developing a common telemetry archiving architecture

Leeha Herrera

Nov 2005L Herrera

Developing a Common Telemetry Archiving Architecture

Presented at: Space Telescope Scientific Institute

Nov 2005L Herrera

Introduction

It is possible and desirable to support multiple missions with a common code infrastructure.For example, a common telemetry archiving system for creating a

long-term repository telemetry in packet format to support science and mission operations was developed.

With a common approach, development costs, staff, and testing can be shared across multiple missions.The initial cost is significant but is then drastically reduced for

subsequent missions.

Nov 2005L Herrera

Overview

• Background on JHU APL ground systems

• The Common Ground Approach• Assessment Product Line• Benefits, Issues, Results

Nov 2005L Herrera

JHU/APL Ground System Software Background

Nov 2005L Herrera

Single Mission Paradigm

• A Dedicated Teams was formed for each mission.+ No conflicts in resource scheduling across programs+ Responsive to the specific needs of the mission- Knowledge transfer & sharing between teams not inherently facilitated- Potential for redundant tasking

• A Snapshot of a previous system was used as basis for next mission No requirement to design for reuse+ A new mission did not have to start from scratch+ Users were familiar with overall functionality of applications- Not a simple or straight forward task- Modifications for some areas could be comparable to complete rework- Occasional need for complete rework or creation of new applications- Fixes or enhancements could easily shared between active missions

• Approximately the same efforts was required to provide the same fundamental functionality for each mission!

Nov 2005L Herrera

Single Mission ParadigmGround System

(in development)

Ground System

(production version)

Nov 2005L Herrera

Need for Change

JHU/APL was awarded with three NASA missions with overlapping schedules.

A Snapshot approach for re-use works for missions with dovetailed schedules, where teams can easily transition between projects.

We needed a solution to address limited resources for three active mission.

Would like to: Reduce overall development costs Reduce required staff per mission Improve quality

Improve estimations in scheduling and cost

Nov 2005L Herrera

The Common Ground Approach

Nov 2005L Herrera

Overview

The primary objective was to support the simultaneous development of multiple NASA missions.• Effort to develop applications that can be used on all

missions with little to no customization

Staff was reorganized into teams centered around “Product Lines.” • Reduce redundancy by dividing into common functional

areas vs. traditional mission-based teams• 4 Product Lines Created:

Assessment, Commanding, Planning, Telemetry

Nov 2005L Herrera

Simplified MOC Overview

Mission Operations Center

Planning(Commanding

Pre-Processing)Commanding

Assessment(Telemetry

Post-Processing)Telemetry

Deep SpaceNetwork

CLTU’s

Telemetry Frame

SFDU’s

Telemetry TransferFrames

Scripts

SupplementedTelemetry

Packets

SpacecraftUsers

CLCW’sMOC statusTelemetry

PlanningInputs

PlanningProducts

PlanningInputs

DataProducts

POC Commanding

POC or SSC Data Processing

WrappedTelecommandPackets

TelemetryPackets

STEREO Ground Software Peer Review: Ground Software Presentation – P. Mckerracher 10/4/2002

Nov 2005L Herrera

Initial Steps and Planning

The legacy code was evaluated for reuse.• Identify adaptable areas of code

A design was established for a new code infrastructure to handle variations in missions but support common core development.• Establish approach to encapsulate mission specific details• Definition of base classes to support new software pattern:

Parallel Classes• Used across all functional areas for all missions

Risk areas were identified.• Example: Telemetry Archiving Architecture

Nov 2005L Herrera

Cartoon Overview

+ =Generic

MissionY

MissionX

MissionZ

Mission…

Nov 2005L Herrera

Code Organization

Parallel Class PatternEx: Other classes use the TelemetryPacket class without knowing which

implementation it is.

Classes abide by the interface defined in the generic version.

The Configuration Management Tool associates the correct version of classes for a particular build.

CCSDSTelemetryPacket

(generic)

CCSDSTelemetryPacket(MESSENGER)

CCSDSTelemetryPacket

(STEREO)

Common GroundSW

Nov 2005L Herrera

Assessment Product Line

Nov 2005L Herrera

Assessment Charter

Encompasses all activities and applications used to archive, play back, and interpret telemetry for assessing spacecraft health and state, identifying trends, and providing telemetry data to external processes and customers for further analysis and processing.

Primary users of the data products include Subsystem Developers, Guidance and Control (G&C) analysts, the Integration & Test team, and the Mission Operations team. The primary users of the data access libraries and higher level data access tools are MOC and testbed software developers and G&C analysts.

Responsible for providing historical data from telemetry (e.g., command histories, voltages, temperatures) in forms appropriate for the tasks being performed by its customers (e.g., trend analysis for MOPS). It is additionally responsible for providing access tools in the form of software libraries for use by developers and stand-alone utilities for data consumers.

