8.15: news.adda/ROS (robot`OS)/summary:
. recently it was conjectured that the best chance for a
prog'lang that features concurrency better than threads
is going to be the use of a configuration co.language
which coordinates the modules of a system
while the user's prefered language defines the modules .
. now it's been reported by Kurzweil
that several university labs have settled on a configuration lang
for the problem of needing a standard platform for robotics development .
. there is currently an impl' for python,
giving that a healthy concurrency model,
and making it mixable with other languages .
12.16: todo.adda/e`promises/lots of interesting unfamiliars here:
In E a future is represented by a pair of valuescalled a "promise/resolver pair".The promise represents the read-only view,and the resolver is required in order to set the future's value.Support for read-only views is consistent with the Principle of Least Authority,since it enables the ability to set the valueto be restricted to subjects that need to set it.In a system that also supports pipelining,the sender of an asynchronous message (with result)receives the read-only promise for the result,and the target of the message receives the resolver.Promise pipelining should be distinguished fromparallel asynchronous message passing.In a system supporting parallel message passingbut not pipelining,the message sends x <- a() and y <- b() in the above examplecould proceed in parallel,but the send of t1 <- c(t2)would have to wait until both t1 and t2 had been received,even when x, y, t1, and t2 are on the same remote machine.The relative latency advantage of pipeliningbecomes even greater in more complicated situationsinvolving many messages.Promise pipelining also should not be confused withpipelined message processing in Actor systems,where it is possible for an actor to specify and begin executinga behaviour for the next messagebefore having completed processing of the current message.. a promise pipelining system has first-class promise valueswhen an argument to, and the value returned by a call or sendcan directly be a promise .
12.19: news.adda/concurrency:
humans are actually quite adept at concurrency
. web for adda/0908's source of
"(humans are actually quite adept at reasoning about concurrent systems)
[PDF] Are new languages necessary for multicore?
ptolemy.eecs.berkeley.edu/.../CGO_Panel_PositionStatement_Lee.pdf
[PDF] Cyber-Physical Systems - Are Computing Foundations Adequate?
mesl.ucsd.edu/gupta/cse237b/Readings/Lee_CPS_PositionPaper.pdf -
Symposium on Code Generation and Optimization (CGO) ` CGO-2007 Panel
www.cgo.org/cgo2007/html/panel.html
No comments:
Post a Comment