Virtualised e-Learning with Real-Time Guarantees on the IRMOS Platform

885 views

Published on

In this paper we focus on how Quality of Service guarantees are provided to virtualised applications in the Cloud Computing infrastructure that is being developed in the context of the IRMOS European Project. Provisioning of proper timeliness guarantees to distributed real-time applications involves the careful use of real-time scheduling mechanisms at the virtual-machine hypervisor level, of QoS-aware networking protocols and of proper design methodologies and tools for stochastic modelling of the application. The paper focuses on how we applied these techniques to a case-study involving a real eLearning mobile content delivery application that has been integrated into the IRMOS platform and its achieved performance.

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
885
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Virtualised e-Learning with Real-Time Guarantees on the IRMOS Platform

  1. 1. IEEE 2010 December 13 – 15, 2010 Perth, AustraliaVirtualised e-Learning with Real-Time Guarantees on the IRMOS PlatformTommaso CucinottaReal-Time Systems LaboratoryScuola Superiore SantAnnaPisa, Italy… and 12 others from 6 institutions:
  2. 2. IntroductionTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  3. 3. Introduction  Towards a new computing paradigm  Computing, network, storage in the cloud  Not only batch computing and storage  but also interactive real-time applicationsTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  4. 4. Web and Clouds Performance Today  How much should I replicate my infrastructure  to meet desired average QoS levels ?  General-purpose technologies  What about the non-average cases and interactivity?Tommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  5. 5. The IRMOS Approach to real-time and stable QoSTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  6. 6. IRMOS  Focus: Interactive Real-time Multimedia on SOIs Application Scenarios SaaS Framework Services PaaS Intelligent Service-Oriented Networking Infrastructure IaaS (ISONI)Tommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  7. 7. IRMOS Two-Phases Approach Design Tools Benchmarking Application Concretion Discovery NegotiationModeling, Reservation Mechanisms for Mechanisms for Methodology for theAnalysis, Methodology for thePlanning precise allocationService precise allocationof of identification of identification of resources resources Instantiation resource requirements resource requirements to applications Service to applications Component Configuration Execution & Monitoring Cleanup Offline 7Tommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  8. 8. IRMOS Two-Phases Approach Design Tools Benchmarking Application Concretion Discovery NegotiationModeling, ReservationAnalysis,Planning Service Instantiation Service Component Configuration Execution & Monitoring Cleanup Offline 8Tommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  9. 9. The IRMOS ISONI Execution Environment  Infrastructure for deployment of  virtualized software components  with stable and precise QoS guarantees  Scheduling mechanisms for temporal isolation  Computing layer  IRMOS Real-Time Scheduler for Linux  Networking layer  QoS-aware protocols (DiffServ, IntServ, MPLS, …)  Storage layerTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  10. 10. The IRMOS ISONI Execution EnvironmentTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  11. 11. Deployment Problem See our work @ SOCA09  Deployment of VSNs on PNs  Given computing/network requirements  Respecting end-to-end timing constraints Physical Host Physical Host Computing Networking Requirements Requirements Physical Subnet Physical Link Physical Host Virtual Service Network Maximum response-time Physical HostTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  12. 12. Case-study: e-LearningTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  13. 13. Real-time e-Learning Synchronising real and virtual worlds Venue locations Access devices Interactive Gallery media and communication terminals Real time synchronization Mobiles Festivals Interactive Virtual Worldswhiteboards Learning Media generation Remote locations Professional Museums Social Networking Personal Classroom HomeTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  14. 14. Mobile e-Learning Architecture and modelTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  15. 15. Mobile e-Learning Architecture and modelTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  16. 16. Deploying e-Learning  Goal Scheduling params (budget, period)  Deploy the e-Learning server  With the application-level parameters specified in the SLA  Detail level (i.e., resolution)  Maximum number of users  Respecting the SLA QoS Physical Host  Statistics on the response-time of individual requests mean Resol. Resol. W x H x fps W x H x fps (mean, max, std dev) std-dev Users Users 10 10 maxTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  17. 17. Modelling e-Learning  Non-IRMOS world  Network performance highly dependent on traffic of other apps  Computing performance highly dependent on workload of other apps  When deployed in IRMOS/ISONI  QoS-aware networking and CPU real-time scheduling limit the interferences among different application instances  Applications can be analysed in isolationTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  18. 18. Modelling e-Learning  Network delays: Erlang distributions  Parameters fitted on benchmark data  Computing delays  Strong dependence on application-level parameters (number of users, resolution)  Black-box approach → Neural NetworksTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  19. 19. Benchmarking Data The real-time scheduler successfully isolates performance of 2 VMUsTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  20. 20. Benchmarking DataTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  21. 21. Neural Network Training  After training, the ANN successfully outputs the mean and standard deviation of the SC response time: prediction error less than 3%Tommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  22. 22. Conclusions and future workTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  23. 23. Conclusions  The IRMOS/ISONI virtualized infrastructure facilitates  benchmarking & modelling  off-line performance prediction  on-line performance stability  allowsfor better server consolidation levels while meeting the timing constraints  We showed the IRMOS way to deploy an e-Learning application with precise QoS guaranteesTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  24. 24. Future work (WiP, actually)  Apply the methodology to the VirtualWorld e-Learning platform  Apply the methodology to the other IRMOS application scenarios  Film post-production  Virtual reality in automotive  Model how the scheduler affects the QoS metrics, to reduce the number of configurations to benchmarkTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  25. 25. References T. Cucinotta, K. Konstanteli, T. Varvarigou, "Advance Reservations for Distributed Real-TimeWorkflows with Probabilistic Service Guarantees", IEEE International Conference on Service-Oriented Computing and Applications (SOCA 2009), December 2009, Taipei, Taiwan K. Kostanteli, D. Kyriazis, T. Varvarigou, T. Cucinotta, G. Anastasi, "Real-time guarantees in flexible advance reservations", 2nd IEEE International Workshop on Real-Time Service-Oriented Architecture and Applications (RTSOAA 2009), Seattle, Washington, July 2009 F. Checconi, T. Cucinotta, D. Faggioli, G. Lipari, "Hierarchical Multiprocessor CPU Reservations for the Linux Kernel", in 5th International Workshop on Operating Systems Platforms for Embedded Real-Time Applications (OSPERT 2009), Dublin, Ireland, June 2009 T. Cucinotta, G. Anastasi, L. Abeni, "Real-Time Virtual Machines", in 29th Real- Time System Symposium (RTSS 2008) -- Work in Progress Session, Barcelona, December 2008 YouTube Video on e-Learning performance isolation:  http://www.youtube.com/watch?v=8FbHZ4ytNoQ IRMOS YouTube channel:  http://www.youtube.com/user/irmosproject IRMOS Project Website: http://www.irmosproject.euTommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it
  26. 26. Thanks for your attention Questions ?Tommaso Cucinotta – Real-Time Systems Lab (RETIS), Pisa, Italy – cucinotta@sssup.it

×