Much has been made lately about the trend toward solid-state drives. Now a new Intel technology, code-named Braidwood, may delay that trend, blending the performance of solid-state drives with the economy of old-style hard drives.
Braidwood--like its predecessor, Intel's Turbo Memory technology (formerly code-named Robson)--is basically a solid-state cache for all the disks in the system.
I heard about Braidwood earlier this summer on CNET (see "Intel 'Braidwood' chip targets snappier software" by Brooke Crothers). But I shrugged it off, assuming it would be no better than Turbo Memory, which left a bad taste in the mouth of many PC makers, end users, and Microsoft execs. Turbo Memory (and Turbo Memory 2.0) wasn't cheap, and it definitely wasn't worth the cost. The PC industry operates on such slim margins that every dollar's worth of hardware has to earn its keep--and Robson didn't.
The 62-page report is titled "Intel's Braidwood: Death to SSDs?"
Handy's report argues persuasively that Braidwood might actually be worthwhile, and that got my attention. I've known him a long time, and he's a very good analyst--he's been covering memory and caching technology a lot longer than I have. He wrote one of the standard references for computer system architects, "The Cache Memory Book."
So I sent Handy a note, and he sent me a copy of the report. And now that I've read it, I'm inclined to agree with his conclusions, assuming the information he's obtained about Braidwood is accurate. It does seem reasonable, at least.
The first thing to understand is why flash memory can be a good disk cache. This boils down to its much faster access times: microseconds, not milliseconds. Flash can actually take much longer to write than a hard disk. But for reads, it's really quick. So if you can be smart about putting the right hard-disk data in the cache, especially by choosing the right time to do those write operations, you can save huge amounts of time on future disk reads.
As Handy explains in his report, Intel was relying on Microsoft to provide the smarts for Turbo Memory, but Vista's support for it was weak to begin with. And when the early version of Turbo Memory fell short in performance, Microsoft saw no reason to invest in more software development.
I think Turbo Memory also failed because Intel didn't provide enough flash memory to help much: just 512MB or 1GB in Turbo Memory 1.0, and up to 4GB in the 2.0 generation. Microsoft's guidelines for ReadyBoost--the generic version of this feature, which worked from ordinary thumb drives--recommended providing one to three times as much flash as the RAM in the system, and 3x worked better than 1x. With most performance-sensitive users already equipping their systems with gigabytes of RAM, Turbo Memory needed a lot more flash than Intel was offering.
Handy's report said Braidwood's new drivers will help solve the software problems (though he describes some potential further improvements), and I assume Moore's Law and the crashing prices of flash memory will solve the capacity problem.
As I noted back in 2007 ("Flash drives--now a pricey but reasonable option"), a moderate amount of flash--say, at least 16GB, a retail value of $15 to $20 in the Braidwood time frame (later this year and into 2010)--can hold every file needed to run the operating system and the popular applications. That's something Turbo Memory could never do.
But if Braidwood is smart enough, it could be more effective to treat that flash as a cache rather than a separate drive, as I suggested back then. A caching solution would reduce boot time and application launch times substantially, and be easier to manage. If there's room enough to cache all the disk directory structures, paging files, and commonly accessed user documents as well, so much the better. We'll have to see just how smart Intel's software really is, but with enough room, these benefits should be realizable. As they used to say about Soviet tanks, quantity has a quality all its own.
Handy's report summarizes all the arguments I've ever heard in favor of flash cache, plus some I hadn't heard before. It also goes on to explain Braidwood's likely effect on system cost (though I think his price estimates are high), to predict how rapidly Braidwood will be adopted, and to show how Braidwood could actually slow the growth rate for DRAM in PCs, since it reduces the need for excess DRAM to act as a disk cache.
The bottom line is that the combination of flash and standard hard disks deliver almost all the advantages of pure solid-state disks, but at a lower overall price.
I think Handy's report is good work. If you're in the mass-storage business, you should take a look.