{"id":37785,"date":"2019-10-31T22:19:45","date_gmt":"2019-10-31T19:19:45","guid":{"rendered":"https:\/\/prohoster.info\/blog\/kak-my-uporolis-po-energoeffektivnosti-mashzala\/"},"modified":"2019-10-31T22:19:45","modified_gmt":"2019-10-31T19:19:45","slug":"kak-my-uporolis-po-energoeffektivnosti-mashzala","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/kak-my-uporolis-po-energoeffektivnosti-mashzala","title":{"rendered":"Si u angazhuam p\u00ebr efikasitetin energjetik t\u00eb sall\u00ebs s\u00eb makin\u00ebs","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"Si
\n
\nI dedicate this post to those who lied on their certificates, which almost led us to install Bengal lights in our halls.<\/p>\n

The story is more than four years old, but I\u2019m publishing it now because the NDA has expired. Back then, we realized we had almost entirely filled the data center (which we lease), but its energy efficiency didn't really improve. The hypothesis was that the more we filled it, the better it would be, as the engineering load would be distributed among all. However, it turned out we were deceiving ourselves in that regard, and with good loading, there were some losses. We worked on many fronts, but our brave team focused on cooling. <\/p>\n

The real life of a data center is a bit different from what\u2019s in the project. Constant adjustments from the operations team to increase efficiency and optimization of settings for new tasks. Take the mythical average rack, for example. In practice, it doesn\u2019t exist; the load distribution is uneven, with some areas dense and others sparse. Thus, we had to reconfigure some things for better energy efficiency.<\/p>\n

Qendra jon\u00eb t\u00eb dh\u00ebnash Kompressor \u00ebsht\u00eb e nevojshme p\u00ebr nj\u00eb gam\u00eb t\u00eb gjer\u00eb klient\u00ebsh. Prandaj, aty, mes raftesh normale prej dy deri n\u00eb kat\u00ebr kilovat, mund t\u00eb gjeni plot\u00ebsisht nj\u00eb raft prej 23 kilovat\u00ebsh ose m\u00eb shum\u00eb. Si pasoj\u00eb, kondicioner\u00ebt ishin rregulluar p\u00ebr t'i ftohur ata, dhe ajri, p\u00ebrmes raftesh m\u00eb pak t\u00eb fuqish\u00ebm, thjesht kalonte p\u00ebrmes. <\/p>\n

Hipoteza e dyt\u00eb ishte se korridoret e ngrohta dhe t\u00eb ftohta nuk p\u00ebrzihen. Pas matjeve, mund t\u00eb them se kjo \u00ebsht\u00eb nj\u00eb iluzion, dhe aerodinamika e v\u00ebrtet\u00eb ndryshon nga modeli n\u00eb \u00e7do aspekt. <\/a><\/noindex><\/p>\n

Inspektimi<\/h3>\n

\nFillimisht, ne filluam t\u00eb shqyrtonim rrjedhat e ajrit n\u00eb sallat. Pse shkuam atje? Sepse e kuptonim se Qendra e t\u00eb Dh\u00ebnave \u00ebsht\u00eb projektuar p\u00ebr pes\u00eb deri n\u00eb gjasht\u00eb kW p\u00ebr raft, por e dinim se ato n\u00eb fakt variojn\u00eb nga 0 deri n\u00eb 25 kW. Rregullimi i gjith\u00eb k\u00ebsaj me pllaka \u00ebsht\u00eb pothuajse e pamundur: matjet e para treguan se ato l\u00ebshonin pothuajse nj\u00ebsoj. Nd\u00ebrsa pllakat p\u00ebr 25 kW n\u00eb t\u00eb v\u00ebrtet\u00eb nuk ekzistojn\u00eb, ato duhet t\u00eb jen\u00eb jo thjesht t\u00eb zbraz\u00ebta, por me vakum t\u00eb l\u00ebngsh\u00ebm. <\/p>\n

Kemi bler\u00eb nj\u00eb anemomet\u00ebr dhe filluam t\u00eb matim flukset midis rafteve dhe mbi raftet. N\u00eb t\u00eb v\u00ebrtet\u00eb, duket se duhet t\u00eb punosh sipas GOST-it dhe shum\u00eb standardeve, t\u00eb cilat jan\u00eb t\u00eb v\u00ebshtira p\u00ebr t'u realizuar pa ndaluar sall\u00ebn e makinave. Na interesonte jo sakt\u00ebsia, por pamja parimore. Pra, mat\u00ebm p\u00ebraf\u00ebrsisht.<\/p>\n

