RAM, 80 GB SATA</small> || [[en:Yongle Emperor of China|Emperor Yongle]]
|- style="background-color:#e1e1e1;"
| 3 || colspan=6 | [http://zwinger.wikimedia.org/ganglia/?c=Paris%20cluster&m=&r=hour&s=descending&hc=4 Paris Cluster]
|-
|[[bleuenn]]|| 12/2004 || Former Squid in France, not in use ||<small>Debian</small> || <small>1U, 1 x Celeron 600 MHz, 640 MB RAM, 20 GB ATA. [http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?contentGroup=NT_CT_Specifications&locale=en_US&docIndexId=178991&taskId=101&prodTypeId=15351&prodSeriesId=62929 HP Hosting Server Appliance sa1100]</small> || donor request
|-
|[[chloe]]|| 12/2004 || Former Squid in France, not in use ||<small>Debian</small> || <small>1U, 1 x Celeron 600 MHz, 640 MB RAM, 20 GB ATA. [http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?contentGroup=NT_CT_Specifications&locale=en_US&docIndexId=178991&taskId=101&prodTypeId=15351&prodSeriesId=62929 HP Hosting Server Appliance sa1100]</small> || donor request
|-
|[[ennael]]|| 12/2004 || Former Squid in France, not in use ||<small>Debian</small> ||<small>1U, 1 x Celeron 600 MHz, 768 MB RAM, 20 GB ATA. [http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?contentGroup=NT_CT_Specifications&locale=en_US&docIndexId=178991&taskId=101&prodTypeId=15351&prodSeriesId=62929 HP Hosting Server Appliance sa1100]</small> || donor request
|- style="background-color:#e1e1e1;"
| 12 || colspan=6 | [http://ganglia.wikimedia.org/small/?c=Kennisnet%20cluster&m=&r=hour&s=descending&hc=4 Kennisnet cluster]
|-
|[[pascal]]||5/2005||WWW, DNS, NFS||<small>Fedora Core</small>||<small>Sun V40z, 2 x 1.8GHz Opteron, 2GB RAM, 2x 70GB SCSI || [[w:en:Blaise Pascal|Blaise Pascal]]
|-
|[[vandale]]||5/2005||Squid||<small>Fedora Core</small>||<small>Sun V40z, 2 x 1.8Ghz Opteron, 8GB RAM, 6x 146GB SCSI || [[w:nl:Johan Hendrik van Dale|Johan Hendrik van Dale]]
<!--T:63-->
6% nuclear
<!--T:64-->
1% hydro/biomass/solar/other
<!--T:65-->
14% wind (Oncor/Ercot)
</translate>
|<translate>
<!--T:66-->
790,000 lb = 400 short tons = 360 metric tons
<small>= 0.23 * 77 kW * 8765.76 hr/yr * 2.1 lb CO<sub>2</sub>/kWh for coal</small>
<!--T:67-->
<small>+ 0.56 * 77 kW * 8765.76 hr/yr * 1.22lb CO<sub>2</sub>/kWh for nat gas</small>
<!--T:68-->
<small>+ 0.06 * 77 kW * 8765.76 hr/yr * 0 lb CO<sub>2</sub>/kWh for nuclear</small>
<!--T:69-->
<small>+ 0.15 * 77 kW * 8765.76 hr/yr * 0 lb CO<sub>2</sub>/kWh for renewables</small>
|?
</translate>
|-
|esams
|[[clematis]]||5/2005||Squid||<small>Fedora Core</small>||<small>Sun V20z, 2 x 2.2GHz Opteron, 3GB RAM, 2x 70GB SCSI || -
|Haarlem
2031 BE
Netherlands
|EvoSwitch ([https://www.evoswitch.com/ <translate><!--T:70--> Website</translate>])
<translate>
<!--T:71-->
|December 2008
|May 2016: < 10
May 2015: 10
|"a combination of wind power, hydro and biomass"
|0
|n.a.
