Open AccessProceedings Article10.1109/RTSS.2006.20

Distributed Utilization Control for Real-Time Clusters with Load Balancing

- 05 Dec 2006

- pp 137-146

TL;DR: Simulation results and control-theoretic analysis demonstrate that DUC-LB can provide robust utilization control and effective load balancing in large-scale clusters.

Abstract: Previous years have seen rapid growth of online services that rely on large-scale server clusters to handle high volume of requests. Such clusters must adaptively control the CPU utilizations of many processors in order to maintain desired soft real-time performance and prevent system overload in face of unpredictable workloads. This paper presents DUC-LB, a novel distributed utilization control algorithm for cluster-based soft real-time applications. Compared to earlier works on utilization control, a distinguishing feature of DUC-LB is its capability to handle system dynamics caused by load balancing, which is a common and essential component of most clusters today. Simulation results and control-theoretic analysis demonstrate that DUC-LB can provide robust utilization control and effective load balancing in large-scale clusters

AI Agents for this Paper

Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps

Most frequently asked questions

1. What have the authors contributed in "Distributed utilization control for real-time clusters with load balancing" ?

This paper presents DUC-LB, a novel distributed utilization control algorithm for cluster-based soft real-time applications.

Figure 5. Utilization and Response Times under DUC-LB when execution time fluctuates globally

Figure 6. Utilization and Response Times under DUC-LB when execution time fluctuates locally

Figure 3. Average Utilization and Deviation of DUC-LB When Execution-time Factor Varies

Figure 4. Stable Regions of DUC-LB, FCU-LB and Analytical Result

Figure 2. Utilization of DUC-LB and FCU-LB at etf = 1.1

Figure 1. DUC-LB Framework Overview(5 × 5 torus)

Citations

•Journal Article•10.1109/TNSM.2009.04.090403

Resource allocation and disturbance rejection in web servers using SLAs and virtualized servers

Martin Ansbjerg Kjær, +2 more

- 01 Dec 2009

- IEEE Transactions on Network and Service...

TL;DR: This article proposes a method where parameters are estimated online, and thus also adapts to the changing environment, and compares it to two other control strategies proposed in the literature, which are based on off-line estimation of certain parameters.

...read moreread less

•Proceedings Article•10.1109/RTSS.2006.20

Distributed Utilization Control for Real-Time Clusters with Load Balancing

Yong Fu, +3 more

- 05 Dec 2006

TL;DR: Simulation results and control-theoretic analysis demonstrate that DUC-LB can provide robust utilization control and effective load balancing in large-scale clusters.

...read moreread less

Journal Article•10.1016/J.AUTOMATICA.2012.02.001

Brief paper: A linear programming approach to routing control in networks of constrained linear positive systems

Heather Arneson, +1 more

- 01 May 2012

- Automatica

TL;DR: A linear programming (LP) characterization of optimal constant routing policies for the infinite-horizon routing problem under weighted l"1 constraints and objective is derived.

...read moreread less

Proceedings Article•10.23919/ECC.2009.7074401

Honoring SLAs on cloud computing services: A control perspective

Christos Yfoulis, +1 more

- 01 Aug 2009

TL;DR: A survey of recent results in the literature indicates that the problem of honoring SLAs on cloud computing services is a new interesting application for control theory and that researchers can benefit significantly from a number of well-known modern control methodologies, such as hybrid, supervisory, hierarchical and model predictive control.

...read moreread less

•Journal Article•10.2478/V10006-012-0054-1

Modeling nonlinear road traffic networks for junction control

Tamás Péter

- 01 Sep 2012

- International Journal of Applied Mathema...

TL;DR: A method of mathematical modeling of high scale road traffic networks, where a new special hypermatrix structure is intended to be used, where the structure describes the inner–inner, inner–outer and outer–outer relations, and laws of a network area is introduced.

...read moreread less

...

Expand

References

Journal Article•10.1145/321738.321743

Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment

C. L. Liu, +1 more

- 01 Jan 1973

- Journal of the ACM

TL;DR: The problem of multiprogram scheduling on a single processor is studied from the viewpoint of the characteristics peculiar to the program functions that need guaranteed service and it is shown that an optimum fixed priority scheduler possesses an upper bound to processor utilization.

...read moreread less

9.4K

•Book

Robust and Optimal Control

John Doyle

- 17 Aug 1995

TL;DR: This paper reviewed the history of the relationship between robust control and optimal control and H-infinity theory and concluded that robust control has become thoroughly mainstream, and robust control methods permeate robust control theory.

...read moreread less

7.1K

Journal Article•10.1016/0967-0661(96)83721-X

Robust and optimal control

Tadeusz Kaczorek

- 01 Aug 1996

- Control Engineering Practice

1.4K

Journal Article•10.1016/0743-7315(89)90021-X

Dynamic load balancing for distributed memory multiprocessors

George Cybenko

- 01 Oct 1989

- Journal of Parallel and Distributed Comp...

TL;DR: This paper completely analyze the hypercube network by explicitly computing the eigenstructure of its node adjacency matrix and shows that a diffusion approach to load balancing on a hypercube multiprocessor is inferior to another approach which is called the dimension exchange method.

...read moreread less

1.1K

Journal Article•10.1109/TAC.2003.816954

Distributed control design for spatially interconnected systems

Raffaello D'Andrea, +1 more

- 15 Sep 2003

- IEEE Transactions on Automatic Control

TL;DR: A state space framework for posing problems of this type, and focus on systems whose model is spatially discrete, is developed using the l/sub 2/-induced norm as the performance criterion.

...read moreread less

904

...

Expand

Distributed Utilization Control for Real-Time Clusters with Load Balancing

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What have the authors contributed in "Distributed utilization control for real-time clusters with load balancing" ?

Figures

Citations

Resource allocation and disturbance rejection in web servers using SLAs and virtualized servers

Distributed Utilization Control for Real-Time Clusters with Load Balancing

Brief paper: A linear programming approach to routing control in networks of constrained linear positive systems

Honoring SLAs on cloud computing services: A control perspective

Modeling nonlinear road traffic networks for junction control

References

Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment

Robust and Optimal Control

Robust and optimal control

Dynamic load balancing for distributed memory multiprocessors

Distributed control design for spatially interconnected systems

Related Papers (5)

Comparative studies of load balancing with control and optimization techniques

New Approach for Modeling, Analysis, and Control of Air Traffic Flow

Decentralized utilization control in distributed real-time systems

An Algorithmic Approach for Airspace Flow Programs

Feedback Control of Computing Systems