A Capacity Planning Process for Performance Assurance of Component-based Distributed Systems

Nilabja Roy, Abhishek Dubey, Aniruddha S. Gokhale, Larry W. Dowdy

ICPE'11 - Second Joint WOSP/SIPEW International Conference on Performance Engineering, Karlsruhe, Germany, March 14-16, 2011 2011

PDF DOI

Why This Matters

Web-scale enterprise systems must deliver consistent performance and high availability to users while managing operational costs and maintaining revenue through service level agreements. This work innovates by developing a systematic capacity planning process that combines analytical performance estimation with component-level profiling to handle increased system activity from multiple processing stages. The two-stage approach enables both accurate performance prediction and efficient resource allocation for complex multi-tier systems.

What We Did

This paper develops a capacity planning process for multi-tiered component-based distributed systems that enables service providers to guarantee performance and availability to customers without impacting revenue. The work presents a two-stage Modeling and Analysis using Queuing-Placement and Replication Optimization (MAQ-PRO) process that combines analytical modeling with profiling techniques to estimate performance requirements and determine optimal component placement. It addresses the challenge of high assurance of performance and availability while minimizing resource consumption.

Key Results

The approach successfully models multi-tier web applications using queuing theory extended with regression models that account for increased system activity and context switching overhead. Experimental results show accurate performance estimation that enables determining required resource allocations to satisfy service level agreements. The framework demonstrates effective capacity planning for real-world systems while maintaining application functionality and performance characteristics.

Full Abstract

Cite This Paper

@inproceedings{Roy2011b,
  author = {Roy, Nilabja and Dubey, Abhishek and Gokhale, Aniruddha S. and Dowdy, Larry W.},
  booktitle = {ICPE'11 - Second Joint {WOSP/SIPEW} International Conference on Performance Engineering, Karlsruhe, Germany, March 14-16, 2011},
  title = {A Capacity Planning Process for Performance Assurance of Component-based Distributed Systems},
  year = {2011},
  pages = {259--270},
  acceptance = {36},
  abstract = {For service providers of multi-tiered component-based applications, such as web portals, assuring high performance and availability to their customers without impacting revenue requires effective and careful capacity planning that aims at minimizing the number of resources, and utilizing them efficiently while simultaneously supporting a large customer base and meeting their service level agreements. This paper presents a novel, hybrid capacity planning process that results from a systematic blending of 1) analytical modeling, where traditional modeling techniques are enhanced to overcome their limitations in providing accurate performance estimates; 2) profile-based techniques, which determine performance profiles of individual software components for use in resource allocation and balancing resource usage; and 3) allocation heuristics that determine minimum number of resources to allocate software components. Our results illustrate that using our technique, performance (i.e., bounded response time) can be assured while reducing operating costs by using 25\% less resources and increasing revenues by handling 20\% more clients compared to traditional approaches.},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  biburl = {https://dblp.org/rec/bib/conf/wosp/RoyDGD11},
  category = {selectiveconference},
  contribution = {colab},
  doi = {10.1145/1958746.1958784},
  file = {:Roy2011b-A_Capacity_Planning_Process_for_Performance_Assurance_of_Component-based_Distributed_Systems.pdf:PDF},
  keywords = {capacity planning, multi-tier systems, performance modeling, quality of service, resource allocation, queuing models},
  project = {cps-middleware},
  tag = {platform},
  timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
  url = {https://doi.org/10.1145/1958746.1958784}
}
Quick Info
Year 2011
DOI 10.1145/1958746.1958784
Keywords
capacity planning multi-tier systems performance modeling quality of service resource allocation queuing models
Research Areas
scalable AI ML for CPS
Search Tags

Capacity, Planning, Process, Performance, Assurance, Component, Distributed, Systems, capacity planning, multi-tier systems, performance modeling, quality of service, resource allocation, queuing models, scalable AI, ML for CPS, 2011, Roy, Dubey, Gokhale, Dowdy