Nga matja e 100 p\u00ebrqind t\u00eb ajrit q\u00eb del nga pllaket, 60 p\u00ebrqind arrin n\u00eb rafte, nd\u00ebrsa e gjitha e tjera kalon p\u00ebrmes. Kjo ndodh p\u00ebr shkak t\u00eb stativ\u00ebve t\u00eb r\u00ebnd\u00eb me 15\u201325 kW, mbi t\u00eb cilat \u00ebsht\u00eb nd\u00ebrtuar ftohja. <\/p>\n

Nuk mund t\u00eb ndalim kondicioner\u00ebt, sepse mbi stativ\u00ebt e ngroht\u00eb n\u00eb zon\u00ebn e server\u00ebve t\u00eb sip\u00ebrm do t\u00eb b\u00ebhet krejt\u00ebsisht nxeht\u00eb. N\u00eb at\u00eb moment kuptojm\u00eb se duhet t\u00eb izolojm\u00eb di\u00e7ka nga di\u00e7ka tjet\u00ebr, q\u00eb ajri t\u00eb mos k\u00ebrcente nga nj\u00eb radh\u00eb n\u00eb nj\u00eb tjet\u00ebr dhe q\u00eb transferimi i nxeht\u00ebsis\u00eb n\u00eb bllok t\u00eb ndodhte gjithsesi. <\/p>\n

P\u00ebrve\u00e7 k\u00ebsaj, shqet\u00ebsohemi p\u00ebr pyetjen, a \u00ebsht\u00eb kjo financiare t\u00eb arsyeshme. <\/p>\n

Me befas n\u00eb m\u00ebnyr\u00eb t\u00eb papritur se konsumi i energjis\u00eb s\u00eb Qendr\u00ebs s\u00eb T\u00eb Dh\u00ebnave n\u00eb p\u00ebrgjith\u00ebsi e kemi, por thjesht nuk mund t\u00eb llogarisim fancoil-at p\u00ebr sall\u00ebn specifike. Dometh\u00ebn\u00eb, analitikisht mundemi, por n\u00eb fakt \u2014 jo. Dhe nuk jemi n\u00eb gjendje t\u00eb vler\u00ebsojm\u00eb kursimin. Detyra b\u00ebhet gjithnj\u00eb e m\u00eb interesante. N\u00ebse kursejm\u00eb 10% t\u00eb fuqis\u00eb s\u00eb kondicioner\u00ebve \u2014 sa para mund t\u00eb ndajm\u00eb p\u00ebr izolimin? Si ta llogarisim?<\/p>\n

Shkuam tek automatizuesit, t\u00eb cil\u00ebt kishin p\u00ebrfunduar sistemin e monitorimit. Faleminderit djemve: t\u00eb gjitha sensor\u00ebt i kishin, duhej vet\u00ebm t\u00eb shkruhej kodi. Ata filluan t\u00eb nxjerrin ve\u00e7mas chillernat, UPS, ndri\u00e7imin. Me pajisjen e re pati mund\u00ebsin\u00eb t\u00eb shihnim se si ndryshonte situata sipas element\u00ebve t\u00eb sistemit.<\/p>\n

Eksperimente me perde<\/h3>\n

\nParalelisht fillojm\u00eb eksperimente me perde (ndarje). Vendosim t\u2019i montojm\u00eb ato n\u00eb shtylla t\u00eb kanaleve t\u00eb kabllove (nuk kemi vend tjet\u00ebr), fatmir\u00ebsisht ato duhet t\u00eb jen\u00eb t\u00eb lehta. Shpejt u vendos\u00ebm p\u00ebr mbulesat ose shiritat. <\/p>\n

\"Si
\n
\n\"Si
\n
\nProblemi \u00ebsht\u00eb se deri tani kemi punuar me shum\u00eb furnizues. T\u00eb gjith\u00eb kan\u00eb zgjidhje p\u00ebr DC-t\u00eb e tyre, por p\u00ebr DC-t\u00eb komerciale, n\u00eb thelb nuk ka zgjidhje t\u00eb gatshme. Klient\u00ebt tan\u00eb vazhdojn\u00eb t\u00eb vijn\u00eb e ikin vazhdimisht. Ne jemi nj\u00eb nga pak DC-t\u00eb 't\u00eb r\u00ebnda' pa kufizime n\u00eb gjer\u00ebsin\u00eb e raftit, me mund\u00ebsin\u00eb p\u00ebr t\u00eb pranuar k\u00ebta server\u00eb-gjak\u00ebsit deri n\u00eb 25 kW. Asnj\u00eb planifikim t\u00eb infrastruktur\u00ebs paraprakisht. Pra, n\u00ebse do t\u00eb merrnim sisteme modulare t\u00eb kafaz\u00ebve nga furnizuesit, gjithmon\u00eb do t\u00eb kishim boshll\u00ebqe p\u00ebr dy muaj. K\u00ebshtu q\u00eb asnj\u00ebher\u00eb sallat tona nuk do t\u00eb jen\u00eb energjikisht efikase n\u00eb thelb. <\/p>\n

Vendos\u00ebm ta b\u00ebjm\u00eb vet\u00eb, fatmir\u00ebsisht kemi inxhinier\u00ebt tan\u00eb. <\/p>\n

E para q\u00eb mor\u00ebm, ishin shirita nga frigorifer\u00ebt industrial\u00eb. K\u00ebto jan\u00eb disa cop\u00ebza t\u00eb fleksib\u00ebl poliethylene q\u00eb mund t'i p\u00ebrdor\u00ebsh. Sigurisht i keni par\u00eb diku n\u00eb hyrje t\u00eb seksionit t\u00eb mishit t\u00eb dyqaneve m\u00eb t\u00eb m\u00ebdha. Filluam t\u00eb k\u00ebrkojm\u00eb materiale jo toksike dhe jo t\u00eb ndezshme. Gjet\u00ebm, blen\u00eb p\u00ebr dy rreshta. I var\u00ebm, filluam t\u00eb shihnim se \u00e7far\u00eb rezultati kishte. <\/p>\n

Kishim, se ishte ndryshe. Por doli krejt shum\u00eb, tamam shum\u00eb ndryshe. Ata fillojn\u00eb, si makarona, t\u00eb shp\u00ebrndahen n\u00eb rrjedha. Gjet\u00ebm bandat magnetike si ato p\u00ebr frigoriferin. I ngjit\u00ebm ato n\u00eb k\u00ebto banda, i ngjit\u00ebm nj\u00ebri me tjetrin, muri doli mjaft monolit. <\/p>\n

Filluan t\u00eb mendojn\u00eb se \u00e7far\u00eb do t\u00eb ishte n\u00eb sall\u00eb. <\/p>\n

Shkuam te nd\u00ebrtuesit, u treguam projektin ton\u00eb. Ata shikojn\u00eb, thon\u00eb: perdet tuaja jan\u00eb disi shum\u00eb t\u00eb r\u00ebnda. 700 kilogram\u00eb n\u00eb t\u00eb gjith\u00eb sall\u00ebn e makinave. Shkoni, thon\u00eb, njer\u00ebz t\u00eb mir\u00eb, te djalli. M\u00eb sakt\u00eb, te ekipi i SCS-s\u00eb. Le t\u00eb llogarisin sa noodles kan\u00eb n\u00eb shporta, sepse 120 kg p\u00ebr met\u00ebr katror \u00ebsht\u00eb maksimumi. <\/p>\n

SCS thot\u00eb: kujtoni, erdhi nj\u00eb klient i madh te ne? Ai ka dhjet\u00ebra mij\u00ebra porte n\u00eb nj\u00eb sall\u00eb. N\u00eb skajet e sall\u00ebs s\u00eb makinave \u00ebsht\u00eb mir\u00eb, por m\u00eb af\u00ebr lidhjes nuk do funksionoj\u00eb: shportat do bien. <\/p>\n

Nd\u00ebrtuesit k\u00ebrkuan gjithashtu nj\u00eb certifikat p\u00ebr materialin. Do t\u00eb theksoj se ne deri at\u00ebher\u00eb kishim punuar me fjal\u00ebn e ndershme t\u00eb furnizuesit, pasi ishte vet\u00ebm nj\u00eb test. K\u00ebrkuam nga ky furnizues, i tham\u00eb: OK, jemi gati p\u00ebr beta, le t\u00eb marrim t\u00eb gjitha dokumentet. Ata d\u00ebrgojn\u00eb di\u00e7ka q\u00eb nuk \u00ebsht\u00eb n\u00eb format t\u00eb caktuar. <\/p>\n

Ne heqim: d\u00ebgjojm\u00eb, por nga e keni marr\u00eb k\u00ebt\u00eb dokument? Ata: na e ka d\u00ebrguar prodhuesi yn\u00eb kinez si p\u00ebrgjigje ndaj k\u00ebrkesave. Nga ky dokument, ky materiale nuk digjet fare. <\/p>\n

