Addressing the Challenges of Tactical Information Management in Net-Centric Systems With DDS


Published on

This paper provides an overview of the advantages provided by the OMG Data Distribution Service for Real-Time Systemts (DDS) for addressing the challenges associated with Tactical Information distribution.

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Addressing the Challenges of Tactical Information Management in Net-Centric Systems With DDS

  1. 1. Software Engineering Technology Addressing the Challenges of Tactical Information Management in Net-Centric Systems With DDS Dr. Douglas C. Schmidt, Dr. Angelo Corsaro, and Hans van’t Hag PrismTech Corporation Recent trends in net-centric systems motivate the development of tactical information management capabilities that ensure the right information is delivered to the right place at the right time to satisfy quality of service (QoS) requirements in heteroge- neous environments. This article presents an architectural overview of the Object Management Group’s (OMG) Data Distribution Service (DDS), which is a standards-based QoS-enabled data-centric middleware platform that enables applica- tions to communicate by publishing information they have and subscribing to information they need in a timely manner. DDS is an important distributed software technology for mission-critical Department of Defense (DoD) net-centric systems because it supports the following: (1) location independence, via anonymous publish/subscribe (pub/sub) protocols that enable commu- nication between collocated or remote publishers and subscribers, (2) scalability, by supporting large numbers of topics, data T readers, and data writers and platform portability, and (3) interoperability, via standard interfaces and transport protocols. actical information management sys- the Common Object Request Broker works. tems increasingly run in net-centric Architecture [CORBA] Event Service and • An orchestrated information bus environments characterized by thousands Notification Service, the Java Message that aggregates, filters, and prioritizes of platforms, sensors, decision nodes, and Service [JMS], and various other propri- the delivery of this information to computers connected together to etary middleware products) have histori- work effectively under the restrictions exchange information, support sense- cally lacked key architectural and QoS of transient and enduring resource making, enable collaborative decision capabilities, such as dependability, surviv- constraints. making, and effect changes in the physical ability, scalability, determinism, security, • Continuous adaptation to changes environment. For example, the Global and confidentiality needed by net-centric in the operating environment, such as Information Grid (GIG) is an ambitious systems for tactical information manage- dynamic network topologies, publisher net-centric environment being designed to ment. To address these limitations – and and subscriber membership changes, ensure that different services and coalition to better support tactical information and intermittent connectivity. partners, as well as individuals participat- management in net-centric systems like • Standard QoS policies and mecha- ing to specific missions, can collaborate the GIG – the OMG has adopted the nisms that enable applications and effectively and deliver appropriate fire- DDS specification, which is a standard for administrators to customize the way power, information, or other essential QoS-enabled data-centric pub/sub com- information is delivered, received, and assets to warfighters in a timely, depend- munication aimed at net-centric tactical processed in the appropriate form and able, and secure manner [1]. Achieving information management systems [2]. level of detail to users at multiple lev- this vision requires the following capabili- DDS is used in a wide range of military els in net-centric systems. ties from the distributed middleware soft- and commercial systems including naval This section describes the key capabil- ware: combat management systems, commercial ities and entities in DDS and shows how • Shared operational picture. A key air traffic control, transportation manage- its QoS policies can be used to specify and requirement for mission-critical net- ment, automated stock trading systems, enforce performance-related requirements centric systems is the ability to share and semiconductor fabrication devices. of net-centric tactical information man- an operational picture with planners, The remainder of this article presents agement systems. Figure 1 shows the vari- warfighters, and operators in real-time. an overview of DDS that is geared to ous profiles and layers in the DDS stan- • Ensure the right data gets to the software architects. We also