[BBLISA] Fwd: Re: Fileserver opinion

[Daniel Feenberg]

We have Netapp equipment, and it has done well for us, but the Netapp fan 
club on this list is seriously underestimating the relative cost of 
home-brew to Netapp equipment, and they should not be ridiculing the OP 
for considering a home-brew store.

A simple Linux or FreeBSD box with 12 2TB drives can be assembled for 
$3,000. I wouldn't put a RAID 5 on it (because Linux and FreeBSD don't do 
a good job of reconstruction after a drive failure - see
http://www.nber.org/sys-admin/linux-nas-raid.html ) but a RAID 1 will 
provide reliable, but not fast storage. Suppose that the formatted storage 
capacity is a third the total drive capacity - that makes the cost about 
$375 per TB.

I have a recent quote from Netapp, including a discount of undetermined 
size (the list prices were not available) for three FAS2040 SATA based 
systems with 12, 24 and 48 GB of disk. With RAID 4 the Netapp is a little 
more space efficient and probably half the raw capacity is usable, The 
cost per TB ranges from $7,500 down to $3,500 and includes 3 years of 
maintainance, but only NFS software - no CIFS. That is about 10 to 20 
times the price of home-brew. Of course the Netapp is much faster, but as 
"features" go, FreeBSD is quite competitive, indeed it can do many things 
the Netapp can't, such as run rsync, or can't do without paying extra, 
such as run CIFS.

The most discouraging aspect of the Netapp is what happens when the 
original 3 year service contract runs out. On our 4 year old FAS3040 with 
about 3TB of storage, the maintainance is $7,000/year and is not on-site. 
The high price is intended to prevent users from keeping old Netapp's, and 
it largely succeeds at that objective. There is rarely any point in adding 
shelves to an existing Netapp head - the maintainance on the head will 
make the shelves totally uneconomic long before they are obsolete. We have 
not decided yet what we will do, but we will not renew maintainance!

It is important not to take the view that if A is better than B, then B is 
no good, or to make decisions based on rules of thumb appropriate for 
situations far different from your own, or to assert that allowing price 
to trump quality is unprofessional. All of those are errors of thought 
that tend to result in systems that are gold-plated, but of insufficient 
capacity. It is common in university settings to offer faculty and 
students totally inadaquate storage quantity, on totally over-engineered 
storage systems. A typical research project can withstand a storage outage 
once a year (but not lost data) far more easily than it can withstand 
bumping against a tiny storage quota every day.

Daniel Feenberg