N\u00eb at\u00eb moment e kuptuam se ishte koha p\u00ebr t\u00eb ndaluar dhe p\u00ebr t\u00eb kontrolluar faktet. Shkruajm\u00eb te vajzat e siguris\u00eb zjarrmore t\u00eb Qendr\u00ebs s\u00eb Datan\u00eb, ato na p\u00ebrmendin laboratorin q\u00eb kontrollon djegshm\u00ebrin\u00eb. \u00c7mimet dhe afatet jan\u00eb mjaft t\u00eb arsyeshme (n\u00eb t\u00eb v\u00ebrtet\u00eb, ne e mallkuam gjith\u00e7ka derisa b\u00ebm\u00eb dokumentacionin e nevojsh\u00ebm). Aty shkenc\u00ebtar\u00ebt thon\u00eb: sillni materialin dhe do t\u00eb b\u00ebjm\u00eb prova. <\/p>\n

N\u00eb p\u00ebrfundim t\u00eb provave shkruhet se nga nj\u00eb kilogram substance mbetet pluhur rreth 50 gram. E gjith\u00eb e tjerat digjet fort, rrjedh dhe p\u00ebrkrah shum\u00eb mir\u00eb djegien n\u00eb nj\u00eb pishin\u00eb. <\/p>\n

E kuptojm\u00eb \u2014 mir\u00eb q\u00eb nuk e blejm\u00eb. Fillojm\u00eb t\u00eb k\u00ebrkojm\u00eb nj\u00eb material tjet\u00ebr. <\/p>\n

Gjet\u00ebm polikarbonat. Ai doli m\u00eb i fort\u00eb. Fleta transparente \u2014 dy mm, dyert \u2014 prej kat\u00ebr mm. N\u00eb thelb, \u00ebsht\u00eb akrilik. S\u00eb bashku me prodhuesin fillojm\u00eb bised\u00ebn p\u00ebr sigurin\u00eb ndaj zjarrit: na jepni nj\u00eb certifikat\u00eb. Ata e d\u00ebrgojn\u00eb. N\u00ebnshkruar nga i nj\u00ebjti institut. Ne i telefonojm\u00eb atyre, themi: \u00e7'a b\u00ebni, njer\u00ebz, a e keni provuar k\u00ebt\u00eb?<\/p>\n

Ata thon\u00eb: po, e kontrolluan. Fillimisht e dogj\u00ebn te vetja, pastaj e soll\u00ebn vet\u00ebm p\u00ebr teste. Nga nj\u00eb kilogram material, n\u00ebse e digjni me nj\u00eb piskatore, mbetet rreth 930 gram pluhur. Ai shkrihet dhe rrjedh, por nj\u00eb pellg nuk do t\u00eb digjet. <\/p>\n

Menj\u00ebher\u00eb kontrollojm\u00eb magnet\u00ebt tan\u00eb (ata jan\u00eb n\u00eb nj\u00eb mb\u00ebshtjell\u00ebs polimer). P\u00ebr habi, digjen keq. <\/p>\n

Nd\u00ebrtimi<\/h3>\n

\nNga kjo fillojm\u00eb t\u00eb mbledhim. Polikarbonati \u00ebsht\u00eb i shk\u00eblqyer, sepse \u00ebsht\u00eb m\u00eb i leht\u00eb se polietileni, i cili \u00ebsht\u00eb shum\u00eb m\u00eb i v\u00ebshtir\u00eb p\u00ebr t'u p\u00ebrkulur. Megjithat\u00eb, ata sjellin flet\u00ebt 2.5 me 3 metra, dhe furnizuesi nuk i intereson \u00e7far\u00eb t\u00eb b\u00ebni me to. Ne kemi nevoj\u00eb p\u00ebr 2.8 me gjer\u00ebsi 20\u201325 centimetra. Dyert i d\u00ebrguam n\u00eb zyrat q\u00eb presin flet\u00ebt si\u00e7 duhet. Dhe lamelat i preknim vet\u00eb. Procesi i prerjes \u00ebsht\u00eb rreth dy her\u00eb m\u00eb i shtrenjt\u00eb se flet\u00ebt.<\/p>\n

