Software-Defined Service Networking: Performance Differentiation in Shared Multi-Tenant Cloud Applications

A composite cloud application can achieve economies of scale by sharing partner services between tenants at runtime, following the Single-Instance Multi-Tenancy (SIMT) model. However, supporting runtime sharing with tenant-specific functional and performance variations in an SIMT application is challenging. We address this challenge by proposing the Software-Defined Service Networking (SDSN) approach. SDSN realizes an SIMT composite cloud application with a managed service network and a set of managed virtual service networks (VSNs) that share the service network. The service network is formed by connecting a set of partner services according to their capabilities and interoperability. To achieve the functional and performance requirements of a particular tenant, a VSN composes a subset of the services under a specific configuration design over the service network, and regulates the interactions between the services concerned according to a specific regulation design. We describe how a service network and its VSNs are designed and enacted to achieve the SIMT objectives. We show the feasibility of SDSN, demonstrate the utilization benefits it achieves, and quantify the runtime overhead it incurs.

[1]  Alan Colman,et al.  Service Orchestration as Organization: Building Multi-Tenant Service Applications in the Cloud , 2014 .

[2]  Wouter Joosen,et al.  Building a Customizable Business-Process-as-a-Service Application with Current State-of-Practice , 2013, SC@STAF.

[3]  Frank Leymann,et al.  Service Networks Modelling: An SOA & BPM Standpoint , 2010, J. Univers. Comput. Sci..

[4]  Jaap Gordijn,et al.  Service Network Approaches , 2012, Handbook of Service Description.

[5]  Frank Leymann,et al.  Defining Composite Configurable SaaS Application Packages Using SCA, Variability Descriptors and Multi-tenancy Patterns , 2008, 2008 Third International Conference on Internet and Web Applications and Services.

[6]  Claus Pahl,et al.  Process as a Service Distributed Multi-tenant Policy-Based Process Runtime Governance , 2010, 2010 IEEE International Conference on Services Computing.

[7]  Wouter Joosen,et al.  Context-oriented programming for customizable SaaS applications , 2012, SAC '12.

[8]  Hai Jin,et al.  QoS-Driven Service Selection for Multi-tenant SaaS , 2012, 2012 IEEE Fifth International Conference on Cloud Computing.

[9]  V. Allee.  Value network analysis and value conversion of tangible and intangible assets , 2008 .

[10]  Jun Han,et al.  Using role-based coordination to achieve software adaptability , 2007, Sci. Comput. Program..

[11]  Anne H. H. Ngu,et al.  QoS-aware middleware for Web services composition , 2004, IEEE Transactions on Software Engineering.

[12]  Jeff Magee,et al.  Dynamic Configuration for Distributed Systems , 1985, IEEE Transactions on Software Engineering.

[13]  Martin F. Arlitt,et al.  Web server workload characterization: the search for invariants , 1996, SIGMETRICS '96.

[14]  Wouter Joosen,et al.  A Middleware Layer for Flexible and Cost-Efficient Multi-tenant Applications , 2011, Middleware.

[15]  Sven Apel,et al.  Granularity in software product lines , 2008, 2008 ACM/IEEE 30th International Conference on Software Engineering.

[16]  M. Brian Blake,et al.  Composition as a service [Web-Scale Workflow] , 2010, IEEE Internet Computing.

[17]  Jun Han,et al.  Software-Defined Service Networking: Runtime Sharing with Performance Differentiation in Multi-tenant SaaS Applications , 2015, 2015 IEEE International Conference on Services Computing.

[18]  Rouven Krebs,et al.  Metrics and techniques for quantifying performance isolation in cloud environments , 2012, QoSA '12.

[19]  Nick McKeown,et al.  OpenFlow: enabling innovation in campus networks , 2008, CCRV.

[20]  Srinath Perera,et al.  A Scalable Multi-Tenant Architecture for Business Process Executions , 2012, Int. J. Web Serv. Res..

[21]  Timothy W. Finin,et al.  Modeling conversation policies using permissions and obligations , 2004, Autonomous Agents and Multi-Agent Systems.

[22]  Manuel Loesch,et al.  Platform-as-a-Service Architecture for Performance Isolated Multi-tenant Applications , 2014, 2014 IEEE 7th International Conference on Cloud Computing.

[23]  Yanbo Han,et al.  Feedback-Control-Based Performance Regulation for Multi-Tenant Applications , 2009, 2009 15th International Conference on Parallel and Distributed Systems.

[24]  Filip De Turck,et al.  An open middleware for proactive QoS-aware service composition in a multi-tenant SaaS environment , 2012 .

[25]  Jun Han,et al.  Dynamic Performance Management in Multi-tenanted Business Process Servers Using Nonlinear Control , 2012, ICSOC.

[26]  Gordon S. Blair,et al.  Models@ run.time , 2009, Computer.

[27]  Ying Chen,et al.  SPIN: Service Performance Isolation Infrastructure in Multi-tenancy Environment , 2008, ICSOC.

[28]  Fernando M. V. Ramos,et al.  Software-Defined Networking: A Comprehensive Survey , 2014, Proceedings of the IEEE.

[29]  Xipeng Xiao,et al.  Internet QoS: a big picture , 1999, IEEE Netw..

[30]  Danny Weyns,et al.  Architecture-centric support for adaptive service collaborations , 2014, TSEM.

[31]  Paola Inverardi,et al.  Model-based performance prediction in software development: a survey , 2004, IEEE Transactions on Software Engineering.