I recently upgraded to a 320GB hard drive for my main OS partition, as I was starting to run very tight on room with the old 160. I've got a lot of stuff on there; there's really no way around it. To get everything over to the new disk I used BootIt Next Generation, a program I've used for similar hard drive upgrades that is at least as good as anything available for twice the price. And when I booted to the new drive, I saw a marked performance increase -- I went from 45 seconds in the boot logo to about 10 seconds, tops. The performance in the rest of the system, though, seems to be about the same.
How to explain this? I think the copy process helped defragment some of the file-system structures that otherwise hadn't been compacted, which might speed boot time. But the rest of the system still seems to be at about the same plateau of speed. Not long ago while doing research for a series of articles on defragmentation (which I'll link here when they're all up), I was told that while hard drives have on the whole gotten faster, the disparity of speed between the hard drive and the rest of the computer has increased. I think I'm experiencing some examples of that firsthand.
But there's no question that having that much more free space -- and a drive which is incrementally faster than the last one -- has paid off. In the same articles, I argued that having more free space on any drive always helps cut down (although not eliminate) the detrimental effects of fragmentation; if you have more than 75% space full on a given drive, you either need to clear it off or upgrade, or you're going to be fighting a performance drain that you can't win by simply defragging. And the vast majority of the time when you buy a bigger drive, you get one that runs quieter, has a bigger onboard cache, and may even support slightly faster throughput to begin with.
As an aside, I've also started running the newest version of a progressive defragmenter called Buzzsaw, which runs silently in the background and defragments individual files when hard drive activity drops be low a certain level. It seems to work well with servers, too, although at a bit of a CPU cost -- it seems to need a lot of processing power when the number of fragments is quite high. But I've recommended it before, and I'm doing so again now.
Finally, I'm laying tentative plans to move the Newsletter off Blogger competely and onto my own personal pages. When this happens I'll try to also migrate the old posts with it, and there may also be a name change to go with it. But the old blog will redirect to the new one for quite some time to come.