K\u00ebtu \u00ebsht\u00eb rezultati:<\/p>\n

\"Si
\n
\nP\u00ebrfundimi \u2014 sistemi i kafazimit rikuperohet p\u00ebr m\u00eb pak se nj\u00eb vit. Kjo \u00ebsht\u00eb, ne kursejm\u00eb rreth 200\u2013250 KWt vazhdimisht nga fuqia e fankojeve. Akoma di\u00e7ka n\u00eb chillere, sa sakt\u00ebsisht \u2014 nuk e dim\u00eb. Server\u00ebt thithin me nj\u00eb shpejt\u00ebsi konstante, fankojet fryjn\u00eb. Chillernat ndiz\u00ebn dhe fik\u00ebn si nj\u00eb greb\u00ebz: \u00ebsht\u00eb e v\u00ebshtir\u00eb t\u00eb nxjerr\u00ebsh t\u00eb dh\u00ebnat nga ajo. Nuk mund t\u00eb ndalosh sall\u00ebn e makineris\u00eb p\u00ebr teste. <\/p>\n

Jemi t\u00eb lumtur q\u00eb n\u00eb at\u00eb koh\u00eb kishte nj\u00eb rregull q\u00eb modulat 5x5 t\u00eb vendoseshin n\u00eb m\u00ebnyr\u00eb q\u00eb konsumimi i tyre mesatar t\u00eb ishte maksimumi gjasht\u00eb kW. K\u00ebshtu, ngrohja nuk \u00ebsht\u00eb e p\u00ebrqendruar n\u00eb nj\u00eb ishull, por e shp\u00ebrndar\u00eb n\u00eb sall\u00ebn e makinerive. Por ka nj\u00eb situat\u00eb ku 10 shtylla 15-kilovat\u00ebsh jan\u00eb pran\u00eb, por p\u00ebrball\u00eb ka nj\u00eb ruajt\u00ebs. Ai \u00ebsht\u00eb i ftoht\u00eb. Balancimi \u00ebsht\u00eb.<\/p>\n

Kudo ku nuk ka shtyll\u00eb \u2013 nevojitet nj\u00eb ndarje deri n\u00eb dysheme. <\/p>\n

Dhe gjithashtu disa nga klient\u00ebt tan\u00eb jan\u00eb t\u00eb izoluar me grila. Me ta gjithashtu kishte disa ve\u00e7ori. <\/p>\n

N\u00eb lamel\u00eb priteshin, sepse gjer\u00ebsia e shtyllave nuk \u00ebsht\u00eb fikse, dhe periodiciteti i greb\u00ebrth\u00ebve t\u00eb montimit \u00ebsht\u00eb i p\u00ebrcaktuar: tre-kat\u00ebr cm ose djathtas, ose majtas gjithmon\u00eb do t\u00eb ket\u00eb. N\u00ebse ke nj\u00eb bllok 600 n\u00ebn vendin p\u00ebr shtyll\u00eb, ka 85 p\u00ebr qind gjasa q\u00eb ai t\u00eb mos rrij\u00eb. Dhe lamel\u00ebt e shkurtra dhe t\u00eb gjata bashk\u00ebjetojn\u00eb dhe ngjiten. Ndonj\u00ebher\u00eb e presim lamel\u00ebn n\u00eb form\u00ebn e shkronj\u00ebs G sipas kontur\u00ebve t\u00eb shtyllave.<\/p>\n

\"Si
\n<\/p>\n

Sensor\u00ebt<\/h3>\n

\nPara t\u00eb reduktosh fuqin\u00eb e fancoil-ve, duhej t\u00eb kishim vendosur nj\u00eb monitorim t\u00eb sakt\u00eb t\u00eb temperatur\u00ebs n\u00eb pika t\u00eb ndryshme t\u00eb sall\u00ebs, p\u00ebr t\u00eb evituar surprizat. K\u00ebshtu lindi ideja p\u00ebr sensor\u00ebt pa tel. Me tel \u2014 p\u00ebr \u00e7do rresht duhej t\u00eb vendosnim nj\u00eb pajisje p\u00ebr kryq\u00ebzimin e k\u00ebtyre sensor\u00ebve dhe ndonj\u00ebher\u00eb p\u00ebr k\u00ebt\u00eb \u2014 zgjatje. Kjo shnd\u00ebrrohet n\u00eb nj\u00eb garland\u00eb. \u00cbsht\u00eb shum\u00eb e keqe. Dhe kur k\u00ebto tela hyjn\u00eb n\u00eb kafazet e klient\u00ebve, siguruar\u00ebt menj\u00ebher\u00eb shqet\u00ebsohen dhe k\u00ebrkojn\u00eb shpjegim me nj\u00eb certifikat\u00eb p\u00ebr at\u00eb q\u00eb kalon n\u00ebp\u00ebr k\u00ebto tela. Duhet t\u00eb kemi parasysh nervat e siguruar\u00ebve. P\u00ebr \u00e7udin\u00eb time, ata nuk shqet\u00ebsohen p\u00ebr sensor\u00ebt pa tel. <\/p>\n

