Feedback performance control in software services T.F. Abdelzaher, J.A. Stankovic, C. Lu, R. Zhang, and Y. Lu, Feedback Performance Control in Software.

Slides:



Advertisements
Similar presentations
Feedback Control Real- time Scheduling James Yang, Hehe Li, Xinguang Sheng CIS 642, Spring 2001 Professor Insup Lee.
Advertisements

CprE 458/558: Real-Time Systems
Feedback Control Theory a Computer Systems Perspective Introduction What is feedback control? Why do computer systems need feedback control? Control design.
Feedback Control Real-Time Scheduling: Framework, Modeling, and Algorithms Chenyang Lu, John A. Stankovic, Gang Tao, Sang H. Son Presented by Josh Carl.
Scheduling in Web Server Clusters CS 260 LECTURE 3 From: IBM Technical Report.
Application Performance in the QLinux Multimedia Operating System Jun Wang Jun Wang.
DEXA 2005 Control-based Quality Adaptation in Data Stream Management Systems (DSMS) Yicheng Tu†, Mohamed Hefeeda‡, Yuni Xia†, Sunil Prabhakar†, and Song.
REAL-TIME COMMUNICATION ANALYSIS FOR NOCS WITH WORMHOLE SWITCHING Presented by Sina Gholamian, 1 09/11/2011.
Backlog Estimation and Management for Real-Time Data Services Kyoung-Don Kang, Jisu Oh, and Yan Zhou Department of Computer Science State University of.
Efficient Autoscaling in the Cloud using Predictive Models for Workload Forecasting Roy, N., A. Dubey, and A. Gokhale 4th IEEE International Conference.
1 The Challenges of Reflexive Control Systems Lui Sha
IP traffic and QoS control : the need for flow aware networking Jim Roberts France Telecom R&D NSF-COST Workshop.
Towards Feasibility Region Calculus: An End-to-end Schedulability Analysis of Real- Time Multistage Execution William Hawkins and Tarek Abdelzaher Presented.
A Case for Relative Differentiated Services and the Proportional Differentiation Model Constantinos Dovrolis Parameswaran Ramanathan University of Wisconsin-Madison.
SIGMETRICS 2008: Introduction to Control Theory. Abdelzaher, Diao, Hellerstein, Lu, and Zhu. CPU Utilization Control in Distributed Real-Time Systems Chenyang.
Dynamic Process Allocation in Apache Server Yu Cai.
Traffic Engineering With Traditional IP Routing Protocols
Energy Management and Adaptive Behavior Tarek Abdelzaher.
AQM for Congestion Control1 A Study of Active Queue Management for Congestion Control Victor Firoiu Marty Borden.
End-to-End Analysis of Distributed Video-on-Demand Systems Padmavathi Mundur, Robert Simon, and Arun K. Sood IEEE Transactions on Multimedia, February.
1 Action Breakout Session Anil, AP, Nina Bhatti, Charles Berdnall, Joe Hellerstein, Wei Hu, Anthony Joseph, Randy Katz, Li, Machi Mukund Kimmo Raatikanen,
Load Adaptation: Options for Basic Services Vance Maverick ADAPT Bologna Feb. 13, 2003.
Fair Scheduling in Web Servers CS 213 Lecture 17 L.N. Bhuyan.
Measuring Performance Chapter 12 CSE807. Performance Measurement To assist in guaranteeing Service Level Agreements For capacity planning For troubleshooting.
Yaksha: A Self-Tuning Controller for Managing the Performance of 3-Tiered Web Sites Abhinav Kamra, Vishal Misra CS Department Columbia University Erich.
LDU Parametrized Discrete-Time Multivariable MRAC and Application to A Web Cache System Ying Lu, Gang Tao and Tarek Abdelzaher University of Virginia.
Misconceptions About Real-time Computing : A Serious Problem for Next-generation Systems J. A. Stankovic, Misconceptions about Real-Time Computing: A Serious.
Bandwidth Allocation in a Self-Managing Multimedia File Server Vijay Sundaram and Prashant Shenoy Department of Computer Science University of Massachusetts.
Dynamic Process Allocation in Apache Server Yu Cai.
© 2008 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Automated Workload Management in.
1 A State Feedback Control Approach to Stabilizing Queues for ECN- Enabled TCP Connections Yuan Gao and Jennifer Hou IEEE INFOCOM 2003, San Francisco,
1 Chapter 7 Dynamic Job Shops Advantages/Disadvantages Planning, Control and Scheduling Open Queuing Network Model.