discuss the dard. The lower layer defines a Data- right place at the right time by satis- DDS QoS policies that are the most rele- Centric Publish Subscribe (DCPS) plat- fying end-to-end QoS requirements, vant for net-centric tactical information form, whose goal is to provide efficient, such as latency, jitter, throughput, management systems. Finally, we explain scalable, predictable, and resource-aware dependability, and scalability. how DDS has been applied in practice to data distribution. The higher layer is the • Interoperability and portability in address key challenges of developing and Data Local Reconstruction Layer (DLRL), heterogeneous environments. Since operating distributed software in current which is an optional interface that pro- net-centric systems are faced with and planned net-centric tactical informa- vides an object-oriented facáde atop the unprecedented challenges in terms of tion management systems. DCPS. The DLRL can be used to map platform and network heterogeneity, topics onto object fields and defines navi- they are necessary. Overview of DDS gable associations between objects. • Support for dynamic coalitions. In DDS provides the following capabilities A separate specification, called the many net-centric tactical information for net-centric tactical information sys- Real-Time Publish/Subscribe (RTPS) management systems, dynamically tems: DDS interoperability wire protocol, formed coalition of nodes will need to • Universal access to information defines the standard network protocol share a common operational picture from a wide variety of sources that run used to exchange data between publishers and exchange data seamlessly. over potentially heterogeneous hard- and subscribers that use different imple- Prior middleware technologies (such as ware/software platforms and net- mentations of DDS [3]. The remainder of 24 CROSSTALK The Journal of Defense Software Engineering March 2008
  2. 2. Addressing the Challenges of Tactical Information Management in Net-Centric Systems With DDS this section describes the conceptual model of DDS and explains the QoS poli- cies that are most relevant for net-centric tactical information management systems. Data Local Reconstruction Layer (DLRL) DDS Conceptual Model DDS applications send and receive data Domains and Partitions within a domain. Domains can be divided into partitions that allow the separation and protection of different data flows. Although DDS entities can belong to dif- ferent domains, only participants within the same domain can communicate, which helps isolate and optimize communication Data Centric Publish Subscribe (DCPS) within communities that share common Figure 1: Profiles and Layers in the DDS Standard interests. For example, each communica- tion layer within the GIG could be associ- object-oriented models. DDS thus allows In net-centric tactical information sys- ated with a DDS domain and further sub- the expression of the system information tems, an information model will be associ- divided into partitions. This approach iso- model as either a 1) topic relational model, ated with every layer in which DDS-based lates domain participants across layers, which can be thought of as an extension data exchange occurs. This information which enables effective use of resources of the familiar entity relationship diagrams model, which can comply with DoD or and helps enforce security and confiden- used in data bases, decorated with QoS, or North American Trade Organization stan- tiality policies. 2) an object-oriented model, which can dards, is the lingua franca used by the dif- also be synthesized as an object-oriented ferent applications in coalitions to B view of the relational model. m exchange information and seamlessly DDS provides a strongly typed global data The DCPS layer provides support for interoperate. Likewise, the QoS policies Global Data Space space within each domain in which appli- relational modeling, while the DLRL decorating the information model deter- A F cations produce and consume the dynam- extends the DCPS with an object-oriented mine how the data is disseminated, per- ically changing portions of a shared infor- facade, so that applications can either sisted, and receivedC in the global data J completely ignore the DCPS relational space.