Just request a quote from the leading and also local vendors. Tell them about the volume and the access pattern you have in mind and collect the offerings. Than you compare the prices. You should consider space (in the data center) network architecture and energy cosumption as well as heat generation in such a cluster which handles some PB on the long run. Have a look into the book: http://www.amazon.de/Hadoop-Operations-Eric-Sammer/dp/1449327052 Cheers, Mirko
2014-07-10 11:10 GMT+02:00 YIMEN YIMGA Gael <gael.yimen-yi...@sgcib.com>: > Thank for your return Mirko, > > > > In my case, I can consider *compression factor of *8* according to the > service in charge of it. > > > > Data, I’m dealing with are : logs only. But it’s many types of logs > (printing logs, USB logs, Remote access logs, Active Directory logs, > database servers logs, Web servers logs, Antivirus logs, etc.) > > I precise that in my case it’s only logs that are stored. Sometime we > could have CSV files. But no videos or images are considered here. > > > > Any advice according to that specific type of data? > > What are the reasons to consider servers with 12 HDD (3TB) per server? > Knowing that, I prefer the LOW-COST. > > What could be the price of a LOW-COST server with 12HDD (3TB) ? > > > > Regards > > > > *From:* Mirko Kämpf [mailto:mirko.kae...@gmail.com] > *Sent:* Thursday 10 July 2014 11:01 > > *To:* user@hadoop.apache.org > *Subject:* Re: Need to evaluate a cluster > > > > I multiply by 1.3 which means I add 30% of the estimated amount to have > reserved capacity for intermediate data. > > In your case with approx. 2TB per day I think, data nodes with 1 to 3 > discs are not a good idea. You should consider servers with more discs and > than add one per week. Start with 10 servers and 12 HDD (3TB) per server. > This allows you to handle approx. 35 TB raw uncompressed data. You have to > evaluated compression in your special case. It can be high, but also not > very high, if raw data is already compressed somehow. What data are you > dealing with? > > Text, messages, logs or more binary data like images, mp3 oder video > formats? > > Cheers, > Mirko > > > > 2014-07-10 10:43 GMT+02:00 YIMEN YIMGA Gael <gael.yimen-yi...@sgcib.com>: > > Hi, > > > > What does « 1.3 for overhead » mean in this calculation ? > > > > Regards > > > > *From:* Mirko Kämpf [mailto:mirko.kae...@gmail.com] > *Sent:* Wednesday 9 July 2014 18:09 > > > *To:* user@hadoop.apache.org > *Subject:* Re: Need to evaluate a cluster > > > > Hello, > > > > if I follow your numbers I see one missing fact: *What is the number of > HDDs per DataNode*? > > Let's assume you use machines with 6 x 3TB HDDs per box, you would need > about 60 DataNodes > > per year (0.75 TB per day x 3 for replication x 1.3 for overhead / ( nr of > HDDs per node x capacity per HDD )). > With 12 HDD you would only need 30 servers per year. > How did you calculate the number of 367 datanodes? > > > > Cheers, > > Mirko > > > > 2014-07-09 17:59 GMT+02:00 YIMEN YIMGA Gael <gael.yimen-yi...@sgcib.com>: > > Hello Dear, > > > > I made an estimation of a number of nodes of a cluster that can be > supplied by 720GB of data/day. > > My estimation gave me *367 datanodes* in a year. I’m a bit afraid by that > amount of datanodes. > > The assumptions, I used are the followings : > > > > - Daily supply (feed) : 720GB > > - HDFS replication factor: 3 > > - Booked space for each disk outside HDFS: 30% > > - Size of a disk: 3TB. > > > > I have two questions. > > > > First, I would like to know if my assumptions are well taken? > > Secondly, could someone help me to evaluate that cluster, to let me be > sure that my results are not to excessive, please ? > > > > Standing by for your feedback > > > > Warm regard > > ************************************************************************* > This message and any attachments (the "message") are confidential, > intended solely for the addressee(s), and may contain legally privileged > information. > Any unauthorised use or dissemination is prohibited. E-mails are > susceptible to alteration. > Neither SOCIETE GENERALE nor any of its subsidiaries or affiliates shall > be liable for the message if altered, changed or > falsified. > Please visit http://swapdisclosure.sgcib.com for important information > with respect to derivative products. > ************ > Ce message et toutes les pieces jointes (ci-apres le "message") sont > confidentiels et susceptibles de contenir des informations couvertes > par le secret professionnel. > Ce message est etabli a l'intention exclusive de ses destinataires. Toute > utilisation ou diffusion non autorisee est interdite. > Tout message electronique est susceptible d'alteration. > La SOCIETE GENERALE et ses filiales declinent toute responsabilite au > titre de ce message s'il a ete altere, deforme ou falsifie. > Veuillez consulter le site http://swapdisclosure.sgcib.com afin de > recueillir d'importantes informations sur les produits derives. > ************************************************************************* > > > > >