Nov 2005L Herrera

AssessmentCSCI

GSE Clients1:N

Archived Telemetry

Ancillary ProductsArchive Files

Files Affected FileIDR File

Archived TelemetryPackets

TelemetryCSCI

Archived Telemetry

Telemetry Packets,archive_server Directives

PlanningCSCI

Archived Telemetry

Assessment CSCI Context Diagram

SOC

archive_serverDirectives


DSMS

Cmd RadiationTimes Files

FrontEndCSCI

Cmd RadiationTimes Files

MOPs/ I&TUsers1:N

archive_serverDirectives,

Requested PointsFile

Archived Telemetry, Decommutated Points

Plots


LHerrera 06/03

Context Diagram

Nov 2005L Herrera

Goals for the Archiving System

Faster and consistent response and processing

A robust and flexible system• Key for development for multiple missions• Allow mission specific modifications• Lower maintenance costs and support time

Allow mission specific requirements and configurations

Support wide variety of users: Developers, I&T, MOps, Science Centers

Interface with EPOCH T&C

Nov 2005L Herrera

Key Design Decisions

Provide 3 types of access to dataReal-time, Instant Playback, Long-term Playback

Ability to “plug in” different processes to convert data from multiple sources and formats

Long running, file based processing vs. socketsApplications monitor input directories for new files

Archive built from real-time and off-line file processing

Access to available archive data while new data is added

Nov 2005L Herrera

System Data Flow

As ReceivedData Files

SortedArchive Files

Ground StationData Files

Front EndData Files

SorterdIndex & Data

Files

SorterdIndex & Data

Files

Real-timeData Client

Instant ReplayData Client

Playback Data Client

Data RouterReal-timeInterface

Spool datato files

InstantReplay

Interface

Sort &Create Index& Data Files

(filter)

DataConverter

DataConverter

Sort &Create Index& Data Files

(filter)

Merge newdata w/currentarchive

AssessmentProcesses

AssessmentProcesses

PlaybackInterface

for archiveddata

Ground Station

EPOCH

Nov 2005L Herrera

Benefits, Issues, Results

Nov 2005L Herrera

Component Based Architecture

Allows for flexibility in configurationEach mission can customize the system

Only necessary “plug in” components used

Small components focused on a single function have proven to be easier to develop, test, and maintain

Future components can be added to create additional layers of functionality

Nov 2005L Herrera

Archiving Architecture

Seamless merging of data from multiple sources with configurable filtering capabilities

Allows multiple points of access to data while processing it into the long term archive

Archive files can be used directly as data products to science centers

Ability to archive by ground receipt and/or spacecraft time

Efficient processing & quick turn around of large amounts of data

Nov 2005L Herrera

Common Ground Approach

• Challenges• Increased cost of initial development• Increased difficulty in coordinating resources across

simultaneous program schedules• Increased resources & staff over commitment on

occasion• Increased need to negotiate requirements among

multiple missions• Sophisticated configuration management system is

needed• Tightly manage modifications to “generic” classes• Complexity of system configuration remains an issue

Problems now reported due to configuration

Nov 2005L Herrera

Common Ground Approach

• Benefits• Shared cost of development across multiple NASA missions

• MErcury Surface, Space ENvironment, GEochemistry, and Ranging (MESSENGER)

• Solar-Terrestrial Relations Observatory (STEREO)

• New Horizons

• Reduced redundancy & capitalized on domain knowledge• Supports the needs of all current missions• Increased reliability through repeated testing & use• Increased familiarity among users• Solid base to free up resources to add new levels functionality

to be layered on top• Decreased development time & cost for subsequent missions

• A telemetry archiving system can be brought on-line for a new mission in a week

Nov 2005L Herrera

Comparison of Total Change Requests

Comparisons of GSW Errors

0

5

10

15

20

25

30

35

40

45

May-02 Dec-02 J un-03 J an-04 Aug-04 Feb-05 Sep-05 Mar-06

MESSENGER

STEREO

New Horizons

Comparisons of GSW Enhancement Requests

0

5

10

15

20

25

30

35


MESSENGER

STERENew Horizons

Comparisons of GSW New Requirements

0

2

4

6

8

10

12


MESSENGER

STEREO

New Horizons

Nov 2005L Herrera

Comparison Shifted for Development Phase

Comparison of Errors, MESSENGER Data Shifted

0

5

10

15

20

25

30

35

40

45

erro

rs p

er m

on

th

MESSENGER

STEREO

New Horizons

All relativeI&T starts

Comparison of Enhancment Requests, MESSENGER Data Shifted

0

5

10

15

20

25

30

35

enh

ance

men

ts p

er m

on

th

MESSENGER

STEREO

New Horizons


Comparison of New Requirements, MESSENGER Data Shifted

0

2

4

6

8

10

12

new

req

uir

emen

ts p

er m

on

th

MESSENGER

STEREO

New Horizons


Nov 2005L Herrera

What To Tackle Next

• Release of complete Ground SW Infrastructure

• Address next level of functionality for Assessment PL• Rely on a sound infrastructure of telemetry data flow &

archiving• Advance satellite/spacecraft State of Health (SOH)

assessment tools

• Search for solutions & technologies offered by other organizations• Identify emerging technologies that JHU/APL could help

cultivate

developing a common telemetry archiving architecture

Documents