(DLRL) D mation model, as shown in Figure 2. DDS’ models or build an object model atop the Data Local Reconstruction Layer information model capabilities are similar to those of relational databases, except DLRL. Data associated with DDS topics Publishers and Subscribers K E that DDS’ global data space is completely are expressed using types defined by the In net-centric tactical information man- distributed, QoS-aware, and allows anony- standard OMG Interface Definition agement systems, publishers and sub- mous and asynchronous sharing of a Language (IDL), which simplifies the scribers correspond to a range of domain common information model. The DDS inter-working between DDS and CORBA. participants such as embedded devices, information model is the only knowledge Relationships between topics can be cap- Unmanned Air Vehicles (UAVs), soldiers’ publishers and subscribers need to com- tured via keys that can be used to distin- equipment, as well as planning and simula- municate, i.e., they need not be aware of guish between different instances of the tion services in operations centers. DDS each other nor be concerned with low- same topic. applications use data writers to publish Name Data Centric Publish Subscribe (DCPS) level network programming details, such Figure 2: DDS Global Data Space in a Domain as Internet protocol addresses, port num- bers, remote object references, or service Type Topic QoS names. By allowing data to flow where and when needed, DDS’s global data space enables the sharing of tactical information and situational awareness information needed to implement net-centric tactical information management systems. B m A DDS topic is an association between a Topic data type, a set of QoS, and a unique A F name, as shown in Figure 3 (see page 26). A topic is also the unit of information J D C contained in DDS’ global data space and is used by applications to define their information model and associate QoS K E policies with it. DDS applications in net- centric systems define their information model by identifying topics that are rele- vant for solving their requirements and organizing them into either relational or March 2008 Name 25
  3. 3. Software Engineering Technology described, DDS also supports content-based capability supports dynamic scenarios subscription, in which a subset of the stan- common in net-centric tactical informa- dard Structured Query Language (SQL) is tion management where cooperating Name used to specify subscription filters. In domain participants join and leave DDS a matching subscription must match throughout system operation. the following two types of a topic’s prop- Type Topic QoS erties: (1) its features, such as its type, Figure 3: DDS Topic name, key, and content; (2) its QoS poli- DDS is designed for mission-critical net- QoS Policies data values to the global data space of a cies, which are described in the QoS Policies centric systems where the right answer domain and data readers to receive data. A section. delivered too late becomes the wrong publisher is a factory that creates and The matching process for QoS uses a answer. To meet timing requirements it is manages a group of data writers with sim- requested/offered (RxO) model shown in essential that the middleware controls and ilar behavior or QoS policies, as shown in Table 1, where the requested QoS must be optimizes the use of resources, such as net- Figure 4. A subscriber is a factory that cre- less than or equal to the offered QoS. For work bandwidth, memory, and CPU time. ates and manages data readers, as shown example, subscribers requesting reliable Table 1 shows the rich set of QoS policies in Figure 4. data delivery cannot communicate with that DDS provides to control and limit Publishers can declare their intent to publishers that only distribute data using topic (T), data reader (DR), data writer produce data on particular topics with best effort delivery. Likewise, subscribers (DW), publisher (P), and subscriber (S) cannot request a topic update whose dead- resources and topic QoS properties, such associated QoS, and they distribute the line is smaller than that declared by any as persistence, reliability, and timeliness [2]. data on those topics to the global data publishers. Below we discuss the DDS QoS policies space. Subscribers receive topic data in the The subscription matching mechanism that are the most relevant for net-centric global data space that match their sub- tactical information management systems. provided by DDS enforces a powerful scriptions (the rules that define what rep- form of design by contract [4], where QoS is resents a matching subscription are used together with type information to described below). QoS policies allow pub- decide whether publishers and subscribers DDS provides the following QoS policies Data Availability lishers and subscribers to define, first, can communicate. This extended form of that control the availability of data to their local behavior, such as the number of design by contract helps ensure that net- domain participants: historical data samples they require and centric systems will operate as intended, • The Durability QoS policy controls the maximum update-rate at which they both from functional and QoS perspec- the lifetime of the data written to the want to receive