NetQoPE: A Model-Driven Network QoS Provisioning Engine ...papers/cpsweek08/papers/rtas08/3C.pdf · (NetQoS), which is a domain-specific modeling lan-guage (DSML) that supports design-time

NetQoPE: A Model-driven Network QoS Provisioning Enginefor Distributed Real-time and Embedded Systems∗

Jaiganesh Balasubramanian†, Sumant Tambe†, Balakrishnan Dasarathy‡,Shrirang Gadgil‡, Frederick Porter‡, Aniruddha Gokhale†, and Douglas C. Schmidt†

†Department of EECS, Vanderbilt University, Nashville, TN, USA‡Telcordia Technologies, Piscataway, NJ, USA

Abstract

This paper provides two contributions to the study ofquality of service (QoS)-enabled middleware that supportsthe network QoS requirements of distributed real-time andembedded (DRE) systems. First, we describe the designand implementation of NetQoPE, which is a model-drivencomponent middleware framework that shields applicationsfrom the details of network QoS mechanisms by (1) speci-fying per-flow network QoS requirements, (2) performingresource allocation and validation decisions (such as ad-mission control), and (3) enforcing per-flow network QoSat runtime. Second, we evaluate the effort required andflexibility of using NetQoPE to provide network QoS assur-ance to end-to-end application flows. Our results demon-strate that NetQoPE can provide network-level differenti-ated performance to each application flow without modify-ing its programming model or source code, thereby provid-ing greater flexibility in leveraging network-layer mecha-nisms.

1 Introduction

Emerging trends. Distributed real-time and embedded(DRE) systems, such as shipboard computing systems, su-pervisory control and data acquisition (SCADA) systems,and enterprise security and hazard sensing subsystems, con-sist of multiple communication-intensive applications withmultiple end-to-end application flows. These systems havenetwork quality of service (QoS) requirements, such as lowend-to-end roundtrip latency and jitter, that must be satis-fied under varying levels of network connectivity and band-width availability. Network QoS mechanisms, such as in-tegrated services (IntServ) [12] and differentiated services(DiffServ) [2], help provide diverse network service levelsfor applications in DRE systems.

For example, applications can use advanced networkQoS mechanisms (e.g., a DiffServ bandwidth broker [3]) to(1) request a network service level and (2) allocate and man-

∗This work is supported in part or whole by DARPA Adaptive and Re-flective Middleware Systems Program Contract NBCH-C-03-0132. Anyopinions, findings, conclusions or recommendations expressed in this ma-terial are those of the author(s) and do not necessarily reflect the viewsof the Department of the Interior, National Business Center, AcquisitionServices Division, Southwest Branch or DARPA.

age network resources for their remote invocations. Appli-cations invoke remote operations by adding a service level-specific identifier (e.g., DiffServ codepoint (DSCP)) to theIP packets. DiffServ-enabled network routers parse the IPpackets and provide the appropriate service level-specificpacket forwarding behavior.

Limitations with current approaches. Although ad-vanced network QoS mechanisms are powerful, it is te-dious and error-prone to develop applications that interactdirectly with low-level network QoS mechanism APIs writ-ten imperatively in third-generation languages, such as C++or Java. To overcome this problem, middleware-based so-lutions [22, 18, 16, 4] have been developed that allow ap-plications to specify their coordinates (source and destina-tion IP and port addresses) and per-flow network QoS re-quirements via higher-level frameworks. The middlewareframeworks—rather than the applications—are responsiblefor converting the higher-level QoS specifications into thelower-level network QoS mechanism APIs.

Although middleware frameworks alleviate many acci-dental complexities of low-level network QoS mechanismAPIs, they can still be hard to evolve and extend. In par-ticular, application source code changes may be necessarywhenever changes occur to the deployment contexts (sourceand destination nodes of the applications), per-flow require-ments, IP packet identifiers, or the middleware APIs. Whatis needed, therefore, are middleware-guided network QoSprovisioning solutions that (1) are not tied to a particularnetwork QoS mechanism and (2) do not modify applica-tion source code to specify and enforce network QoS re-quirements. These solutions should ideally operate on well-defined system abstractions (e.g., per-flow requirements andsource/destination nodes) that do not require programmat-ically modifying application source code, thereby facilitat-ing application reuse across a wide range of deployment andnetwork QoS contexts.

Solution approach → A model-driven componentmiddleware network QoS provisioning framework thatuses declarative domain-specific techniques [1] to raise thelevel of abstraction of DRE system design higher than us-ing imperative third-generation programming languages. Amodel-driven framework allows system engineers and soft-ware developers to perform deployment-time analysis (suchas schedulability analysis [10]) of non-functional system

IEEE Real-Time and Embedded Technology and Applications Symposium

1080-1812/08 $25.00 © 2008 IEEEDOI 10.1109/RTAS.2008.32

113

IEEE Real-Time and Embedded Technology and Applications Symposium

1080-1812/08 $25.00 © 2008 IEEEDOI 10.1109/RTAS.2008.32

113

IEEE Real-Time and Embedded Technology and Applications Symposium

1080-1812/08 $25.00 © 2008 IEEEDOI 10.1109/RTAS.2008.32

113

properties (such as network QoS assurances for end-to-end application flows). Such an analysis helps providedeployment-time assurance that application QoS require-ments will be satisfied.

This paper describes the Network QoS Provisioning En-gine (NetQoPE), which is a model-driven component mid-dleware framework that deploys and configures applicationsin DRE systems and enforces their network QoS require-ments using the four-stage (i.e., design-, pre-deployment-, deployment-, and runtime) approach shown in Figure 1.The innovative elements of NetQoPE’s four-stage architec-

Figure 1: NetQoPE’s Four-stage Architecture

ture include the following:• The Network QoS Specification Language

(NetQoS), which is a domain-specific modeling lan-guage (DSML) that supports design-time specification ofper-flow network QoS requirements, such as bandwidthand delay across a flow. By allowing application devel-opers to focus on functionality—rather than the differentdeployment contexts (e.g., different bandwidth and delayrequirements) where they will be used—NetQoS simplifiesthe deployment of applications in contexts that havedifferent network QoS needs, e.g., different bandwidthrequirements.

• The Network Resource Allocation Framework (Ne-tRAF), which is a middleware-based resource allocatorframework that uses the network QoS requirements cap-tured by NetQoS as input at pre-deployment time to helpguide QoS provisioning requests on the underlying net-work QoS mechanism at deployment time. By providingapplication-transparent, per-flow resource allocation capa-bilities at pre-deployment-time, NetRAF minimizes runtimeoverhead and simplifies validation decisions, such as admis-sion control.

• The Network QoS Configurator (NetCON), whichis a middleware-based network QoS configurator that pro-vides deployment-time configuration of component mid-dleware containers. NetCON adds flow-specific identifiers(e.g., DSCPs) to IP packets at runtime when applications in-voke remote operations. By providing container-mediated

and application-transparent capabilities to enforce runtimenetwork QoS, NetCON allows DRE systems to leverage theQoS services of configured routers without modifying ap-plication source code.

As shown in the Figure 1, the output of each stage inNetQoPE serves as input for the next stage, which helpsautomate the deployment and configuration of DRE appli-cations with network QoS support.

Paper organization. The remainder of the paper is or-ganized as follows: Section 2 describes a case study thatmotivates common requirements associated with provision-ing network QoS for DRE systems; Section 3 explains howNetQoPE addresses those requirements via its model-drivencomponent middleware framework; Section 4 evaluates thecapabilities provided by NetQoPE; Section 5 compares ourwork on NetQoPE with related research; and Section 6presents concluding remarks and lessons learned.

2 Motivating NetQoPE’s Network QoS Pro-visioning Capabilities

Figure 2 shows a representative DRE system in an of-fice enterprise security and hazard sensing environment,which we use as a case study to demonstrate and evaluateNetQoPE’s model-driven, middleware-guided network QoSprovisioning capabilities. Enterprises often transport net-

Figure 2: Network Configuration in an Enterprise Securityand Hazard Sensing Environment

work traffic using an IP network over high-speed Ethernet.Network traffic in an enterprise can be grouped into sev-eral classes, including (1) e-mail, videoconferencing, andnormal business traffic, and (2) sensory and imagery traf-fic of the safety/security hardware (such as fire/smoke sen-sors) installed on office premises. Our case study makesthe common assumption that safety/security traffic is morecritical than other traffic, and thus focuses on model-driven,middleware-guided mechanisms to assure the specified QoSfor this type of traffic in the presence of other traffic thatshares the same network.

114114114

As shown in Figure 2, our case study uses softwarecontrollers to manage hardware devices, such as sensorsand monitors. Each sensor/camera software controller fil-ters the sensory/imagery information and relays them to themonitor software controllers that display the information.An office enterprise could have lot of sensors and mon-itors deployed, and each of these communications couldhave a variety of network QoS requirements. These soft-ware controllers were developed using Lightweight CCM(LwCCM) [14] and the traffic between these software con-trollers uses a bandwidth broker [3] to manage networkresources via DiffServ network QoS mechanisms. Al-though the case study in this paper focuses on DiffServand LwCCM, NetQoPE is designed for use with other net-work QoS mechanisms (e.g., IntServ) and component mid-dleware technologies (e.g., J2EE).

Component-based applications in our case study usebandwidth broker services via the following middleware-guided steps: (1) network QoS requirements are specifiedon each application flow, along with information on thesource and destination IP and port addresses, (2) the band-width broker is invoked to reserve network resources alongthe network paths for each application flow, configure thecorresponding network routers, and obtain per-flow DSCPvalues to help enforce network QoS, and (3) remote oper-ations are invoked with appropriate DSCP values added tothe IP packets so that configured routers can provide per-flow differentiated performance. Section 3 describes thechallenges we encountered when implementing these stepsin the context of our case study and shows how NetQoPE’sfour-stage architecture shown in Figure 1 resolves thesechallenges.

3 NetQoPE’s Multistage Network QoS Provi-sioning Architecture

As discussed in Section 1, conventional techniques forproviding network QoS to applications incur several keylimitations, including modifying application source code to(1) specify deployment context-specific network QoS re-quirements, and (2) integrate functionality from networkQoS mechanisms at runtime. This section describes howNetQoPE addresses these limitations via its model-driven,middleware-guided network QoS provisioning architecture.

3.1 Challenge 1: Alleviating Complexities in QoS Re-quirements Specification

Context. Each application flow in a DRE system canspecify a required level of service (e.g., high priority vs. lowpriority), the source and destination IP and port addresses,and bandwidth and delay requirements. This information isused to allocate and configure network resources to providethe required QoS.Problem. Network QoS requirements (such as the band-

width and delay requirements mentioned above) can changedepending on a deployed context. For example, in our casestudy from Section 2, multiple fire sensors are deployed atdifferent importance levels and each sensor sends its sen-sory information to its corresponding monitors. Fire sen-sors deployed in the parking lot have a lower importancethan those in the server room. The sensor-monitor flowsthus have different network QoS requirements, even thoughthe reusable software controllers managing the fire sensorand the monitor have the same functionality.

Conventional techniques, such as hard-coded API ap-proaches [4], require application source code modificationsfor each context. Writing this code manually to specify net-work QoS requirements is tedious, error-prone, and non-scalable. In particular, it is hard to envision at developmenttime all the contexts in which the source code will be de-ployed.

Sidebar 1: Overview of LightweightCORBA Component Model (LwCCM)

Application functionality in LwCCM is provided throughcomponents which collaborate with other components viaports to create component assemblies. Assemblies inLwCCM are described using XML descriptors (mainly thedeployment plan descriptor) defined by the OMG D&C [15]specification. The deployment plan includes details aboutthe components, their implementations, and their connec-tions with other components. The deployment plan alsohas a placeholder configProperty that is associated with ele-ments (e.g., components, connections) to specify their prop-erties (e.g., priorities) and resource requirements. Compo-nents are hosted in containers, which provide the appropri-ate runtime operating environment (e.g., transactions sup-port) for components to invoke remote operations.

Solution approach → Model-driven visual network re-quirements specification. NetQoPE provides a DSMLcalled the Network QoS Specification Language (NetQoS).DRE system developers can use NetQoS to (1) model com-ponent assemblies, (2) assign components to target nodes,and (3) declaratively specify the following deploymentcontext-specific network QoS requirements on the modeledapplication flows: (a) network QoS classes, such as HIGH

PRIORITY (HP), HIGH RELIABILITY (HR), MULTIMEDIA

(MM), and BEST EFFORT (BE), (b) bi-directional bandwidthand delay requirements, and (c) selection of transport pro-tocol.

In the context of our case study, NetQoS’s network QoSclasses correspond to the DiffServ levels of service providedby our Bandwidth Broker [3].1 For example, the HP class

1NetQoS’s DSML capabilities can be extended to provide requirementsspecification conforming to a different network QoS mechanism, such asIntServ.

115115115

represents the highest importance and lowest latency traf-fic (e.g., fire detection reporting in the server room). TheHR class represents traffic with low drop rate (e.g., surveil-lance data). NetQoS also supports the MM class for sendingmultimedia data and the BE class for sending traffic with noQoS requirements.

After a model has been created, NetQoS’s model in-terpreter traverses the modeled application structure andgenerates a deployment plan (described in Sidebar 1).NetQoS’s model interpreter also traverses each modeled ap-plication flow and augments the deployment plan config-Property tags (also described in Sidebar 1) to express net-work QoS requirement annotations on the component con-nections. Section 3.2 describes how NetQoS allocates net-work resources based on requirements specified in the de-ployment plan descriptor.

Our case study has certain application flows (e.g., amonitor requesting location coordinates from a fire sen-sor) where the client needs to control the network priori-ties at which the requests and replies are sent. If a networkQoS provisioning engine provides this capability for appli-cations, real-time actions can be controlled irrespective ofnetwork congestion. There are other examples (e.g., a tem-perature sensor sends temperature sensory information tothe monitors), where the servers (in this case the monitors)could control how to receive and act on client requests.

To support these two models, NetQoS can assignthe following priority attributes to connections: (1) theCLIENT_PROPAGATED network priority model that allowsthe clients to dictate the bi-directional priorities, and (2) theSERVER_DECLARED network priority model that allowsservers to dictate the bi-directional priorities. NetQoS’smodel interpreter updates the deployment plan with thesepriority models for each flow. Section 3.3 explains howNetQoPE’s runtime mechanisms honor these priority mod-els when applications invoke remote operations.Application to the case study. Figure 3 shows a NetQoSmodel that highlights many of its key capabilities. Multiple

Figure 3: NetQoS Capabilities

instances of the same reusable application components (e.g.,FireSensorParking and FireSensorServer components) canbe annotated with different QoS attributes using an intu-itive drag and drop technique. This method of specify-ing QoS requirements is thus much simpler than modifyingapplication code for each deployment context, as demon-strated in Section 4.2. Moreover, the same QoS attribute(e.g., HR_1000 in Figure 3) can be reused across multipleconnections. NetQoS thus increases the scalability of ex-pressing requirements for large numbers of connections thatare prevalent in large-scale DRE systems, such as our casestudy.

3.2 Challenge 2: Alleviating Complexities in NetworkResource Allocation and Configuration

Context. DRE systems must allocate and configure net-work resources based on the QoS requirements specified ontheir application flows so that network QoS assurance canbe provided at runtime.Problem. In our case study, the temperature sensory in-formation from the server room is more important than theinformation from a conference room. It is undesirable, how-ever, to modify the temperature sensor software controllercode to directly interact with a middleware API or net-work QoS mechanism API since certain deployment con-texts (such as the deployment in a conference room) mightnot require network QoS assurances. Moreover, if applica-tion source code is modified to provide resource allocations,decisions on whether to allocate resources or not cannot bedetermined until the applications are deployed and opera-tional. This approach forces DRE system deployers to stopapplication components and deploy them on different nodesif required resources cannot be allocated across source anddestination nodes.Solution approach → Middleware-based Resource Al-locator Framework. NetQoPE’s Network Resource Allo-cator Framework (NetRAF) is a resource allocator enginethat allocates network resources for DRE systems using avariety of network QoS mechanisms, such as DiffServ andIntServ. As shown in Figure 4, the NetQoS DSML de-scribed in Section 3.1 captures the modeled per-flow net-work QoS requirements in the form of a deployment planthat is input to NetRAF.

The modeled deployment context could have many in-stances of the same reusable source code, e.g., the temper-ature sensor software controller could be instantiated twotimes: one for the server room and one for the conferenceroom. When using NetQoS, however, application develop-ers annotate only the connection between the instance atthe server room and the monitor software controller. SinceNetRAF operates on the deployment plan that capturesthis modeling effort, network QoS mechanisms are usedonly for the connection on which QoS attributes are added.

116116116

Figure 4: NetRAF’s Network Resource Allocation Capabil-ities

NetRAF thus improves conventional approaches [18] thatmodify application source code to work with network QoSmechanisms, which can become complex when source codeis reused in a wide range of deployment contexts.

NetRAF’s Network Resource Allocator Manager acceptsapplication QoS requests at pre-deployment-time and de-termines the network QoS mechanism (e.g., DiffServ orIntServ) to use to serve the requests. As shown in Figure 4,NetRAF’s Network Resource Allocator Manager workswith QoS mechanism-specific allocators (e.g., DiffServ Al-locator), which shields it from interacting directly withcomplex APIs for network QoS mechanisms (e.g., DiffServBandwidth Broker), thereby enhancing NetQoPE’s flexibil-ity.

Multiple allocators (e.g., IntServ Allocator and DiffServAllocator) can be used by NetRAF’s Network ResourceAllocator Manager to serve the needs of small-scale de-ployments (where IntServ and DiffServ are both suitable)and large-scale deployments (where DiffServ often pro-vides better scalability). For example, the shaded cloudconnected to the Network Resource Allocator Manager inFigure 4 shows how NetRAF can be extended to work withother network QoS mechanisms, such as IntServ.Application to the case study. Since our case study isbased on DiffServ, NetRAF uses the DiffServ Allocatorto allocate network resources. This allocator invokes theBandwidth Broker’s admission control capabilities [3] byfeeding it one application flow at a time. If all flows cannotbe admitted, NetRAF allows developers an option to mod-ify the deployment context since applications have not yetbeen deployed. Example modifications include changingcomponent implementations to consume fewer resources orchange the source and destination nodes. As demonstratedin Section 4.2, this capability helps NetRAF incur loweroverhead than conventional approaches [22, 18] that per-form validation decisions when applications are deployedand operated at runtime.

NetRAF’s DiffServ Allocator instructs the Bandwidth

Broker to reserve bi-directional resources in the specifiednetwork QoS classes as described in Section 3.1. The Band-width Broker determines the bi-directional DSCPs and Ne-tRAF encodes those values as connection attributes in thedeployment plan. In addition, the Bandwidth Broker usesits Flow Provisioner [3] to configure the routers to provideappropriate per-hop behavior when they receive IP pack-ets with the specified DSCP values. Section 3.3 describeshow component containers are auto-configured to add theseDSCPs when applications invoke remote operations.

3.3 Challenge 3: Alleviating Complexities in NetworkQoS Settings Configuration

Context. After network resources are allocated and net-work routers are configured, applications in DRE systemsneed to invoke remote operations using the chosen networkQoS settings (e.g., DSCP markings) so that the networklayer can differentiate application traffic and provision ap-propriate QoS to each flow.Problem. Application developers have historically writtencode that instructs the middleware to provide the appro-priate runtime services, e.g., DSCP markings in IP pack-ets [16]. For example, fire sensors in our case study fromSection 2 can be deployed in different QoS contexts thatare managed by reusable software controllers. Modifyingapplication code to instruct the middleware to add networkQoS settings is tedious, error-prone, and non-scalable be-cause (1) the same application code could be used in differ-ent contexts requiring different network QoS settings and(2) application developers might not (and ideally shouldnot) know the different QoS contexts in which the applica-tions are used during the development process. Application-transparent mechanisms are therefore needed to configurethe middleware to add these network QoS settings depend-ing on the application deployment context.Solution approach → Deployment and runtime com-ponent middleware mechanisms. Sidebar 1 describeshow LwCCM containers provide a runtime environmentfor components. NetQoPE’s Network QoS Configurator(NetCON) can auto-configure these containers by addingDSCPs to IP packets when applications invoke remote op-erations. As shown in Figure 5, NetRAF performs networkresource allocations, determines the bi-directional DSCPvalues to use for each application flow and encodes thoseDSCP values in the deployment plan.

During deployment, NetCON parses the deploymentplan and its connection tags to determine (1) source anddestination components, (2) the network priority model touse for their communication, (3) the bi-directional DSCPvalues, and (4) the target nodes on which the componentsare deployed. NetCON deploys the components on theirrespective containers and creates the associated object ref-erences for use by clients in a remote invocation. When a

117117117

Figure 5: NetCON’s Container Auto-configurations

component invokes a remote operation in LwCCM, its con-tainer’s context information provides the object reference ofthe destination component. Other component middlewareprovides similar capabilities via containers, e.g., EJB appli-cations interact with containers to obtain the right runtimeoperating environment.

NetCON’s container programming model can trans-parently add DSCPs and enforce the network prior-ity models described in Section 3.1. To support theSERVER_DECLARED network priority model, NetCON en-codes a SERVER_DECLARED policy and the associatedrequest/reply DSCPs on the server’s object reference. Whena client invokes a remote operation with this object refer-ence, the client-side middleware checks the policy on theobject reference, decodes the request DSCP, and includes itin the request IP packets. In the server-side middleware, be-fore sending the reply, the policy is checked again, and thereply DSCP is added on the IP packets.

To support the CLIENT_PROPAGATED network prior-ity model, NetCON configures the containers to applya CLIENT_PROPAGATED policy at the point of bind-ing an object reference with the client. In contrast tothe SERVER_DECLARED policy, the CLIENT_PROPAGATED

policy can be changed at runtime and different clients canaccess the servers with different network priorities. Whenthe source component invokes a remote operation using thepolicy-applied object reference, NetCON adds the associ-ated forward and reverse DSCP markings on the IP packets,thereby providing network QoS to the application flow. Acontainer can therefore transparently add both forward andreverse DSCP values when components invoke remote op-erations using the container services.Application to the case study. NetCON allows DRE sys-tem developers to focus on their application business logic,rather than wrestling with low-level mechanisms for provi-sioning network QoS. Moreover, NetCON provides thesecapabilities without modifying application code, therebysimplifying development and avoiding runtime overhead.

4 Evaluating NetQoPEThis section evaluates the flexibility of using NetQoPE

to provide network QoS assurance to end-to-end application

flows and demonstrates how NetQoPE’s network QoS pro-visioning capabilities significantly reduce application de-velopment effort incurred by conventional approaches.

4.1 Evaluation Scenario

Figure 6 shows key component interactions in the mod-ern office enterprise case study shown in Figure 2 that mo-tivated the design of these evaluations using NetQoPE. Ourscenario consists of software components (e.g., a fire sensorcontroller, monitor controller) developed using the CIAOmiddleware, which is an open-source LwCCM implemen-tation developed on top of TAO real-time CORBA ObjectRequest Broker (ORB). Figure 6 also shows the underly-ing network topology with Diffserv enabled routers (e.g.,P, Q), which we configure using an associated BandwidthBroker [3] software hosted on C.

Figure 6: Experimental SetupIn our evaluation scenario, a number of sensory and im-

agery software controllers sent their monitored informa-tion to monitor controllers so that appropriate control ac-tions could be performed by enterprise supervisors moni-toring abnormal events. For example, Figure 6 shows twofire sensor controller components deployed on hosts A andB. These components sent their monitored information tomonitor controller components deployed on hosts D andF. Communication between these software controllers usedone of the traffic classes (e.g., HIGH PRIORITY (HP)) de-fined in Section 3.1 with the following capacities on alllinks: HP = 20 Mbps, HR = 30 Mbps, and MM = 30 Mbps.The BE class used the remaining available bandwidth in thenetwork. The goal of our experiments was to evaluate theflexibility of using NetQoPE to provide network QoS assur-ance to end-to-end application flows, such as the applicationflow between the fire sensor controller component on hostA and the monitor controller component on host D.

4.2 Evaluating NetQoPE’s Model-driven QoS Provi-sioning Capabilities

Rationale. As discussed in Section 3, NetQoPE is designedto provide network QoS to applications in an extensiblemanner. This experiment evaluates NetQoPE’s application-

118118118

transparent network QoS provisioning capabilities.Methodology. We first identified four flows from Figure 6whose network QoS requirements are described as follows:(1) a fire sensor controller component on host A uses thehigh reliability (HR) class to send potential fire alarms inthe parking lot to the monitor controller component on hostD, (2) a fire sensor controller component on host B uses thehigh priority (HP) class to send potential fire alarms in theserver room to the monitor controller component on hostF, (3) a camera controller component on host E uses themultimedia (MM) class and sends imagery information ofthe break room to the monitor controller component on hostG, and (4) a temperature sensor controller component onhost A uses the best effort (BE) class and sends tempera-ture readings to the monitor controller component on hostF. The clients dictated the network priority for the requestsand replies in all flows except for the temperature sensorand monitor controller component flow, where the serverdictated the priority. We used TCP as the transport pro-tocol and 20 Mbps of forward and reverse bandwidth wasrequested for each type of network QoS traffic.

To compare NetQoPE’s methodology of provisioningnetwork QoS for these flows with other existing solutions,we also define a taxonomy for evaluating technologies thatprovide network QoS assurances to end-to-end DRE ap-plication flows. Conventional approaches can be classi-fied as being (1) object-oriented [8, 18, 22, 16], (2) aspect-oriented [7], and (3) component middleware-based [4, 19].We now describe how each approach provides the followingfunctionality needed to leverage network QoS mechanismcapabilities:

• Requirements specification. In conventional ap-proaches applications use (1) middleware-based APIs [8,22], (2) contract definition languages [18, 16], (3) run-time aspects [7], or (4) specialized component middle-ware container interfaces [4] to specify network QoS re-quirements. Whenever the deployment context and the as-sociated QoS requirements change, however, applicationsource code must also change, thereby limiting reusabil-ity. In contrast, as described in Section 3.1, NetQoS pro-vides domain-specific, declarative techniques that increasereusability across different deployment contexts and allevi-ate the need to programmatically specify QoS requirements.

• Network resource allocation. Conventional ap-proaches require the deployment of applications before theirper-flow network resource requirements can be provisionedby network QoS mechanisms. If the required resources can-not be allocated for those applications they must be stopped,their source code must be modified to specify new resourcerequirements, and the resource reservation process must berestarted. This approach is tedious since it involves de-ploying and re-deploying applications multiple times (po-tentially on different nodes). In contrast, NetRAF handles

deployment changes through NetQoS models, as describedin Section 3.2. This process occurs during pre-deploymentbefore applications have been deployed, which reduces theeffort needed to change deployment topology or applicationQoS requirements.

• Network QoS enforcement. Conventional approachesmodify application source code [16] or programmingmodel [4] to instruct the middleware to enforce runtimeQoS for their remote invocations. Applications must there-fore be designed to handle two different usecases—to en-force QoS and when no QoS is required—thereby limitingapplication reusability. In contrast, as described in Sec-tion 3.3, NetCON uses a container programming model thattransparently enforces runtime QoS for applications withoutchanging their source code or programming model.

Using the conventional approaches and the NetQoPE ap-proach, we now compare the manual effort required to pro-vide network QoS to the 4 end-to-end application flows de-scribed above. We decompose the manual effort across thefollowing general steps: (1) implementation, where soft-ware developers write code, (2) deployment, where systemdeployers map (or stop) application components on theirtarget nodes, and (3) modeling tool use, where applica-tion developers use NetQoPE to model a DRE applicationstructure and specify per-flow QoS requirements. In ourevaluation, a complete QoS provisioning lifecycle consistsof specifying requirements, allocating resources, deployingapplications, and stopping applications when they are fin-ished.

To compare NetQoPE with manual efforts, we deviseda realistic scenario for the 4 end-to-end application flowsdescribed above. In this scenario, three sets of experimentswere conducted with the following deployment variants:

• In the first variant, all 4 end-to-end application flowswere configured with the network QoS requirements as de-scribed above.

• In the second variant, to demonstrate the effect ofchanges in QoS requirements on manual efforts we modi-fied the bandwidth requirements from 20 Mbps to 12 Mbpsfor each end-to-end flow.

• In the third variant, we demonstrate the effect ofchanges in QoS requirements and resource (re)reservationstaken together on manual efforts. We modified bandwidthrequirements of all flows from 12 Mbps to 16 Mbps. Wealso changed temperature sensor controller component touse the high reliability (HR) class instead of the best effortBE class. Finally, we increased the background HR classtraffic across the hosts so that the resource reservation re-quest for the flow between temperature sensor and monitorcontroller components fails. In response, deployment con-texts (e.g., bandwidth requirements, source and destinationnodes) were changed and resource re-reservation was per-formed.

119119119

For the first deployment, the manual effort required us-ing conventional approaches involved 10 steps: (1) mod-ify source code for each of the 4 components to specifytheir QoS requirements (8 implementation steps), (2) de-ploy all components (1 deployment step), and (3) shutdownall components (1 deployment step). Conversely, the effortrequired using NetQoPE involved the following 4 steps: (1)model the DRE application structure of all 4 end-to-end ap-plication flows using NetQoS (1 modeling step), (2) anno-tate QoS specifications on each end-to-end application flow(1 modeling step), (3) deploy all components (1 deploy-ment step), and (4) shutdown all components (1 deploymentstep).

For the second deployment, the effort required usinga conventional approach is also 10 steps since sourcecode modifications are needed as the deployment contextschanged (in this case, the bandwidth requirements changedacross 4 different deployment contexts). In contrast, the ef-fort required using NetQoPE involves 3 steps: (1) anno-tate QoS specifications on each end-to-end application flow(1 modeling step), (2) deploy all components (1 deploy-ment step), and (3) shutdown all components (1 deploymentstep). Application developers also reused NetQoS’s appli-cation structure model created for the initial deployment,which helped reduce the required efforts by a step.

For the third deployment, the effort required using a con-ventional approach is 13 steps: (1) modify source code ofeach of the 8 components to specify their QoS requirements(8 implementation steps), (2) deploy all components (1 de-ployment step), (3) shutdown the temperature sensor com-ponent (1 deployment step – resource allocation failed forthe component), (4) modify source code of temperature sen-sor component back to use BE network QoS class (deploy-ment context change) (1 implementation step), (5) redeploythe temperature sensor component (1 deployment step), and(6) shutdown all components (1 deployment step).

In contrast, the effort required using NetQoPE for thethird deployment is 4 steps: (1) annotate QoS specificationson each end-to-end application flow (1 modeling step), (2)begin deployment of all the components, but NetRAF’s pre-deployment-time allocation capabilities determined the re-source allocation failure and prompted the NetQoPE appli-cation developer to change the QoS requirements (1 pre-deployment step), (3) re-annotate QoS requirements for thetemperature sensor component flow (1 modeling step) (4)deploy all components (1 deployment step), and (5) shut-down all components (1 deployment step).

Table 1 summarizes the step-by-step analysis describedabove. These results show that conventional approachesincur roughly an order of magnitude more effort thanNetQoPE to provide network QoS assurance for end-to-end application flows. Closer examination shows that inconventional approaches, application developers spend sub-

stantially more effort developing software that can workacross different deployment contexts. Moreover, this pro-cess must be repeated when deployment contexts and theirassociated QoS requirements change. Additionally, imple-mentations are complex since the requirements are speci-fied using middleware [22] and/or network QoS mechanismAPIs [12].

Further, application (re)deployments are required when-ever reservation requests fail. In this experiment, only oneflow required re-reservation and that incurred additional ef-fort of 3 steps. If there are large number of flows—and en-terprise DRE systems like our case study often have dozensor hundreds of flows—the amount of effort required is sig-nificantly more than for conventional approaches.

Approaches # Steps in Experiment VariantsFirst Second Third

NetQoPE 4 3 5Conventional 10 10 13

Table 1: Comparison of Manual Efforts Incurred in Con-ventional and NetQoPE Approaches

In contrast, NetQoPE’s “write once, deploy multipletimes for different QoS” capabilities increase deploymentflexibility and extensibility for environments where manyreusable software components are deployed. To provide thisflexibility, NetQoS generates XML-based deployment de-scriptors that capture context-specific QoS requirements ofapplications. For our experiment, communication betweenfire sensor and monitor controllers was deployed in multi-ple deployment contexts, i.e., with bandwidth reservationsof 20 Mbps, 12 Mbps, and 16 Mbps. In DRE systems likeour case study, however, the same communication patternsbetween components could occur in many deployment con-texts.

For example, the same communication patterns coulduse any of the four network QoS classes (HP, HR, MM, andBE). The communication patterns that use the same networkQoS class could make different forward and reverse band-width reservations (e.g., 4, 8, or 10 Mbps). As shown inTable 2, NetQoS auto-generates as much as 1,325 lines ofXML code for these scenarios, which would otherwise behandcrafted by application developers.

Deployment contextsNumber of communications2 5 10 20

1 23 50 95 1855 47 110 215 425

10 77 185 365 72520 137 335 665 1325

Table 2: Generated Lines of XML Code

These results demonstrate that NetQoPE’s network QoS

120120120

provisioning capabilities significantly reduce applicationdevelopment effort incurred by conventional approachesand provides increased flexibility in deploying and provi-sioning multiple application end-to-end flows under multi-ple deployment and network QoS contexts.

5 Related WorkThis section compares our R&D activities on NetQoPE

with related work on middleware-based QoS managementand model-based design tools.

Network QoS management in middleware. Priorwork on integrating network QoS mechanisms with mid-dleware [22, 18, 16, 8] focused on providing middle-ware APIs to shield applications from directly interactingwith complex network QoS mechanism APIs. Middlewareframeworks transparently converted the specified applica-tion QoS requirements into lower-level network QoS mech-anism APIs and provided network QoS assurances. Theseapproaches, however, modified applications to dictate QoSbehavior for the various flows. NetQoPE differs from theseapproaches by providing application-transparent and au-tomated solutions to leverage network QoS mechanisms,thereby significantly reducing manual design and develop-ment effort to obtain network QoS.

QoS management in middleware. Prior research hasfocused on adding various types of QoS capabilities to mid-dleware. For example, [11] describes J2EE container re-source management mechanisms that provide CPU avail-ability assurances to applications. Likewise, 2K [24] pro-vides QoS to applications from varied domains using acomponent-based runtime middleware. In addition, [4] ex-tends EJB containers to integrate QoS features by provid-ing negotiation interfaces which the application developersneed to implement to receive desired QoS support. Syn-ergy [17] describes a distributed stream processing middle-ware that provides QoS to data streams in real time by ef-ficient reuse of data streams and processing components.These approaches are restricted to CPU QoS assurances orapplication-level adaptations to resource-constrained sce-narios. NetQoPE differs by providing network QoS assur-ances in a application-agnostic fashion.

Deployment-time resource allocation. Prior work hasfocused on deploying applications at appropriate nodes sothat their QoS requirements can be met. For example, priorwork [13, 21] has studied and analyzed application commu-nication and access patterns to determine collocated place-ments of heavily communicating components. Other re-search [6, 9] has focused on intelligent component place-ment algorithms that maps components to nodes while satis-fying their CPU requirements. NetQoPE differs from theseapproaches by leveraging network QoS mechanisms to allo-cate network resources at pre-deployment-time and enforc-ing network QoS at runtime.

Model-based design tools. Prior work has been

done on model-based design tools. PICML [1] enablesDRE system developers to define component interfaces,their implementations, and assemblies, facilitating deploy-ment of LwCCM-based applications. VEST [20] andAIRES [10] analyze domain-specific models of embeddedreal-time systems to perform schedulability analysis andprovides automated allocation of components to processors.SysWeaver [5] supports design-time timing behavior ver-ification of real-time systems and automatic code genera-tion and weaving for multiple target platforms. In contrast,NetQoPE provides model-driven capabilities to specify net-work QoS requirements on DRE system application flows,and subsequently allocate network resources automaticallyusing network QoS mechanisms. NetQoPE thus helps as-sure that application network QoS requirements are met atdeployment-time, rather than design-time or runtime.

6 Concluding Remarks

This paper describes the design and evaluation ofNetQoPE, which is a model-driven component middlewareframework that manages network QoS for applications inDRE systems. The lessons we learned developing NetQoPEand applying it to a representative DRE system case studythus far include:

• NetQoPE’s domain-specific modeling languages helpcapture per-deployment network QoS requirements of ap-plications so that network resources can be allocated ap-propriately. Application business logic consequently neednot be modified to specify deployment-specific QoS re-quirements, thereby increasing software reuse and flexibil-ity across a range of deployment contexts.

• Programming network QoS mechanisms directly in ap-plication code requires the deployment and running of ap-plications before they can determine if the required net-work resources are available to meet QoS needs. Con-versely, providing these capabilities via NetQoPE’s model-driven, middleware framework helps guide resource alloca-tion strategies before application deployment, thereby sim-plifying validation and adaptation decisions.

• NetQoPE’s model-driven deployment and configura-tion tools help transparently configure the underlying com-ponent middleware on behalf of applications to add context-specific network QoS settings. These settings can be en-forced by NetQoPE’s runtime middleware framework with-out modifying the programming model used by applica-tions. Applications therefore need not change how theycommunicate at runtime since network QoS settings can beadded transparently.

• NetQoPE’s strategy of allocating network resources toapplications before they are deployed may be too limitingfor certain types of DRE systems. In particular, applica-tions in open DRE systems [23] might not consume their re-source allotment at runtime, which may underutilize system

121121121

resources. We are therefore extending NetQoPE to overpro-vision resources for applications on the assumption that notall applications will use their allotment. If runtime resourcecontentions occur, we are also developing dynamic resourcemanagement strategies that can provide predictable networkperformance for mission-critical applications.

NetQoPE’s model-driven middleware platforms andtools, except the Bandwidth Broker used in the experi-ments, are available in open-source format from www.dre.vanderbilt.edu/cosmic, and along with the CIAO com-ponent middleware available at www.dre.vanderbilt.edu.

References[1] K. Balasubramanian, J. Balasubramanian, J. Parsons,

A. Gokhale, and D. C. Schmidt. A Platform-IndependentComponent Modeling Language for Distributed Real-timeand Embedded Systems. Journal of Computer Systems Sci-ence, 73(2):171–185, 2007.

[2] S. Blake, D. Black, M. Carlson, E. Davies, Z. Wang, andW. Weiss. An Architecture for Differentiated Services. In-ternet Society, Network Working Group RFC 2475, pages1–36, Dec. 1998.

[3] B. Dasarathy, S. Gadgil, R. Vaidhyanathan,K. Parmeswaran, B. Coan, M. Conarty, and V. Bhanot. Net-work QoS Assurance in a Multi-Layer Adaptive ResourceManagement Scheme for Mission-Critical Applicationsusing the CORBA Middleware Framework. In RTAS 2005,San Francisco, CA, Mar. 2005. IEEE.

[4] M. A. de Miguel. Integration of QoS Facilities into Compo-nent Container Architectures. In Proceedings of the FifthIEEE International Symposium on Object-Oriented Real-Time Distributed Computing (ISORC 2002), 2002.

[5] D. de Niz, G. Bhatia, and R. Rajkumar. Model-BasedDevelopment of Embedded Systems: The SysWeaver Ap-proach. In Proc. of RTAS’06, pages 231–242, Washington,DC, USA, August 2006.

[6] D. de Niz and R. Rajkumar. Partitioning Bin-Packing Al-gorithms for Distributed Real-time Systems. InternationalJournal of Embedded Systems, 2005.

[7] G. Duzan, J. Loyall, R. Schantz, R. Shapiro, and J. Zinky.Building Adaptive Distributed Applications with Middle-ware and Aspects. In Proc. of AOSD ’04, pages 66–73, NewYork, NY, USA, 2004.

[8] M. A. El-Gendy, A. Bose, S.-T. Park, and K. G. Shin. Pavingthe First Mile for QoS-dependent Applications and Appli-ances. In Proc. of IWQOS’04, Montreal, Canada, June 2004.

[9] S. Gopalakrishnan and M. Caccamo. Task Partitioning withReplication upon Heterogeneous Multiprocessor Systems.In RTAS ’06, pages 199–207, San Jose, CA, USA, 2006.

[10] Z. Gu, S. Kodase, S. Wang, and K. G. Shin. A Model-Based Approach to System-Level Dependency and Real-time Analysis of Embedded Software. In RTAS’03, pages78–85, Washington, DC, May 2003. IEEE.

[11] M. Jordan, G. Czajkowski, K. Kouklinski, and G. Skin-ner. Extending a J2EE Server with Dynamic and Flex-ible Resource Management. In Proceedings of the

ACM/IFIP/USENIX International Middleware Conference,Toronto, Canada, 2004.

[12] L. Zhang and S. Berson and S. Herzog and S. Jamin. Re-source ReSerVation Protocol (RSVP) Version 1 FunctionalSpecification. Network Working Group RFC 2205, pages1–112, Sept. 1997.

[13] D. Llambiri, A. Totok, and V. Karamcheti. Efficiently Dis-tributing Component-Based Applications Across Wide-AreaEnvironments. In Proc. of ICDCS’03, 2003.

[14] Object Management Group. Light Weight CORBAComponent Model Revised Submission, OMG Documentrealtime/03-05-05 edition, May 2003.

[15] OMG. Deployment and Configuration of Component-basedDistributed Applications, v4.0, Document formal/2006-04-02 edition, Apr. 2006.

[16] R. Schantz and J. Loyall and D. Schmidt and C. Rodriguesand Y. Krishnamurthy and I. Pyarali. Flexible and AdaptiveQoS Control for Distributed Real-time and Embedded Mid-dleware. In Proc. of Middleware’03, Rio de Janeiro, Brazil,June 2003. IFIP/ACM/USENIX.

[17] T. Repantis, X. Gu, and V. Kalogeraki. Synergy: Sharing-Aware Component Composition for Distributed Stream Pro-cessing Systems. In Proc. of Middleware 2006.

[18] R. Schantz, J. Zinky, D. Karr, D. Bakken, J. Megquier, andJ. Loyall. An Object-level Gateway Supporting Integrated-Property Quality of Service. ISORC, 00:223, 1999.

[19] P. Sharma, J. Loyall, G. Heineman, R. Schantz, R. Shapiro,and G. Duzan. Component-Based Dynamic QoS Adapta-tions in Distributed Real-time and Embedded Systems. InProceedings of the International Symposium on DistributedObjects and Applications (DOA), Agia Napa, Cyprus, Oct.2004.

[20] J. A. Stankovic, R. Zhu, R. Poornalingam, C. Lu, Z. Yu,M. Humphrey, and B. Ellis. Vest: An aspect-based com-position tool for real-time systems. In Proc. of RTAS’03,page 58, Washington, DC, USA, 2003.

[21] C. Stewart and K. Shen. Performance Modeling and SystemManagement for Multi-component Online Services. In Proc.of NSDI’05, Boston, MA, pages 71–84, May 2005.

[22] P. Wang, Y. Yemini, D. Florissi, and J. Zinky. A DistributedResource Controller for QoS Applications. In Proceedingsof the Network Operations and Management Symposium(NOMS 2000). IEEE/IFIP, Apr. 2000.

[23] X. Wang, D. Jia, C. Lu, and X. Koutsoukos. DEUCON: De-centralized End-to-End Utilization Control for DistributedReal-Time Systems. Parallel and Distributed Systems, IEEETransactions on, 18(7):996–1009, 2007.

[24] D. Wichadakul, K. Nahrstedt, X. Gu, and D. Xu. 2K: An In-tegrated Approach of QoS Compilation and Reconfigurable,Component-Based Run-Time Middleware for the UnifiedQoS Management Framework. In Proc. of Middleware’01,2001.

122122122