[evla-sw-discuss] Schedule for MIB development

[Bill Sahr]

A .html version of an MS Project worksheet/schedule for MIB 
software and hardware development, along with some dates for 
the test antenna is included with this email. Additionally, 
for those of you whose email programs do not handle html 
files, I have sent a copy of the file to Terry Romero with 
a request that it be added to the Computing Working Documents 
web page.  It should be available on that web page sometime 
this coming Monday (9/16) or Tuesday (9/17). The URL for that 
page is:

http://www.aoc.nrao.edu/evla/techdocs/computer/workdocs/index.shtml

The dates for MIB hardware & lower level software development
were supplied by George Peck (as of 8/26/2002).  The dates
for the test antenna come from a conversation I had in late
August, 2002 with Gene Cole.  I encourage everyone connected 
with AMCS and MIB design to examine this schedule.  

I'd like to highlight a few points in the schedule.  To begin
near the end, a designated test antenna is scheduled to go
into the antenna barn on 4/1/2003.  Approximately 10 weeks
will be spent in the barn and on the master pad outfitting the
antenna with new electronics, including a new monitor and 
control system.  Fiber optics for the EVLA M&C network will
be available on the master pad.  There are no plans to extend
the EVLA M&C network to the antenna barn.  If this schedule is
met, outfitting of the antenna will be completed on/about
6/10/2003.  Approximately 8 to 9 weeks have been set aside for
tests of the antenna on the master pad - 6/11 through 8/11/2003.
At that point, the plan is to move the test antenna into the
array.  The period 8/19/2003 - 11/19/2003 has been set aside
for array tests of the antenna, approximately 13 weeks.

What will be needed w.r.t. the EVLA M&C effort to support this
schedule ?  An EVLA M&C network which which includes the master
pad and the control building must be in place, tested and 
functioning by the time the test antenna is placed on the 
master pad, i.e., sometime around 4/1/2003.  Software capable
of producing, displaying, and archiving monitor data from the
test antenna will be needed in the same time frame, as will
software capable of controlling the test antenna.  Computing 
platforms to run this software must have been specified,
purchased, delivered, installed, and configured by the same
date - 4/1/2003.  Actually, the computing platforms are needed
earlier as it would be wise to have a network and computing 
platforms which include, at a minimum, the master pad and 
control building in place and operating at least one month 
before the test antenna is due to be set down on the master pad.
(My schedule does NOT reflect this one month lead time.)
At least a primitive form of hybrid array software will be
needed for the array tests of the test antenna, i.e., by
early August, 2003.

Moving backward, another very significant milestone is the
availability of a MIB prototype board.  The MIB prototype
board is scheduled to be available, but perhaps not fully
tested and developed, on 11/14/2002, with MIB prototype
design and development finalized by 12/3/2002.  The MIB
module/device software (developed by hardware engineers
with assistance as needed from the Computer Division) is
scheduled for the period 12/06/2002 - 04/03/2002.  The 
end date of that period coinciding with the start of
outfitting of the test antenna.  Development of the 
MIB communication protocol (in the CMP processor and on
the TC11IB TriBoard) is scheduled for the period 9/16/2002
- 11/29/2002, hopefully bringing it to completion at 
approximately the same time as the MIB prototype board
becomes available.  (The term "MIB communication protocol
is meant to cover the protocol(s) for getting commands
to MIB connected devices and data from MIB connected
devices.)

There is also a need to have a substantially completed
overall M&C software architecture and design available by
the time the EVLA Advisory Committee next meets - probably
sometime in mid-December, 2002.

I must add that so far we are not meeting our deadlines 
for software.  It was hoped that the software development
toolset would be licensed by Fri, 9/13/2002.  Well, we
did have one licensed toolset available before that date,
but as of 9/13/2002, the toolset was not fully functional 
due to software/hardware configuration issues.  So, while
we can claim to have met the date in a technical sense,
we have not met it in the functional sense, which is the
sense of the claim that really counts.  Also, we cannot 
claim to have met the date of 9/13/2002 for specification 
of the MIB communication protocol.  It is still under 
development.  Additionally, Accelerated Technology is 
currently stalled in it's efforts to develop the ethernet 
driver for the TC11IB chip, so we are unlikely to meet the
9/18/2002 date for systems software ported to the TriBoard.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/evla-sw-discuss/attachments/20020913/03be3081/attachment.html>