<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
+1 definitely. Thanks Fernando.<BR>
It's impossible to make it up to Berkeley during the workday otherwise I'd come.<BR>
<BR>
Stephen<BR> <BR>> Date: Sat, 14 Nov 2009 02:18:32 -0800<BR>> From: fperez.net@gmail.com<BR>> To: baypiggies@python.org<BR>> Subject: [Baypiggies] [ANN] Scientific Python at UC Berkeley talk for November 18<BR>> <BR>> Hi all,<BR>> <BR>> I would like to know if in general people want to get these<BR>> announcements regarding the UC Berkeley py4science seminars on this<BR>> list or not, I don't want to spam people. But last week's event with<BR>> Guido drew a huge crowd, some from this list, so perhaps there is<BR>> interest (obviously Guido's visit was a special event, but still<BR>> there may be interest in these talks amongst this group). If there<BR>> isn't let me know and I won't post them again, as you can always see<BR>> them at the group page:<BR>> <BR>> https://cirl.berkeley.edu/view/Py4Science/WebHome<BR>> <BR>> But if you find these useful, I'm happy to send the announcements<BR>> here, it will be limited to just one email every 2 weeks at most.<BR>> <BR>> So for now here's next week's, I'll refrain from posting these again<BR>> if so requested.<BR>> <BR>> Next week, we will have a presentation by Bryan Catanzaro from EECS<BR>> on Copperhead, the system we got a glimpse of last week from his<BR>> lightning talk. Details below.<BR>> <BR>> <BR>> Thanks,<BR>> <BR>> f<BR>> <BR>> When: November 18, 2pm.<BR>> <BR>> Where: UC Berkeley, Evans 508-20<BR>> <BR>> Title: Copperhead: Data-Parallel Python<BR>> <BR>> Abstract:<BR>> The need for productive programming languages which can avail<BR>> themselves of parallel hardware has never been more acute. The<BR>> Copperhead project attempts to address this problem by defining a<BR>> subset of Python which can be compiled and executed in a data-parallel<BR>> fashion. Copperhead procedures are expressed as standard, fully-legal<BR>> Python procedures operating on Numpy datatypes, which are intercepted,<BR>> specialized, and compiled to parallel C code at runtime, and then<BR>> executed on a high-performance parallel platform. Since the<BR>> Copperhead runtime supports only a subset of Python, the runtime will<BR>> revert to standard Python execution if specialization fails. The<BR>> current Copperhead runtime targets Nvidia Graphics Processors, which<BR>> are highly suited for data-parallel computation and provide high<BR>> performance. In this talk, I will be discussing the current state of<BR>> Copperhead, as well as plans for future development.<BR>> _______________________________________________<BR>> Baypiggies mailing list<BR>> Baypiggies@python.org<BR>> To change your subscription options or unsubscribe:<BR>> http://mail.python.org/mailman/listinfo/baypiggies<BR> <br /><hr />Windows 7: It works the way you want. <a href='http://www.microsoft.com/Windows/windows-7/default.aspx?ocid=PID24727::T:WLMTAGL:ON:WL:en-US:WWL_WIN_evergreen:112009v2' target='_new'>Learn more.</a></body>
</html>