On 8/21/07, Andrew Lentvorski <[EMAIL PROTECTED]> wrote: > Bob La Quey wrote: > > Again the innovative parts of NetKernel have nothing to do with > > TCP/IP or NAT. They are operating well below that level of the > > system. > > That's not what NetKernel claims: > http://www.1060research.com/netkernel/features/index.html > > The diagram shows NetKernel sitting *above* the Java VM and OS, not > below. Quoting:
Agreed. Though the next generation goes lower. The "target" is primarily down not up. The really important innovation is that it manages the thread pool in an automagical way by using load balancing. All of this is done for any resource oriented applications, including those that never leave the chip. The use of the Java VM and OS at this point in the development is just a convenience and _not_ an essential point. > "NetKernel is built on a precision-crafted micro-kernel. It resolves > logical URIs to physical service implementations. A Unix-like process > table holds pending requests for scheduling onto threads held in a > tightly managed thread pool. The micro-kernel enables both synchronous > and asynchronous processing and its sophisticated algorithms optimizes > CPU throughput while minimizing thread contention and context switching > costs. Developing applications at a logical level means it is impossible > to write non-threadsafe systems." That says nothing about level. "resolves logical URIs to physical level services" is the key. This says nothing about tcp/ip or NAT. In fact they have URI resolutions that work at the level of the VM (and once the implementations are finished the physical hardware of specific cpu architectures.) Yes it is true that they have URI resolutions that are conventional and run on the tcp/ip but that is _not_ innovative. The point, which I will repeat is to build a resource oriented load balancing OS. This model has been shown to work well with the distributed client server architecture of the web. I suspect the analog will work well with mulit-cpu chips. Will the transport be different? Of course. Will the mechanisms of resource resolution be something other than DNS? Of course. But the resource oriented model looks good to me. > Now maybe their marketing just sucks, but, given the slickness of their > website, that is not an immediate assumption I would make. Especially > since they have their own license. > > I stand by my NAT hole punching test. If you can't use the framework to > do that, it doesn't have enough flexibility. Other than a narrow band interpretation of the term URI I cannot see why you are being so dogmatic, Again it appears you miss or do not understand the main point. If you agree fine. If not fine. I know of no way to make my point any clearer except to say, "Spend some time working through the excellent material at their websites." BTW it is _not_ really marketing. Almost all of it has been produced by one of the two main developers of NetKernel, Peter Rogers, has personally produced almost all of the material on the two web sites. Peter has a Ph.D. in physics from Cambridge and is very funny guy. You will find a lot of allusion to Monty Python in the tutorials. His only lament seems to be that it is hard to get many people to think very abstractly about these issues. This seems to be a problem with getting people to wade through Roy Fielding's thesis as well. So it goes, BobLQ -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
