Now that I have the time, I’d like to
respond extend Terry Walsh’s article reporting on rumors of WHS v2.
The potential of a Windows Media Server (possibly a WHS v2 box?) is an exciting proposition for many people – my concern would be a [boosting] of WHS’ minimum hardware requirements. One of the main reasons Windows Media Center was such a problem to self build was TV tuner drivers – I’m no longer a big Media Center user (I’m a big fan of the UI and TV Guide, but it’s a cumbersome solution compared to Sky HD), and I’m sure tuner drivers and the built in MPEG decoder in Vista Media Center are helping drive a better experience, but I love WHS’ simplicity – both hardware and software – and I’d hate to see it be compromised in the future through more complex hardware.
The fact is that Home Server and Media Center are separate entities. The WHS team was right to do it this way since they both perform entirely different function. WHS is a backup and file server while MCE is a TV and recorded media server. I can see the sense of having both and syncing files back and forth at 4am or something.
Terry makes a valid point about hardware. Could this server backup and stream media at the same time? It’ll probably be HD stuff by the time it comes out so we’re talking some pretty large data files here. Sure, the network will probably sweat a little, but that’s not the problem.
If you’re going to actually write this thing, its going to have to be a pretty large undertaking. Its going have to be able run on two or more cores efficiently. So in the four core environment that’s going to be pretty common two years or so from now ( given Redmond’s release cycle) , its going to have to distribute the load across all four cores. Different hard disks will have to be used by each core ( i.e one for backups and one for media to avoid I/O conflicts and such). That’s one hell of a lot of programming logic right there,and that’s before we get to using network bandwidth efficiently.
It’ll be using the Vista networking stack, so we can expect a modest improvement (bearing in mind that I actually haven’t seen this with Vista yet). One possible answer to this is Sun’s Project Neptune:
That’s why we just introduced Project Neptune – a silicon project that marries the parallelism of the microprocessor (for Intel, AMD and SPARC systems), with the parallelism of the underlying operating system (Solaris, Linux or Windows), with parallelism in the network itself. Which in concert with some software magic (which goes by the name of the Crossbow project) allows enterprises to collapse cabling, ports, cards and spending – by bringing parallelism to basic network infrastructure (for geeks, you can take multiple TCP streams and allocate them to different processor threads, spreading out load and freeing up CPU’s/ports). Ports become a physical convenience, just like a server – what’s happening inside depends upon rules or policies set by the user/administrator to automate such decisions. Like I said, the network is the computer, and the computer’s virtualized, so why not the network?
Now I’m assuming that we’ll see a less-than-enterprise offering of that could be part of the hardware.
Then there’s the task of combining two codebases, the logic of how we combine WHS’s interface and WMS’s, how WHS’ duplication is going to work with WMS, how back-ups are gong to work and so on for a thousand and more items.