data, and, second, how tives. These assurances are essential in the global data space in a domain. data should be treated once in transit with development, deployment, and operation Supported durability levels include the respect to reliability, urgency, importance, of mission-critical net-centric tactical following: (1) volatile, which specifies and durability. Topics can also be annotat- information management systems. that once data is published it is not ed with these QoS policies to drive the maintained by DDS for delivery to late behavior of the data-distribution. The joining applications; (2) transient local, QoS policies of pre-defined topics serve Another key feature of DDS is that all which specifies that publishers store Discovery as defaults for publishers and subscribers information needed to establish commu- data locally so that late joining sub- and can therefore ensure consistency nication can be discovered automatically, scribers get the last published item if a between requested and offered QoS. in a completely distributed manner. publisher is still alive; (3) transient, Applications dynamically declare their which ensures that the global data intent to become publishers and/or sub- space maintains the information out- A subscription is an operation that associ- scribers of one or more topics to the DDS side the local scope of any publishers Subscriptions and Matching ates a subscriber to its matching publish- middleware, which uses this information for use by late joining subscribers; and ers, as shown in the center of Figure 4. In to establish the proper communication (4) persistent, which ensures that the addition to the topic-based subscriptions paths between discovered entities. This global data space stores the informa- Figure 4: DDS Publisher/Writer Subscriber/Reader and Subscription/QoS Matching tion persistently so it is available to late joiners even after the shutdown and restart of the system. Durability is achieved by relying on a durability ser- QoS matching vice whose properties are configured by means of the DURABILITY_SER- QoS QoS QoS VICE QoS. • The LIFESPAN QoS policy controls QoS Topic QoS Name the interval of time during which a data sample is valid. The default value DataWriter writes Type reads DataReader is infinite, with alternative values being Publisher ... ... ... Subscriber the time-span for which the data can be considered valid. DataWriter writes Type reads DataReader • The HISTORY QoS policy controls Name the number of data samples (i.e., sub- Topic sequent writes of the same topic) that must be stored for readers or writers. QoS QoS QoS Possible values are the last sample, the QoS matching 26 CROSSTALK The Journal of Defense Software Engineering March 2008
  4. 4. Addressing the Challenges of Tactical Information Management in Net-Centric Systems With DDS last n samples, or all samples. These QoS policies provide the DDS QoS Policy Applicability RxO Modifiable global data space with the ability to coop- T, DR, DW Y N erate in highly dynamic environments DURABILITY characterized by continuous joining and DURABILITY T, DW N N Data leaving of publisher/subscribers. This SERVICE capability makes it possible for net-centric Availability LIFESPAN T, DW - Y tactical information management systems HISTORY T, DR, DW N N to share a common operational picture regardless of the dynamism that charac- PRESENTATION P, S Y N terizes portions of the systems, such as RELIABILITY T, DR, DW Y N coalitions of soldiers collaborating in P, S N Y urban environments or coordinated UAVs PARTITION in support of tactical operations. DESTINATION T, DR, DW Y N Data Delivery ORDER OWNERSHIP T, DR, DW Y N DDS provides the following QoS policies Data Delivery DW - Y that control how data is delivered and OWNERSHIP STRENGTH which publishers are allowed to write a specific topic: DEADLINE T, DR, DW Y Y • The PRESENTATION QoS policy LATENCY T, DR, DW Y Y gives control on how changes to the Data BUDGET information model are presented to Timeliness TRANSPORT T, DW - Y subscribers. This QoS gives control on PRIORITY the ordering as well as the coherency of data updates. The scope at which it TIME BASED DR - Y is applied is defined by the access FILTER Resources scope, which can be one of RESOURCE T, DR, DW N N INSTANCE, TOPIC, or GROUP LIMITS level. USER_DATA DP, DR, DW N Y • The RELIABILITY QoS policy con- trols the level of reliability associated TOPIC_DATA T N Y Configuration with data diffusion. Possible choices GROUP_DATA P, S N Y are RELIABLE and BEST_EFFORT Table 1: Key QoS Policies for Net-Centric Systems distribution. offered by the DDS content-awareness or with a topic instance, thus allowing • The PARTITION QoS policy gives profile that allows applications to select a DDS implementation to prioritize control over the association between information of interest based upon their more important data relative to less DDS partitions (represented by a content. important data. string name) and a specific instance of These QoS policies make it possible to a publisher/subscriber. ensure that tactical information needed to • The DESTINATION_ORDER QoS DDS provides the following QoS policies reconstruct the shared operational picture Data Timeliness policy controls the order of changes that control the timeliness properties of is delivered in a timely manner. made by publishers to some instance distributed data: of a given topic. DDS allows the • The DEADLINE QoS policy allows ordering of different changes accord- applications to define the maximum DDS defines the following QoS policies Resources ing to source or destination time- inter-arrival time for data. DDS can be to control the network and computing stamps. • The OWNERSHIP QoS policy con- configured to automatically notify resources that are essential to meet data trols which writer owns the write- applications when deadlines are dissemination requirements: access to a topic when there are multi- missed. • The TIME_BASED_FILTER QoS ple writers and ownership is EXCLU- • The LATENCY_BUDGET QoS pol- policy allows applications to specify SIVE. Only the writer with the highest icy provides a means for applications the minimum inter-arrival time OWNERSHIP_STRENGTH can to inform DDS of the urgency associ- between data samples, thereby publish the data. If the OWNERSHIP ated with transmitted data. The latency expressing their capability to consume QoS policy value is shared, multiple budget specifies the time period with- information at a maximum rate. writers can concurrently update a in which DDS must distribute the Samples that are produced at a faster topic. OWNERSHIP thus helps to information. This time period starts pace are not delivered. This policy manage replicated publishers of the from the moment the data is written helps a DDS implementation optimize same data. by a publisher until it is available in the network bandwidth, memory, and pro- These QoS policies control the relia- subscriber’s data-cache ready for use cessing power for subscribers that are bility and availability of the data, thus by reader(s). connected over limited bandwidth net- allowing the delivery of the right data to • The TRANSPORT_PRIORITY QoS works or which have limited comput- the right place at the right time. More elab- policy allows applications to control ing capabilities. orate ways of selecting the right data are the importance associated with a topic • The RESOURCE_LIMITS QoS poli- March 2008 27
  5. 5. Software Engineering Technology cy allows applications to control the next-generation combat management sys- The DDS implementation used on amount of message buffering per- tems, and Defense Information Systems TACTICOS supports a data-centric formed by a DDS implementation. Agency as the standard technology for approach where at the start of the system DDS’s QoS policies support the vari- publish/subscribe to be used in all new or design, the information model can be cap- ous elements and operating scenarios that upgraded systems [5, 6]. Several major tured, annotated with proper QoS poli- constitute net-centric tactical information defense programs, such as the U.S. Navy’s cies, and then shared between multiple management. By controlling these QoS DDG-1000 land attack destroyer, U.S. parties. This federated architecture is com- policies it is possible to scale DDS from Army’s Future Combat Systems (FCS), mon in existing and planned coalition-based low-end embedded systems connected and the Thales Tactical Information And developments where multiple parties with narrow and noisy radio links, to high- Command System Operating System jointly implement the overall combat sys- end servers connected to high-speed (TACTICOS), also adopted DDS even tem. DDS provides the fault-tolerant fiber-optic networks. before it was mandated, underscoring information backbone onto which all DDS’ ability to address the data distribu- these applications are deployed and is thus tion challenges of next generation net- responsible for providing each application The QoS policies described above provide centric defense systems. with the right information at the right Configuration control over the most important aspects For example, the TACTICOS combat time. of data delivery, availability, timeliness, management system developed by Thales Along with the rapid adoption of and resource usage. In addition, DDS also Naval Netherlands is based on an imple- DDS in the defense domain, its use is also supports the definition and distribution of mentation of DDS that allows them to steadily growing in other domains, such as user specified bootstrapping information achieve very good scalability, from small transportation, telecommunications, and via the following QoS policies: finance. For example, in the context of • The USER DATA QoS policy allows Air Traffic Control and Management, applications to associate a sequence of DDS provides the DDS has been selected as the octets to domain participant data read- publish/subscribe middleware for distrib- ers and data writers. This data is then fault-tolerant information uting flight data plans in CoFlight [9], distributed by means of the DCPS which is the next generation European participant built-in topic. This QoS backbone onto which Flight Data Processor. In general, DDS is policy is commonly used to distribute an appropriate middleware technology for security credentials. all these applications are application domains that require rich sup- • The TOPIC_DATA QoS policy port for QoS policies and high-perfor- allows applications to associate a deployed and is thus mance and dependability standards-based, sequence of octets with a topic. This commercial-off-the-shelf implementa- bootstrapping information is distrib- responsible for providing tions. uted by means of the DCPS Topic built-in topic. A common use of this each application with Concluding Remarks QoS policy is to extend topics with DDS is a standards-based QoS-enabled additional information, or meta-infor- the right information data-centric publish/subscribe middle- mation, such as eXtensible Markup ware that provides a feature rich data-cen- Language schemas. at the right time. tric real-time platform to support the • The GROUP_DATA QoS policy needs of current and planned net-centric allows applications to associate a ships to aircraft carrier grade, as well as tactical information management systems. sequence of octets with publishers and high performance, availability, and deter- Its powerful set of QoS policies, together subscribers. This bootstrapping infor- minism even under temporary overload with its scalable architecture, makes it an mation is distributed by means of the conditions [7, 8]. TACTICOS is currently effective and mature choice for solving the DCPS subscription,and DCPS publi- in use in 15 navies worldwide serving 20 data distribution and information manage- cation built-in topics, respectively. A ships-classes ranging from small patrol ment problems net-centric systems [10]. typical use of this information is to boats up to large frigates. The utilization Next, we summarize how DDS addresses allow additional application control of DDS is instrumental in its success the key challenges outlined in the intro- over subscriptions matching. since it provides both the scalability to duction in a standard and interoperable support thousands of applications run- manner: ning on more than 150 distributed com- • Shared operational picture. DDS puters on a frigate size system. Another DDS Success Stories Although DDS is a relatively new standard provides effective support for these (adopted by the OMG in 2004), it has key feature of DDS is its battle-damage types of applications via its QoS poli- been adopted quickly due to its ability to resistance, meaning that software can be cies for defining the scope, content, address key requirements of data distribu- dynamically re-allocated to the remaining and QoS of the data model that under- tion in net-centric systems, as well as the computer pool in case of an error on a lies the operational picture. maturity and quality of available imple- specific computer. The DDS Persistence • The right data at the right time at mentations, which are based on decades of Profile support is instrumental in this the right place via DDS QoS policies experience developing data-centric mid- dynamic reallocation since it allows appli- that enable a fine-grained control over dleware for mission-critical systems. cations to store their internal state into the information delivery, such as the abili- Moreover, DDS has been mandated by DDS middleware, which manages this ty to control many aspects of data dis- the U.S. Navy’s Open Architecture state in a distributed and fault-tolerant way semination to ensure timely delivery Computing Environment as the standard so that restarted applications can continue and optimal resource usage. publish/subscribe technology to use in what they were doing before they crashed. • Heterogeneous environment. By 28 CROSSTALK The Journal of Defense Software Engineering March 2008
  6. 6. Addressing the Challenges of Tactical Information Management in Net-Centric Systems With DDS providing standard QoS policies that dardized. The OMG will be addressing Protocol Specification.” <www.omg. control the bandwidth used for pro- this area of standardization starting in the org/cgi-bin/apps/doc?ptc/06-08 viding data to interested parties, DDS spring of 2008. -02.pdf>. runs in heterogeneous platforms while With multiple COTS and open-source 4. Bertrand, Meyer. Object Oriented providing different elements with a implementations and a solid track record Software Construction. 2nd ed., common operational picture. of success in mission-critical military and Prentice Hall, 2001. • Dynamic coalitions. The highly commercial projects, DDS has a bright 5. “Open Architecture Computing dynamic nature of DDS, such as its future as the standards-based middleware Environment.” < support for dynamic discovery, pro- of choice for net-centric tactical informa- wwwDL/B/OACE>. vides an effective platform for sup- tion systems. More information on DDS 6. Defense Systems Information Agency. porting ad hoc interactions. and its application in practice are available “DoD Information Technology DDS continues to evolve to meet new in online forums [11, 12] where experts Standards Registry.” <https://disron operational and technical challenges of discuss advanced features of the DDS>. net-centric tactical information manage- standard and new directions for the tech- 7. THALES. “TACTICOS Combat ment systems. Three types of extensions nology, while DDS beginners can learn Management System, Exploiting the are currently being pursued for DDS by from past experiences and ask questions Full DDS Potential.” <www.omg. the OMG. The first involves adding new about patterns and best practices for platform-specific models that fully lever- org/docs/dds/06-12-06.pdf.>. applying DDS in their net-centric sys- age programming language features, such tems.N 8. “OpenSplice DDS.” <www.prismtech. as standard C++ containers. The second com/openplice-dds>. extension deals with extensible topics that 9. CoFlight eFDP < dds/07-07-04.pdf>. References enable incremental system updates by 1. “The United States Department of ensuring that changes in the data model Defense Quadrennial Defense Review 10. Xiong, Ming, et al. “Evaluating do not break interoperability. The final set Report.” Feb. 2006 <www.defense Technologies for Tactical Information of extensions focus on network data rep- Management in Net-Centric Systems.” resentation and the syntax used to define pdf>. Proceedings of the Defense Transfor- topics. For example, upcoming versions of 2. OMG. “Data Distribution Service for mation and Net-Centric Systems Con- the DDS standard will likely allow the def- Real-Time Systems Specification.” ference, Apr. 9-13, 2007, Orlando, FL. inition of topics using XML, as well as the < 11. OMG DDS SIG Portal <portals. use of XML or Java Script Object -02.pdf>.>. Notation as the network data representa- 3. OMG. “Real-Time Publish Subscribe 12. OMG DDS Forum < tion. DDS security has not yet been stan- Protocol – DDS Interoperability Wire org>. About the Authors Douglas C. Schmidt, Angelo Corsaro, Ph.D., Hans Van’t Hag is the Ph.D., is a professor of is the OpenSplice DDS OpenSplice DDS prod- computer science at product marketing man- uct manager, at Vanderbilt University ager at PrismTech and PrismTech. He has and is the chief technical co-chairs the OMG DDS extensive experience in officer of PrismTech. Special Interest Group applying an information His expertise focuses on distributed and the Real-Time Embedded and approach towards mission-critical and computing middleware, object-oriented Specialized Services task force. He is real-time net-centric systems. Hag is a patterns and frameworks, and distrib- well-known in the distributed real-time contributor to the OMG DDS specifica- uted real-time and embedded systems. and embedded systems middleware tion and has presented numerous papers He has authored nine books and more community and has a wealth of experi- on DDS and publish/subscribe middle- than 350 papers in top technical journals, ence in hard real-time embedded sys- ware technologies. Prior to joining conferences, and books that cover high- tems, large-scale, and very large-scale PrismTech, he worked at Thales Naval performance communication software distributed systems such as defense, Netherlands (TNN) where he was systems, real-time distributed comput- aerospace, homeland security and trans- responsible for the development of the ing, and object-oriented patterns for portation systems. data-centric real-time middleware as concurrent and distributed systems. applied in TNN’s combat system in ser- PrismTech Corporation vice with 15 navies worldwide. PrismTech Corporation 6 Lincoln Knoll LN 6 Lincoln Knoll LN STE 100 PrismTech Corporation STE 100 Burlington, MA 01803 6 Lincoln Knoll LN STE 100 Burlington, MA 01803 Phone: (781) 270-1177 Burlington, MA 01803 Phone: (781) 270-1177 Fax: (781) 238-1700 Phone: (781) 270-1177 Fax: (781) 238-1700 E-mail: angelo.corsaro Fax: (781) 238-1700 E-mail: E-mail: hans.vanthag March 2008 29