GOESR Ground Segment Architecture Overview Presentation to NSOF

1
GOES-R Ground SegmentArchitecture
OverviewPresentation to NSOF

• 17 September 2009

2
Ground Segment Overview
3
Flexible Architecture
Service Based Architecture uses very high
performance Distributed Data Fabric for
generating and moving large environmental
products.
SBA
Messaging between Elements uses industry standard
ESB to reduce development cost and support future
flexibility
4
GS Functional Flow Diagram
5
Mission Management
6
OS/COMET Integrated System
Mission Planning Harris Provided, COTS, or
Customer Provided
OS/COMET COTS Software
Mission Requirements
ASI / ASO
OS/Comet socket interface and API

• 3rd party products
• Data Analysis and Visualization

Orbit Determination Data
FDS or other Flight Dynamics Products
7
Product Generation and Distribution

• Product Generation and Distribution represents a
  set of complex requirements that require more
  than big computers and high data rate throughput
• Examples of issues dealt with in the
  architecture
• Processing of a continuous data stream (e.g.
  non-stop processing)
• Dealing with data errors
• Interdependence of data and products to create
  the final output product set
• Finite limitations of available computers and
  software
• Architecture obsolescence in the face of evolving
  COTS computer hardware/software products

8
Block Processing
ABI in Mode 4
S/C Sensor Observations
RF to WCDAS
Level 1b Processing
Level 1b / GRB Products
Last Block of Full Disk
RF Relay to NSOF (via GRB)
Level 2,2 Processing
Block Processing
Level 2,2 Products
L1b
Algorithm Precedence
Latency
Each L2

• A block is defined as the amount of data used for
  computation by a Level x algorithm. Block sizing
  needs to account for
• The minimum amount of data needed for computation
• Inter-scan dependencies

9
Data and Product Interdependence

• This is an example of a previously generated
  precedence
• Products are in most cases not standalone
• A complex precedence exists between products and
  intermediate products
• To be operationally efficient, we must be able to
  build new products by using services that can be
  developed off-line and efficiently linked into
  the precedence chain on-line
• WITH LOW OVERHEAD

10
PG Functional ViewFunctional Flow Block Diagram
11
PG Software Reference DesignDistributed Data
Fabric

• The Distributed Data Fabric provides an abstract
  interface to data
• Algorithms do not need to know how data is stored
• Algorithms can request blocks of data in the size
  they desire without having to understand how it
  is stored or originally received
• Distributed Data Fabric contains
• Event Manager Allows clients (Algorithms) to
  register and be notified when their desired data
  is available in the Distributed Data Fabric
• High Performance Data Cache RAM based data cache
  makes access to data fast and reduces potential
  bottlenecks by distributing the data across a
  configurable number of servers
• Storage Manager Moves data from Cache into Disk
  Based Storage and Databases

12
Metadata Flow
13
PG SBA Prototype V1 Operation
14
PG SBA Prototype V2 Operation
15
Operational SBA

• Development
• SBA Software Development Kit for round trip
  engineering
• Plug n play in both operational and research
  environment
• Infrastructure
• Workload Schedule Optimization
• Metadata Integration
• Variable Blocks Sizing
• User Interface
• Full Operator/Management Interface
• Remote maintenance capability
• Comprehensive Log Tracing (By Service or by
  Product)
• Reliability
• Improved Service fault detection
• Improved Service Failover Handling

16
Enterprise Management
Service Desk Operations ITIL Service Management
Best Practices Self-Service Web Enabled
GUIs Incident and Problem Reporting Moves,
Adds, Change Requests Auto Trouble Ticket
Generation and Escalation Procedures Configuratio
n Asset Management SLA Management
Reports Knowledgebase Convert Emails to Trouble
Tickets

• Enterprise Operations Console ETE Health
  Status Management
• Product Impact Status
• PG/PD Status Management
• can drill down to Product and user impact screens
• Storage DB Management
• Server Management
• Performance Management
• Space Segment Management View
• Report Management

Architecture ensures integration of Mission and
IT information
17
EM Functional Interaction
Each functional activity interacts with each
other in a unified model so that both real-time
operational data and back-office data remain
accurate and in sync.
18
Provides Detailed Product Monitoring

• Products and User Services can be viewed in a
  configurable Dashboard showing the Health and
  Status products and the ability of the users to
  receive them.
• If an anomaly occurs in any element of the
  Ground Segment, an will appear and an
  operator can drill down to see the components and
  the actual root cause that has impacted this
  service. Complete redundancy will not allow
  service interruptions to occur due to redundant
  configurations, but restoral procedures can begin
  and the proper personnel notified of the actual
  failed component or service.

X
19
Resources Monitoring Capability
20
Summary

• The GS Architecture has been designed for
  Flexibility, Scalability, Expandability, and
  Availability
• Evolution of the GS will continue to follow these
  tenets
• Additional requirement refinement and definition,
  consistent with change management/configuration
  management directives, will continue
• In particular OPSCON and Security related
  requirements