Dhe akoma, raftet arrijn\u00eb dhe largohen. Nj\u00eb sensor me magnet \u00ebsht\u00eb m\u00eb i leht\u00eb p\u00ebr t'u rimontuar, sepse duhet ta varim \u00e7do her\u00eb m\u00eb lart ose m\u00eb posht\u00eb. N\u00ebse server\u00ebt jan\u00eb n\u00eb tre t\u00eb tretat e poshtme t\u00eb raftit \u2014 duhet ta varim m\u00eb posht\u00eb, dhe jo sipas standartit n\u00eb nj\u00eb met\u00ebr e gjysm\u00eb nga toka n\u00eb der\u00ebn e raftit n\u00eb koridori t\u00eb ftoht\u00eb. Atje, matja \u00ebsht\u00eb e kot\u00eb, duhet t\u00eb matim at\u00eb q\u00eb \u00ebsht\u00eb n\u00eb pajisje. <\/p>\n

Nj\u00eb sensor p\u00ebr tre rafte \u2013 m\u00eb shpesh nuk \u00ebsht\u00eb e nevojshme t\u00eb vendoset tjet\u00ebr. Temperatura nuk ndryshon. Kishim frik\u00eb se do t\u00eb kishte kalim ajri p\u00ebrmes vet\u00eb rafteve, \u2013 kjo nuk ndodhi. Por megjithat\u00eb ofrojm\u00eb pak m\u00eb shum\u00eb aj\u00ebr t\u00eb ftoht\u00eb se vlerat e llogaritura. B\u00ebm\u00eb hapje n\u00eb lamela 3, 7 dhe 12, mbi raft krijojm\u00eb nj\u00eb vrim\u00eb. N\u00eb t\u00eb vendosim anemometrin gjat\u00eb kontrollit: shikojm\u00eb se fluksi shkon aty ku duhet.<\/p>\n

\"Si
\n
\nM\u00eb pas vendos\u00ebm fije t\u00eb ndritshme: nj\u00eb praktik\u00eb e vjet\u00ebr p\u00ebr snajper\u00ebt. Duket e \u00e7uditshme, por lejon t\u00eb zbuloni nj\u00eb problem t\u00eb mundsh\u00ebm m\u00eb shpejt. <\/p>\n

\"Si
\n <\/p>\n

Arg\u00ebtues<\/h3>\n

\nNd\u00ebrsa ne po b\u00ebnim t\u00eb gjitha k\u00ebto n\u00eb heshtje, erdhi furnizuesi q\u00eb prodhon inxhinierin\u00eb p\u00ebr Qendrat e t\u00eb Dh\u00ebnave. Thot\u00eb: le t\u00eb vijm\u00eb dhe t'ju flasim p\u00ebr efikasitetin energjetik. Vijn\u00eb, fillojn\u00eb t\u00eb flasin p\u00ebr sall\u00ebn jo optimale, flukset e ajrit. Ne b\u00ebjm\u00eb nj\u00eb kok\u00eb n\u00eb shenj\u00eb miratimi. Sepse e kemi instaluar q\u00eb tre vjet. <\/p>\n

Ata\u00e7ik, p\u00ebr \u00e7do raft vendosin tre sensor\u00eb. Pamjet e monitorimit jan\u00eb fantastike, shum\u00eb t\u00eb bukura. M\u00eb shum\u00eb se gjysma e k\u00ebsaj zgjidhjeje \u00ebsht\u00eb \u00e7mimi \u2014 softi. N\u00eb nivelin e 'alarmit n\u00eb Zabbix', por pron\u00ebsor dhe shum\u00eb i shtrenjt\u00eb. Problemi \u00ebsht\u00eb se ata kan\u00eb sensor\u00ebt, softin, dhe pastaj k\u00ebrkojn\u00eb nj\u00eb kontraktor n\u00eb vend: nuk kan\u00eb ofrues t\u00eb tyre p\u00ebr kase.<\/p>\n

Zbulohet se punimet e tyre kushtojn\u00eb pes\u00eb deri n\u00eb shtat\u00eb her\u00eb m\u00eb shum\u00eb se sa kemi b\u00ebr\u00eb ne.<\/p>\n

Linke<\/h3>\n

<\/p>\n