</translate>
|-
|ulsfo
|[[fuchsia]]||5/2005||DNS, WWW as http://www2.knams.wikimedia.org, statistics||<small>Fedora Core</small>||<small>Sun V20z, 2 x 2.2GHz Opteron, 3GB RAM, 2x 70GB SCSI ||
|San Francisco, CA
|-
94124
|<small>hawtorn, iris, lily, mayflower, mint, sage</small> ||5/2005|| 6x Squid||<small>Fedora Core</small>||<small>Sun V20z, 2 x 2.2GHz Opteron, 3GB RAM, 2x 70GB SCSI || -
USA
|-
|UnitedLayer ([https://unitedlayer.com/ <translate><!--T:72--> Website</translate>])
|[[ragweed]]||5/2005||Squid/[[OTRS]]||<small>Fedora Core</small>||<small>Sun V20z, 2 x 2.2GHz Opteron, 3GB RAM, 2x 70GB SCSI || -
<translate>
|-
<!--T:73-->
|[[zedler]]||8/2005||[[Toolserver]]||<small>Solaris 10</small>||<small>Sun V40z, 2 x 2.2GHz Opteron, 8GB RAM, 6x 146GB SCSI || [[en:Johann Heinrich Zedler|Johann Heinrich Zedler]]
|June 2012
|- style="background-color:#e1e1e1;"
|May 2016: < 5
| 23 || colspan=6 | Korea Yahoo! cluster
May 2015: < 5
|-
|2016: <br/> 25% natural gas
|[[amaryllis]]||9/2005||NFS||<small>Fedora Core</small>||<small>HP DL385, 2x Opteron 250 (2.4GHz), 8GB RAM, 6x 146GB disk</small>||-
|-
|[[dryas]]||9/2005||MySQL master||<small>Fedora Core</small>||<small>HP DL385, 2x Opteron 250 (2.4GHz), 8GB RAM, 6x 146GB disk</small>||-
|-
|[[henbane]]||9/2005||MySQL slave||<small>Fedora Core</small>||<small>HP DL385, 2x Opteron 250 (2.4GHz), 8GB RAM, 6x 146GB disk</small>||-
|-
|yf1000 ... yf1004||9/2005||5x Squid||<small>Fedora Core</small>||<small>HP DL140, 2x Opteron 246 (2.0GHz), 4GB RAM, 1x 80GB HDD</small>||
|-
|yf1005 ... yf1016||9/2005||12x Apache ||<small>Fedora Core</small>||<small>HP DL140, 2x Opteron 246 (2.0GHz), 4GB RAM, 1x 80GB HDD</small>||-
|-
|yf1017||9/2005|| Search server ||<small>Fedora Core</small>||<small>HP DL140, 2x Opteron 246 (2.0GHz), 4GB RAM, 1x 80GB HDD</small>||-
|-
|yf1018||9/2005|| LVS load balancer ||<small>Fedora Core</small>||<small>HP DL140, 2x Opteron 246 (2.0GHz), 4GB RAM, 1x 80GB HDD</small>||-
|-
|yf1019||9/2005|| Squid ||<small>Fedora Core</small>||<small>HP DL140, 2x Opteron 246 (2.0GHz), 4GB RAM, 1x 80GB HDD</small>||-
|- style="background-color:#e1e1e1;"
| 240 || colspan=6 | Total
|- style="background-color:#a0a0dd;"
|}
<!--T:74-->
==Old servers==
23% nuclear
The old web servers, which are currently not in service, were also named after historical encyclopedists:
* "Pliny" ([[en:Pliny the Elder|Pliny the Elder]])
* "Geoffrin" ([[en:Marie Thérèse Rodet Geoffrin|Marie Thérèse Rodet Geoffrin]])
<!--T:75-->
==Donations==
30% renewable
While Wikipedia is free in both the "free speech" and "no charge" senses of the word, running the web site does cost money. You can help with purchasing new server hardware by donating to the non-profit [[Wikimedia Foundation]]: '''http://wikimediafoundation.org/wiki/Fundraising'''
<!--T:76-->
6% hydro
<!--T:77-->
17% unspecified (PG&E)
</translate>
|<translate><!--T:78-->
13,000 lb = 6.7 short tons = 6.1 metric tons (+ unspecified)
<small>= 0.00 * 5 kW * 8765.76 hr/yr * 2.1 lb CO<sub>2</sub>/kWh for coal</small>
<!--T:79-->
==Status and problems==
<small>+ 0.25 * 5 kW * 8765.76 hr/yr * 1.22lb CO<sub>2</sub>/kWh for nat gas</small>
<!--T:80-->
You can check one of the following sites if you want to know if the Wikimedia servers are overloaded, or if you just want to see how they are doing.
<small>+ 0.23 * 5 kW * 8765.76 hr/yr * 0 lb CO<sub>2</sub>/kWh for nuclear</small>
<!--T:81-->
*[http://ganglia.wikimedia.org/ Wikimedia server Ganglia monitoring]
<small>+ 0.36 * 5 kW * 8765.76 hr/yr * 0 lb CO<sub>2</sub>/kWh for hydro/renewable</small>
*[http://noc.wikimedia.org/~midom/squidmatrix/ Squid matrix] (down?)
<!--T:82-->
<small>+ 0.17 * 5 kW * 8765.76 hr/yr * ? lb CO<sub>2</sub>/kWh for unspecified</small>
</translate>
|?
|-
|eqsin
|Singapore
|Equinix ([https://www.equinix.com/ <translate><!--T:84--> Website</translate>])
|?
|?
|?
|?
|?
|-
|drmrs
|Marseille
|Digital Realty ([https://www.digitalrealty.com/ <translate><!--T:114--> Website</translate>])
|?
|?
|?
|?
|?
|}
<translate>
If you are seeing errors in real time, visit #wikimedia-tech on irc.freenode.net. Check the topic to see if someone is already looking into the problem you are having. If not, please report your problem to the channel. It would be helpful if you could report specific symptoms, including the exact text of any error messages, what you were doing right before the error, and what server(s) are generating the error, if you can tell. The #wikipedia channel may be better populated and the topic there may contain more updated information about the status of the problem. (But this channel is for general conversation about Wikipedia.)
== See also == <!--T:90-->
=== More hardware info === <!--T:91-->
If you don't use IRC, or don't get a reply, you can also send e-mail to noc@wikimedia.org. Someone will create a report in [http://bugzilla.wikimedia.org Bugzilla] (under the "Issues" product) so others will know the problem has been reported, and what the status is. You can [http://bugzilla.wikimedia.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=&product=Issues&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&keywords_type=allwords&keywords=&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&bug_status=RESOLVED&bug_status=VERIFIED&bug_status=CLOSED&emailassigned_to1=1&emailtype1=substring&email1=&emailassigned_to2=1&emailreporter2=1&emailcc2=1&emailtype2=substring&email2=&bugidtype=include&bug_id=&votes=&chfieldfrom=&chfieldto=Now&chfieldvalue=&cmdtype=doit&order=Reuse+same+sort+as+last+time&field0-0-0=noop&type0-0-0=noop&value0-0-0= check here] to see if your problem has already been reported.
<!--T:92-->
If you're wondering if it's only you experiencing this problem, you can check the following sites. Unfortunately, the Wikimedia administrators do not monitor these sites for problems.
* <tvar name=wt>[[wikitech:Clusters]]</tvar> – technical and usually more up-to-date information on the Wikimedia clusters
*[http://www.thewritingpot.com/wikistatus/ WikiStatus]
*[http://openfacts.berlios.de/index-en.phtml?title=Wikipedia_Status Wikipedia Status - OpenFacts]
=== Admin logs === <!--T:93-->
If you are getting a "connection refused" error, that is a squid problem. Determine which IP address you are trying to connect to, and ask someone to look at that host.
<!--T:94-->
==See also==
* [[<tvar name="1">wikitech:Server admin log</tvar>|Server admin log]] – Documents server changes (especially software changes)
{{Commons|Category:Wikimedia servers}}
=== Offsite traffic pages === <!--T:95-->
===More hardware info===
*[[Wikimedia servers/hardware orders|Recent hardware orders]], accounting summary
*[[Wikimedia partners and hosts]]
*[[en:Wikipedia:Technical_FAQ#How_about_the_hardware.3F|Technical FAQ - How about the hardware?]]
<!--T:96-->
===Admin logs===
* [<tvar name="1">https://www.semrush.com/website/wikipedia.org/overview/</tvar> Semrush: Wikipedia.org Traffic Ranking and Analysis]
*[http://wikitech.leuksman.com/view/Server_admin_log Server admin log] - Documents server changes (especially software changes)
== Historical information == <!--T:99-->
===Offsite traffic pages===
</translate>
*[http://www.alexa.com/data/details/traffic_details?&range=1y&size=large&y=t&url=wikipedia.org Alexa traffic rank]
* <translate><!--T:100--> [[<tvar name="1">Special:MyLanguage/Cache strategy</tvar>|Cache strategy]] (draft from 2006)</translate>
===Long-term planning===
* <translate><!--T:109--> [[<tvar name="2">Special:MyLanguage/PHP caching and optimization</tvar>|PHP caching and optimization]] (draft from 2007)</translate>
*[http://mail.wikipedia.org/mailman/listinfo/wikitech-l wikitech-l mailing list]
* <translate><!--T:110--> [[<tvar name="3">Special:MyLanguage/Hardware orders up to 2007</tvar>|Hardware orders up to 2007]]</translate>
*[http://www.livejournal.com/community/wikitech/ Wikimedia technical discussions' Journal]
*[https://wikitech.leuksman.com/view/Main_Page wikitech.leuksman.com]
<translate>
==Useful information about other sites==
== References == <!--T:103-->
*Evolution of LiveJournal systems:
</translate>
**04/2004 [http://www.danga.com/words/2004_mysqlcon/ MySQLCon 2004] PDF/SXI
<references />
**07/2004 [http://www.danga.com/words/2004_oscon/ OSCON 2004] PDF/SXI
**11/2004 [http://www.danga.com/words/2004_lisa/ LISA 2004] PDF/SXI
**04/2005 [http://www.danga.com/words/2005_mysqlcon/ MySQLCon 2005] PDF/PPT/SXI
** journals to watch for system details: [http://www.livejournal.com/userinfo.bml?user=brad Brad (Fitzpatrick) ] [http://www.livejournal.com/userinfo.bml?user=lj_backend lj_backend] [http://www.livejournal.com/userinfo.bml?user=lj_maintenance lj_maintenance]
*[http://www.search3w.com/Siteresources/data/MediaArchive/files/Google%2015000%20servers%20secrest.pdf Google cluster architecture] (PDF)
*[http://www.vbmysql.com/uc2004/ MySQL User's Conference 2004 blog highlights]
[[Category:Wikimedia hardware]]
[[Category:Wikimedia servers administration{{#translation:}}]]
| 