End-to-End Delay Analysis for Fixed Priority Scheduling in WirelessHART Networks Abusayeed Saifullah, You Xu, Chenyang Lu, Yixin Chen.
Computer System Lifecycle Chapter 1. Introduction Computer System users, administrators, and designers are all interested in performance evaluation. Whether.
Active Network Applications Tom Anderson University of Washington.
Distributed Real-Time systems 1 By: Mahdi Sadeghizadeh Website: Sadeghizadeh.ir Advanced Computer Networks.
Adaptive Control of Virtualized Resources in Utility Computing Environments HP Labs: Xiaoyun Zhu, Mustafa Uysal, Zhikui Wang, Sharad Singhal University.
Dynamic and Decentralized Approaches for Optimal Allocation of Multiple Resources in Virtualized Data Centers Wei Chen, Samuel Hargrove, Heh Miao, Liang.
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jennifer Rexford Princeton University With Jiayue He, Rui Zhang-Shen, Ying Li,
Applying Feedback Control to QoS management - an introduction -
Tufts Wireless Laboratory School Of Engineering Tufts University “Network QoS Management in Cyber-Physical Systems” Nicole Ng 9/16/20151 by Feng Xia, Longhua.
Distributed Multimedia March 19, Distributed Multimedia What is Distributed Multimedia?  Large quantities of distributed data  Typically streamed.
Computers on Cruise Control Creating Adaptive Systems with Control Theory Ricardo Portillo The University of Texas at El Paso
Brief Review of Control Theory
Computer Networks Performance Metrics. Performance Metrics Outline Generic Performance Metrics Network performance Measures Components of Hop and End-to-End.
Quality of Service Karrie Karahalios Spring 2007.
임규찬. 1. Abstract 2. Introduction 3. Design Goals 4. Sample-Based Scheduling for Parallel Jobs 5. Implements.
Large Scale Deeply Embedded Networks Jack Stankovic, Tarek Abdelzaher, Sang Son, Chenyang Lu Department of Computer Science University of Virginia Fall.
1 IEEE Meeting July 19, 2006 Raj Jain Modeling of BCN V2.0 Jinjing Jiang and Raj Jain Washington University in Saint Louis Saint Louis, MO
Load Shedding in Stream Databases – A Control-Based Approach Yicheng Tu, Song Liu, Sunil Prabhakar, and Bin Yao Department of Computer Science, Purdue.
Empirical Quantification of Opportunities for Content Adaptation in Web Servers Michael Gopshtein and Dror Feitelson School of Engineering and Computer.
© 2006 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Injecting Realistic Burstiness to.
Measuring the Capacity of a Web Server USENIX Sympo. on Internet Tech. and Sys. ‘ Koo-Min Ahn.
Managing Web Server Performance with AutoTune Agents by Y. Diao, J. L. Hellerstein, S. Parekh, J. P. Bigus Presented by Changha Lee.
Quality Is in the Eye of the Beholder: Meeting Users ’ Requirements for Internet Quality of Service Anna Bouch, Allan Kuchinsky, Nina Bhatti HP Labs Technical.
Control-based Quality Adaptation in Data Stream Management Systems (DSMS) Yicheng Tu†, Song Liu‡, Sunil Prabhakar†, and Bin Yao‡ † Department of Computer.
Optimization-based Cross-Layer Design in Networked Control Systems Jia Bai, Emeka P. Eyisi Yuan Xue and Xenofon D. Koutsoukos.
Spark on Entropy : A Reliable & Efficient Scheduler for Low-latency Parallel Jobs in Heterogeneous Cloud Huankai Chen PhD Student at University of Kent.
Dynamic Resource Allocation for Shared Data Centers Using Online Measurements By- Abhishek Chandra, Weibo Gong and Prashant Shenoy.
Feedback Control Real- Time Scheduling C. Lu, J.A. Stankovic, G. Tao, and S.H. Son, Design and Evaluation of a Feedback Control EDF Scheduling Algorithm,
OPERATING SYSTEMS CS 3502 Fall 2017
Action Breakout Session
Standards and Patterns for Dynamic Resource Management
Provision of Multimedia Services in based Networks
Dynamic Process Allocation in Apache Server
CSE 550 Computer Network Design
Feedback Control Real-time Scheduling
Towards Predictable Datacenter Networks
Presentation transcript:

