1. D. Schmidt DARPA Example: Navy UAV Concept & Representative Scenario 1. Video feed from off-board source (UAV) 2. Video distributor sends video to hosts on ship’s network 3. Users’ hosts receive video and display it 4. Users interact with UAV to control it in real time uav.navair.navy.mil/home.htm 2 1 3 4 Prototype Interactions

2. D. Schmidt DARPA QuO/HIPER-D Integration Architecture Distributor Video Forwarding Code UAV SIMULATION HOST Video Source Process UAV Video File Video Forwarding Code Instrumentation Input Queue Size Host Monitor Process CPU load Net load Video Distribution Process VIDEO DISTRIBUTION HOST Contract Resource Manager Adaptive Control Process Output Queue Size Video Display Code Video Display Process VIDEO DISPLAY HOST 1 Contract Output Rate Control (TBD) Other parameter (TBD) CPU load Output Rate Control VIDEO DISPLAY HOST N Video Forwarding Code Video Acceptor Process UAV Priority Host Monitor Code Instrumentation Other parameters Instrumentation

3. D. Schmidt DARPA Current Adaptation responses I P B I I I I I I I I P P P P P P P P P P B B B B B B B BB B B B B B B B BB B B B B B B B B B B B B B B B B B B B B B BBBBBBBBBBBBBB B BBB BBBBBBBBBBBBBBBBBBBBBBBBBBBB P P P P P P P P P P I BB P BB P BB P BB P BB P BB I BB P B NETWORK RESERVATION Under excessive Network load - Use IntServ to reserve bandwidth... P BB P BB P BB II I I DATA FILTERING Excessive network or CPU load - Drop selective frames LOAD BALANCING Excessive CPU load - Move distributor to more lightly loaded host Variations in Operating Conditions Variations in Mission Requirements Timeliness Pilot or targeting officer must have an out-of-the- window view of UAV imagery Quantity Surveillance officer must record complete UAV imagery for off-line analysis

4. D. Schmidt DARPA Layers of QoS Specification & Adaptation in UAV Systems IntServ & DiffServ A/V Streaming Service, HiperD RM, Open QoS Testbed RM, RT ARM, DeSiDeRata, QuO, Proteus Mission doctrine contracts (TBMD, AAW, CFF) Emerging Alternatives Candidate Technologies Linux RKRT Java QuO Gateway RT CORBA (ACE+TAO), Distributed RT Java Adaptive feedback loops can run at multiple layers Distribution Middleware QoS Network QoS Common Middleware Service QoS System-wide QoS Operating System QoS Application or Domain- specific QoS UAV capabilitiesCEC/SIAP Hybrid & MultiChannel

5. D. Schmidt DARPA Current Layers of QoS Specification & Adaptation in Navy UAV Systems Network QoS Common Middleware Service QoS Operating System QoS Application or Domain- specific QoS Distribution Middleware QoS Video Forwarding Code Video Source Video Display Hyper-D Resource Manager ACE/TAO RT ORB IntServ/RSVP AQoSA IntServ/RSVP AQoSA QoS Adaptive Control A/V Streaming Service

6. D. Schmidt DARPA AdHoc Integration of Components for QoS Adaptation and Control Video Forwarding Code Video Source Video Display Distributed Resource Management Coordination ACE/TAO RT ORB A/V Streaming Service Data Path ACE/TAO RT ORB LAN Network Operating System Operating System Wireless Network Control Path Data Path Control Path ACE/TAO RT ORB A/V Streaming Service A/V Streaming Service AQoSA IntServ/RSVP AQoSA Virtual Information Collection IntServ/RSVP Operating System

7. D. Schmidt DARPA UAV/HIPER-D Requirements (Previous Experiment)  Low latency to support interaction (users see images at the same time as the UAV)  Displayed frame rate can be less than 30/second, providing that targets remain clear and no jitter  HIPER-D Resource Manager determines where and when applications run  Management techniques focused on discrete problem and remedy  New experiments will extend these basic ideas: – Individual and composite bottleneck identification and adaptations – End-to-end behavior – Aggregate and Coordinated behavior – Scaling and Redundancy – Varied anomalies and operating conditions – More resources under control/coordination, including “soft” resources – Intercluster coordination and feedback

8. D. Schmidt DARPA UAV demonstration illustrates some of the software engineering challenges with reusing and adding QoS to current off-the-shelf component software  We used an off-the-shelf video player in the UAV demonstration – Developed for playing MPEG video from a file – Had to convert it to accept input from a stream  Developers of the video player had recognized the need for adaptation to handle changes in QoS – The video player included code to compensate for slow video cards (i.e., falling behind in the video) – Unfortunately, this code is intertwined throughout the functional code (i.e., there is no separation of concerns)  Reusing this code presented some challenges because the QoS code was intertwined and specific to a different use-case – We had to “turn off” the file-specific adaptation in order to use the video player effectively with a video stream – This was difficult because the adaptive code was intertwined throughout the functional code  In contrast, the adaptive code specified separately in the QuO middleware was easy to change

9. D. Schmidt DARPA ClientObject ORB endsystem ORB endsystem Resource Applying Reflection to Optimize Multi-level Resource Management Key System Characteristics Integrate observing & predicting of current status & delivered QoS to inform the meta-layer Meta-layer applies reflection to adapt system policies & mechanisms to enhance delivered QoS QoS Doctrine Applying reflection as an optimization is even more relevant to middleware than compilers due to dynamism & global resources: Piggybacked Measurements Expected QoS Measured QoS Correlate Probes Status Resource Management Service Collect Translate Integrate Infer/Adapt Feedback Loop Interceptor Monitors

10. D. Schmidt DARPA Key Research Challenge: Providing & Organizing QoS Guarantees for Multiple Adaptive Feedback Loops Multi-level distributed resource management middleware Support stable QoS at varying granularity & scope levels for integrated, multi- property feedback paths across different locations & time scales Patterns, protocols, & architectures needed to integrate COTS components ClientObject Combined System-level & Application-level Management Feedback Local Resource- centric Feedback Local Resource- centric Feedback End-to-End Application-centric Feedback End-to-End Application-centric Feedback End-to-End Network-centric Feedback End-to-End Network-centric Feedback Solution Approach

11. D. Schmidt DARPA Integrated Adaptive System Concept System-wide QoS Distribution Middleware QoS Network QoS Common Middleware Services QoS Operating System QoS Application or Domain-specific QoS ACE/TAO RT ORB IntServ/RSVP Operating System IntServ/RSVP Operating System IntServ/RSVP Operating System IntServ/RSVP Operating System IntServ/RSVP Operating System IntServ/RSVP Operating System ACE/TAO RT ORB

12. D. Schmidt DARPA Fundamental Research Challenges Addressed by UAV Demo Scenario Develop, integrate, & showcase DARPA technologies that: Provide “time-bound dependability” that is (1) specified declaratively & (2) respects (re)configured battle force (multi)mission doctrines Illustrate management & control of multiple resources that ensures system stability in the presence of adaptivity & feedback at multiple layers e.g., network adaptivity, application-adaptivity, & end-to-end system adaptivity Demonstrate the feasibility of “pluggable” components & adapters that enforce QoS-enabled services at various layers Show integrated failure management & information assurance across multiple DARPA+COTS technologies Integrate with advanced NSWC & Open QoS Testbed instrumentation, monitoring, & dynamic resource management capabilities