送信者: Tetsuro Kondo <kondo@crl.go.jp>
宛先: Alan R. Whitney <awhitney@haystack.mit.edu>
Cc: B Carlson <Brent.Carlson@hia.nrc.ca>; Wayne Cannon <wayne@sgl.crestech.ca>; Dick Ferris <dferris@atnf.csiro.au>; Dave Graham <p062gra@mpifr-bonn.mpg.de>; Nori Kawaguchi <kawagu@hotaka.mtk.nao.ac.jp>; Sergei Pogrebenko <svp@jive.nl>; Misha Popov <mpopov@asc.rssi.ru>; Jon Romney <jromney@nrao.edu>; <weh@gemini.gsfc.nasa.gov>; Ralph Spencer <res@jb.man.ac.uk>; <sekido@crl.go.jp>; JUNICHI NAKAJIMA <nakaji@crl.go.jp>; <koyama@crl.go.jp>
件名 : Re: Strawman VSI-S spec
日時 : 2001年2月28日 13:43

Dear Strawman,

Comments of Kashima group for the strawman VSI-S spec have been compiled by
Nakajima-san and Sekido-san who have been engaged in the actual development
VSI hardware. I attach them below.
Best regards,

Tetsuro Kondo /Kashima Space Research Center

********

Thanks to the excellent strawman VSI-S spec. With the well organized
document
we belive further VSI-S argument will be finished faster than VSI-H.
Followings are the CRL/Kashima-Group initial discussion to the VSI-S version
1.0.

1. Targeting tape-based VLBI correlation, it is not clear how to synchronize
   the DOMs with the existing commands. We suppose that a combination of
   'rotset' 'transmit' and 'domsync' will work for this.
   We can not figured out how to establish typical synchronized play back
   with these command. On the other hand correlation for the real-time
   The command sequence is very clear. 'rotset' and 'domsync' are whole
   procedure before real-time correlation.

2. We would like to introduce time-line operation idea into the VSI-S.
   Example: 'act hhmmss <command>'
   the 'act' command works with several keywords and series of acts are
   programmable and cancelations are possible too. For example in the
   tape-based operation,
   'act hhmmss receive' will start the synchronized recording.
   'act hhmmss transmit' will start the synchronized playback.
   This seems complex, but heterogeneous environment and large scale system,
   time critical control to many DIMs and DOMs are solved with the
   time line operation.


3. PDATA driven DIM(DOM) control should be included.
   Currently, the PDATA is only used for parameter forwarding. In some case,
   it is expected that an upstream DAS has an ability to control the
successive
   DIMs. Thus, several DIM(DOM) setup and operational commands should be
activated
   via the PDATA input. This will increase the VSI standard ability.


4. How far we step into DTS specific and how to treat them through our
discussion?
   Several commands defined by {} are highly DTS specific and may be less
important
   to discuss with other group. But some of them, we will find out common
usage
   or useful combination. In this VSI-S, How we will discuss common part
accompanying
   the each DTS specific. Of course we are inetersted in other DTS specific
functions.
   But we should determine how far we will describe the DTS specific.

5. 'friendly' Keywords
   'Keywords' are difficult to imagine their function especially to newcomer
in
   their first glance. We will propose some alternatives.


Japaneses group proposal contains the detailed items will be submitted
separately.
We will collate the VSI-S with our idea soon. The situation is different
from when
we discussed the VSI-H and we agreed to the version 1.0 bottom line.
The proposal will be a moderate one with supplemantal Keywords.

****
 