[BBLISA] Fileserver opinion

[Eugene Gorelik]

Hi Ian,

Though it may seem irrelevant, but before answering your questions I'd like
to understand
what is your web content caching strategy ?

With heavy traffic web sites like the one you described caching may
tremendously improve your performance
at all tiers and especially for storage.

There are several caching options available: browser caching(Cache-Control
and Expires HTTP headers...), memory caching (mod_memcached,mod_mem_cache),
caching proxy (nginx, varnish).

Static pages caching is relatively easy to implement in the web server layer
without code modification, for dynamic resources caching code modification
could be required.

So, I'd strongly suggest to explore all your caching options and make
decision on storage update after that.


To answer your questions:

> 1. Should we consider running a VM on this same server and host e.g. the
web server on a VM which accesses files through the virtualization
layer, rather than a physical network interconnect.

I'd recommend to keep storage separate from web servers, it will let your
web tire to scale.
It is also more secure that way.

> 2. What combination of network filesystem and local file system
combination makes sense? (currently NFS + ext4 is on the cards)
> 3. Should we consider alternatives to GigE for interconnect.

It depends on several factors, one of the important ones is how NFS server
on the storage side is implemented.
On the client side NFS+ext4 over GB interface are usually sufficient. There
are several tools available which allow to simulate
network traffic and measure performance, you can also write your own script
for basic measurements.

> 4. How can we estimate our IOPs and throughput requirements?

That's a tough one. Try collecting web server logs for several days and from
logs calculate total size of downloaded data per day.