ABSTRACT
SEDA (Staged Event-Driven Architecture) is a middleware architecture designed to support massive concurrency demands of internet services. However, managing the resources manually to achieve high performance in such a computing system has proved difficult, time-consuming, error-prone and non-QoS-guaranteed. In this paper, we propose an adaptive control approach to automatic resource management and performance control for SEDA-based applications. This approach is based on a combination of a load balancing strategy and feedback auto-tune stages for global optimal performance. In addition, our control algorithms are able to automatically optimize the control parameters at runtime. The design has been built into a SEDA-based web sever and validated by benchmarking this web server. The experimental results demonstrate that our auto-tune design is able to yield superior adaptation performance for SEDA applications in dynamic working environments, achieving desired performance targets with simple control algorithms and automatic parameter tuning.
- Beltran V., Carrera D., et al.: Evaluation the scalability of java event-driven web servers. Proc. of International Conference on Parallel Processing (ICPP'04), IEEE. (2004) Google ScholarDigital Library
- Benjamin C. Kuo and Golnaraghi F.: Automatic Control Systems (8th edition), Wiley, ISBN: 0471134767. (2002) Google ScholarDigital Library
- Chen H. and Mohapatra P.: Session-based overload control in QoS-aware Webservers. Proc. of IEEE INFOCOM2002, pages 516--524. (2002)Google Scholar
- Diao Y., Hellerstein J. L., et al.: Managing web server performance with autotune agents. IBM System journal Vol 42, No 1, pages 136--149. (2003) Google ScholarDigital Library
- Diao Y., Gandhi N., et al.: Using MIMO feedback control to enforce policies for interrelated metrics with application to the Apache Web server. Proc. of the Network Operations and Management Symposium, Florence, Italy. (2002)Google Scholar
- Diao Y., Hellerstein J. L., et al.: Optimizing quality of service using fuzzy control. Proc. of the 13th IFIP/IEEE International Workshop on Distributed Systems: Operations and Management. Springer-Verlag, pages 42--53. (2002) Google ScholarDigital Library
- Liu X., Lui Sha, et al.: Online Response Time Optimization of Apache Web Server. Proc. of the 11th International Workshop on Quality of Service (IWQoS 2003), pages 461--478. (2003) Google ScholarDigital Library
- Lu C., Abdelzaher T. F., et al.: A feedback control architecture and design methodology for service delay guarantees in web servers. Technical Report CS-2001-06, University of Virginia, Department of Computer Science. (2001) Google ScholarDigital Library
- Menasce D. A. and F. Almeida V. A.: Capacity Planning for Web Services Metrics, Models, and Methods, Prentice Hall PTR Upper Saddle River, N.J. 07458, ISBN: 0-13-065903-7. (2002) Google ScholarDigital Library
- Hellerstein J. L., Diao Y., et al.: IBM Research Report: applying control theory to computing systems. Proc. of Computer Science RC23459 (W0412-008). (2004)Google Scholar
- Hellerstein J. L., Diao Y., et al.: Feedback control of computing systems, IEEE Press Wiley-Interscience, ISBN:0-471-26637-X. (2004) Google ScholarDigital Library
- Ljung, L.: System identification: theory for the user 2nd ed. Upper Saddle River, N. J: Prentice Hall, ISBN: 0136566952. (1999) Google ScholarDigital Library
- the MathWorks Inc: System Identification ToolboxGoogle Scholar
- SUN Microsystems INC. New I/O APIs. http://java.sun.com/j2se/1.4.2/docs/guide/nio. (2002)Google Scholar
- Oppenheim A. V., Schafer R. W., et al.: Discrete-time signal processing, Prentice Hall Signal Processing Series, ISBN: 0137549202. (1999). Google ScholarDigital Library
- Vivek S. Pai, Peter Druschel, et al. Flash: An efficient and portable Web server. Proc. of the USENIX 1999 Annual Technical Conference. (1999). Google ScholarDigital Library
- Welsh M.: NBIO: Nonblocking I/O for Java http://www.eecs.harvard.edu/~mdw/proj/java-nbio/Google Scholar
- Welsh M.: An architecture for highly concurrent, well-conditioned internet services (Thesis). Computer Science, University of California at Berkeley. (2002) Google ScholarDigital Library
- Welsh M. and Culler D.: Adaptive Overload Control for Busy Internet Servers. Proc. of the Fifth USENIX Symposium on Internet Technologies and Systems (2003) Google ScholarDigital Library
- Welsh M. and Culler D.: Virtualization considered harmful: OS design directions for well-conditioned Services. Proc. of the 8th Workshop on Hot Topics in Operating Systems (HotOS VIII). (2001) Google ScholarDigital Library
- Welsh M., Culler D., et al.: SEDA: An architecture for well-conditioned scalable internet services. Proc. of the 18th ACM Symposium on Operating Systems Principles, anff, Canada. (2001) Google ScholarDigital Library
- SPECweb99 Benchmark, Copyright © 1995 - 2006 Standard Performance Evaluation Corporation http://www.spec.org/web99/Google Scholar
Index Terms
- Auto-tune design and evaluation on staged event-driven architecture
Recommendations
Some investigations on fuzzy P + fuzzy I + fuzzy D controller for non-stationary process
The paper addresses the adaptive behaviour of parallel fuzzy proportional plus fuzzy integral plus fuzzy derivative (FP+FI+FD) controller. The parallel FP+FI+FD controller is actually a non-linear adaptive controller whose gain changes continuously with ...
The Design and Simulation of Fuzzy PID Controller
IFITA '10: Proceedings of the 2010 International Forum on Information Technology and Applications - Volume 03Taking error and error rate of change as input, this paper realized the online slef-tuning of PID parameter with the method of fuzzy inference, and made a modeling design and simulation on the controller in the Matlab environment, and realized the non-...
On designing self-tuning controllers for AQM routers supporting TCP flows based on pole placement
This paper revisits the simple pole placement technique in the classical control theory, and exploits this technique to propose two kinds of controllers for active queue management (AQM) in Internet protocol (IP) routers: the self-tuning proportional ...
Comments