Distributed Replicated Storage Across Four Storage Nodes With GlusterFS On Ubuntu 9.1
 

table align=left cellpadding=0 cellspacing=0 width=42 height=40 style=margin-top:0px;margin-bottom:0px;trtdimg class=teaser-image-odd src=http://images.howtoforge.com/images/teaser/ubuntu.gif width=39 height=40 alt= //td/tr/tablepbDistributed Replicated Storage Across Four Storage Nodes With GlusterFS On Ubuntu 9.10/b/ppThis tutorial shows how to combine four single storage servers (running Ubuntu 9.10) to a distributed replicated storage with GlusterFS. Nodes 1 and 2 (span class=systemreplication1/span) as well as 3 and 4 (span class=systemreplication2/span) will mirror each other, and span class=systemreplication1/span and span class=systemreplication2/span will be combined to one larger storage server (distribution). Basically, this is RAID10 over network. If you lose one server from span class=systemreplication1/span and one from span class=systemreplication2/span,the distributed volume continues to work. The client system (Ubuntu9.10 as well) will be able to access the storage as if it was a localfilesystem. GlusterFS is a clustered file-system capable of scaling toseveral peta-bytes. It aggregates various storage bricks overInfiniband RDMA or TCP/IP interconnect into one large parallel networkfile system. Storage bricks can be made of any commodity hardware suchas x86_64 servers with SATA-II RAID and Infiniband HBA./p

عفوا ,,, تم حجب الروابط و المرفقات عن الزوار و الأعضاء غير المتفاعلين [ للتسجيل اضغط هنا ]