The following use cases are intended to illustrate some of the complexities of the storage of instrument metadata in a device called a ÔpuckÕ which travels with the instrument. This document does not capture all of the possible use cases. The MOOS system may decide to support some, all or none of these use cases.
A scientist has an instrument
that is deployed sub-surface on a mooring far offshore. The instrument relies
on power from a solar-charged battery pack on the surface buoy. The scientist is able to communicate
with the instrument via radio modem when they are near the buoy. During a
semi-annual research cruise to the region the scientist discovers that due to
the biological conditions the instrument needs to be changed from the default
settings to collect scientifically meaningful data. She has also planned to have a diver attach an exciting new
sensor she has been developing to a spare analog channel on the instrument.
After the changes are made, the scientist is able to verify that the instrument
and sensor are collecting great data and the ship departs.
Scenario 1a. Non-writeable puck, retrieve and
replace puck in the field.
Divers remove the instrument
from the mooring line (or possibly must recover the entire mooring) and return
it to the ship. There, they
install the new sensor. Since they
planned ahead and brought a Òpuck-burnerÓ, they are able to create a new puck
that has the new instrument settings and also reflects the new sensorÕs output.
Divers then replace the instrument on the mooring string (or re-deploy the
mooring).
Scenario 1b. Field-writeable puck
Divers attach new sensor to
the instrument in-place. The scientist transmits the necessary commands to
adjust instrument settings for the environmental conditions and to output data
from the new sensor as well. The
new configuration parameters are stored to the instrumentÕs puck.
Scenario 1c. Non-writeable puck. Instrument command persistence is
available elsewhere on the platform.
There is no capability
on-board the ship to create a new puck. The scientist decides to make the
necessary changes to the instrument without replacing the puck. Divers attach
new sensor to the instrument in-place. The scientist transmits the necessary
commands to adjust instrument settings for the environmental conditions and to
output data from the new sensor as well. The new configuration parameters are
stored somewhere on the platform other than the instrumentÕs puck.
Scenario 1d Non-writeable puck. No instrument command persistence is
available on the platform
There is no capability
on-board the ship to create a new puck. The scientist decides to gamble and
make the necessary changes to the instrument without replacing the puck. Divers
attach new sensor to the instrument in-place. The scientist transmits the
necessary commands to adjust instrument settings for the environmental
conditions and to output data from the new sensor as well. The new
configuration parameters are running on the system but the will be lost if the
instrument ever resets.
A few weeks later, after an unusual week of fog, the solar-charged
batteries, dip below level required to power the platform so it shuts down
temporarily. Finally, after the
sun returns, the batteries recover and the system and the instrument power back
up.
Scenario 2a. Non-writeable
puck, retrieve and replace puck in the field.
Since a new and correct puck
was created on-board ship and re-installed with the instrument, the
instrumentÕs device driver restores the correct settings.
Scenario 2b. Field-writeable puck
Since changes were written to
the puck automatically in the field, the instrument device driver restores the
correct settings.
Scenario 2c. Non-writeable puck. Command persistence is available
elsewhere on the platformÉ
Since changes were written to
the platformÕs persistent storage in the field, the ISI system reissues
commands to the instrument to restore the correct settings.
Scenario 2d Non-writeable puck. No command persistence is available on
the platform
Since there was no way to
store the non-default settings for the instrument they are lost. The instrument powers-up using itÕs
default settings. The system
collects five months of degraded or useless data and nothing is collected from
the added sensor.
A scientist at WHOI has an
instrument that is deployed on a benthic platform anchored offshore. The
instrument relies on power from a battery pack which is replaced periodically
by an observatory maintenance team.
The scientist is able to communicate with the instrument via the internet.
Previously collected data indicate that the instrumentÕs sensitivity needs to
be increased from the default setting to produce meaningful data. From her desk, the scientist connects
to her instrument through the ISI portal and sends the necessary commands. She then verifies that the instrument
is working properly and logs off. Weeks later she departs for a two-month
research cruise to Barbados. While
she is away, the MBARI operations team goes out the platform and replaces the
battery packs using an ROV. This brief power interruption causes all platform
systems and instruments to reset.
The scientist is notified of the reset event but she will not see the
message until she returns from the cruise.
Scenario 3a. Non-writeable puck, retrieve and
replace puck in the field.
To make the new settings
persist she would have to organize a cruise to go recover the instrument,
replace the puck, and redeploy the instrument on the benthic platform.
Scenario 3b. Field-writeable puck
Parameters are stored on the
puck as part of the remote update process. The instrument device driver
restores the correct settings for the instrument.
Scenario 3c. Non-writeable puck. Instrument command persistence is
available elsewhere on the platform.
Parameters are stored on the
platform as part of the remote update process. Since changes were written to
the platformÕs persistent storage in the field. The ISI system re-issues
commands to the instrument to restore the correct settings
Scenario 3d Non-writeable puck. No instrument command persistence is
available on the platform
Since there was no way to
store the non-default settings for the instrument they are lost. The instrument powers-up using itÕs
default settings. The system
collects a month or more of degraded or useless data.
An experiment is designed to
automatically monitor a signal looking for a predefined event to occur. Once the event is detected, a number
instruments are commanded to begin an intense sampling program. Shortly after detecting the event and
commanding the response something causes a system reset.
Scenario 4a. Non-writeable puck, retrieve and replace puck in the field.
Not applicable.
Scenario 4b. Field-writeable puck
Parameters are stored on the
puck as part of the remote update process. The instrument device driver
restores the correct settings for the intensive sampling program.
Scenario 4c. Non-writeable puck. Instrument command persistence is
available elsewhere on the platform.
Parameters are stored on the
platform as part of the automatic event response. Since changes were written to
the platformÕs persistent storage in the field. The ISI system re-issues
commands to all instruments to restore the correct settings for the intensive
sampling program.
Scenario 4d Non-writeable puck. No instrument command persistence is
available on the platform
Since there was no way to
store the non-default settings for the instruments they are lost. The all instruments powers-up using
their default settings. The system
collects data as if the event never happened.
Changes to the instrument
parameters are stored on the platform (not the puck) as part of the remote
update process as described earlier.
Divers visit the platform to perform routine maintenance of the
instrument in which they remove it, make some changes (calibrations, new settings
etc.). They ÒburnÓ a new puck and replace the instrument with itÕs puck on the
platform.
In scenarios 1c and 3c, the
ISI system re-issues the old commands to the instrument to restore what it
ÔthinksÕ are the correct settings.
In scenarios 1a,b and 3a,b
the ISI system relies on the information in the puck to yield correct operation
of the instrument.