0

I have several dedicated servers with unused space on them. 16GB+ RAM, some have SSD, other HDD. All have minimum 100Mbps connection. OS is Debian 7 64bit. There is no LAN connection between machines (only WAN).

I want to create self replicating storage, because it's hard to manually upload chosen files and remember where they are when you need these files. Something like S3 buckets would be perfect. It must me free (as in beer), relatively easy to install and have authentication between nodes (WAN connection only).

The files that I need to store are up to several gigabytes in size (50MB - 15GB backup *.tar.gz's).

Is hadoop or HDFS in it a good choice?

1 Answer 1

0

Hadoop isn't really meant for distributed FS access over the WAN. It's use is geared towards fast local networks with minimal latency.

If you're just syncing to a few remote locations where it's a 1-to-many data relationship (ie, data is produced in one location and you just want to distribute it for redundancy), you're probably better off just using rsync.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .