if you are doing computations using hadoop on a miniscale yes this hardware is good enough.
Normally hadoop clusters are pre-occupied with the heavy loads so they are not shared for multiple usage unless your utilization of hadoop is on lower side and then you want to reuse the hardware. On Mon, Jun 4, 2012 at 5:52 PM, praveenesh kumar <praveen...@gmail.com>wrote: > On a very high level... we would be utilizing cluster not only for hadoop > but for other I/O bound or in-memory operations. > That is the reason we are going for SAS hard disks. And we also need to > perform lots of computational tasks for which we have RAM kept to 32 GB, > which can be increased. So on a high level just wanted to know does these > hardware specs make sense ? > > Regards, > Praveenesh > > On Mon, Jun 4, 2012 at 5:46 PM, Nitin Pawar <nitinpawar...@gmail.com> > wrote: > > > if you tell us the purpose of this cluster, then it would be helpful to > > tell exactly how good it is > > > > On Mon, Jun 4, 2012 at 3:57 PM, praveenesh kumar <praveen...@gmail.com > > >wrote: > > > > > Hello all, > > > > > > I am looking forward to build a 5 node hadoop cluster with the > following > > > configurations per machine. -- > > > > > > 1. Intel Xeon E5-2609 (2.40GHz/4-core) > > > 2. 32 GB RAM (8GB 1Rx4 PC3) > > > 3. 5 x 900GB 6G SAS 10K hard disk ( total 4.5 TB storage/machine) > > > 4. Ethernet 1GbE connection > > > > > > I would like the experts to please review it and share if this sounds > > like > > > a optimal/deserving hadoop hardware configuration or not ? > > > I know without knowing the actual use case its not worth commenting, > but > > > still in general I would like to have the views. Also please suggest > if I > > > am missing something. > > > > > > Regards, > > > Praveenesh > > > > > > > > > > > -- > > Nitin Pawar > > > -- Nitin Pawar
