The jini meeting is being webcast and recorded. I've not seen it yet but the program has a number of interesting sessions.
"I have a mind like a steel... uh... thingy." Patrick Logan's weblog.
Friday, September 15, 2006
Their benchmarks also show that many applications are not able to exploit 4 cores very well, never mind 8. Now, where did I leave that Erlang disk image....We're busy cramming everything into a common language runtime. Shared memory, critical sections, locks, common object models... it's the wave of the future?
Well, it's a tsunami that is catching up with us while we're busy cramming everything into a common language runtime.
Update: Steve Dekorte writes that no language will save these high core count MISD systems from the Von Nuemann bottleneck.
But the advantage of a shared-nothing programming model like Erlang's is that systems can more easily accomodate multiple shared-memory cores as well as multiple shared-nothing nodes. The languages that assume there is no shared memory are better positioned to get around the Von Nuemann bottleneck.
Common language runtimes that assume shared-memory are a diversion at this point. But I think it will be another five years or so before the JVM and CLR leaders start to think about this. Ideas just take a long time to catch on. With JRuby and such now, we're seeing the beginning of the fruition of an idea that has been around as long as the JVM. The JVM itself was the fruition of UCSD Pascal byte code from the early 1980's.
From what I have seen the hardware manufacturers have a difficult time getting their heads around this direction too. They understand that a single, faster CPU is not going to sell. And so they understand the reason for putting multiple CPUs on a core. I am not sure they know what to do with them in software... Maybe web services?
Tuesday, September 12, 2006
Avi Bryant makes the connection between JRuby, Ruby performance generally, and how fast dynamic languages can really be, e.g. the Strongtalk implementation.
As it happens, the Strongtalk VM would make an excellent basis for the next generation Ruby VM. Given that Sun still employs some of the original Strongtalk engineers, and has just hired the JRuby guys, maybe they can make that happen?It gets even better... Strongtalk was the precursor to the Java Hotspot runtime. Java is essentially already running a pretty good dynamic runtime under the hood. The language-specificness of the JVM just has to be peeled back a notch and Ruby (and Smalltalk, and Lisp) would be good to go. Or at least much better off than they are now.
Maybe a bit more than a notch. Please start peeling.
Then someone over at James Robertson's blog asks about the performance of a good dynamic dispatch mechanism relative to C++.
The line was crossed some time ago when the nature of most C++ applications on the hardware at the time was better suited to dynamic dispatch. When you talk about "C++" once again you have to talk about which implementation. Some C++ implementations have been able to do a more dynamic dispatch (i.e. again implement a significant dynamic language mechanism) in order to get better performance than the more traditional vector-based fixed offset indexing.In general, Smalltalk can do message dispatch faster than C++.Is that one of those in theory things?
Dynamic *will* continue to win more and more for most applications.
Sunday, September 10, 2006
James Robertson on something I've found true as well...
I rent [cars] on travel a fair bit. A decade ago, it took me less time to get out of the rental lot. Now, I always make sure to spend a few minutes locating things like the seat adjustment, the radio controls, the wipers, defroster... you get the picture.I remember the first time I found myself driving with a "sports" transmission without purposely shifting into it, I was on the road and the damned thing wouldn't get out of first. So off to the side of the road where I spent a few minutes figuring out how the supposedly "automatic" transmission worked.
- ► 2011 (19)
- ► 2009 (40)
- ► 2008 (402)
- ► 2007 (388)
- ▼ 2006 (261)
- ► 2005 (335)
- ► 2004 (534)
- ► 2003 (286)
- Patrick Logan
- Portland, Oregon, United States
- I'm usually writing from my favorite location on the planet, the pacific northwest of the u.s. I write for myself only and unless otherwise specified my posts here should not be taken as representing an official position of my employer. Contact me at my gee mail account, username patrickdlogan.