Feedback performance control in software services T.F. Abdelzaher, J.A. Stankovic, C. Lu, R. Zhang, and Y. Lu, Feedback Performance Control in Software Services, IEEE Control Systems, 23(3): 74-90, June 2003.Feedback Performance Control in Software Services,

Overview SW systems become larger and bigger Performance guarantee required, e.g., in web-based e-commerce Control theory  Promising theoretical foundation for perf control in complex SW applications, e.g., real- time scheduling, web servers, multimedia control, storage mangers, power management, routing in computer networks, …

Overview Software performance assurance problems -> Feedback control problems focused on web server performance guarantee problems

SW performance control Less rigorous guarantees on perf and quality Most SW eng. research deals with the development of functionally correct SW Functional correctness is not enough!  Timeliness in embedded systems Correct but delayed action can be disastrous  Non-fucntional QoS attributes, e.g., timeliness, security, availability, …

Traditional approaches for perf guarantees Worst case estimates of load & resource availability  Recall EDF, RM, DM, Priority Ceiling Protocol, …

New demand for performance assurance QoS guarantees required in a broader scope of applications run in open, unpredictable environments  Global communication networks enabling online banking, trading, distance learning, …  Points of massive aggregation suffering unpredictable loads, potential bottlenecks, DoS attacks, … -> Precise workload/system model unknown a priori  Failure to meet QoS requirements -> loss of customers or financial damages  Worst case analysis/overdeisgn could be overly pessimistic or wasteful  Solid analytic framework for cost-effective perf assurance required

Challenges How to model SW architecture? How to map a specific QoS problem into a feedback control system? How to choose proper SW sensors and actuators to monitor and adjust perf and workloads/resource allocation? How to design controllers for servers? -> This paper focuses on web servers

QoS metrics Delay metrics  Proportional to time: queuing delays, execution latencies, service response time Rate metrics  Inversely proportional to time  Connection bandwidth, throughput, packet rate

Time-related perf attributes Can be controlled by adjusting resource allocation  Queuing theory can predict perf given a particular resource allocation or vice versa  Queuing theory only works for Poisson arrival patterns Queuing theory can only predict average perf even if this assumption holds  Arrival patterns in web applications follow heavy-tailed distribution -> Bursty arrival patterns

Service architecture Fig. 1 Server architecture: (a) computing model (b) control-oriented representation Liquid task model

C i << D i  Takes C i units of time to serve request i  D i is the max tolerable response time  Tolerable response time is finite  Service times are infinitesimal Progress of requests through the server queues ≈ Fluid flow Service rate at stage k = dN k (t)/dt where N k is #requests processed by stage k

Liquid task model Volume at time T≈ #requests queued at stage k = ∫ T (F in – F k )  F k : service rate at stage k  F in : request arrival rate to this stage Valves: points of control, i.e., manipulated variables such as the queue length Liquid model does not describe how individual requests are prioritized Control theory can be combined with queuing theory or real-time scheduling

Server modeling Difference equation to model web servers  y(k): perf, e.g., delay or throughput, measured at the k th sampling period  U(k): control input at the k th sampling period  ARMA (AutoreRressive Moving Average) model y(k) = a 1 y(k-1) + a 2 y(k-2) + … + a n y(k-n) + b 1 u(k-1) + b 2 u(k-2) + … + b n u(k-n) Transfer function can be derived  Web proxy cache model [4]  TCP dynamics [5]

Resource allocation for QoS guarantees Allocate more/less resource = open/close a valve Need actuators to control resource allocation or QoS provided by the system

SW system actuators Input flow actuators  Admission control  Control queue length, server utilization, …  Reject some requests under overload

SW system actuators Quality adaptation actuators  Change processing requirements to increase server rate under overload  E.g., Return abbreviated web page under overload  Tradeoff btwn delay & quality  Service level m in a range [0, M] where 0 is rejection

Resource reallocation actuator Alter the amount of allocated resources Usually applicable to multiple classes of clients, e.g., dynamically reallocate disk space to support the service delay ratio 1:2 between two service classes [4,7]

QoS Mapping Convert common resource management & SW perf assurance problems to FC problems Absolute convergence guarantee Relative guarantee Resource reservation guarantee Prioritization guarantee Statistical multiplexing guarantee Utility optimization guarantee

Absolute convergence guarantee Convergence to the specified problem Overshoot: Maximum deviation Settling time: Time taken to recover the desired perf

Absolute convergence guarantee Rate & queue length control  Result in linear FC  (Flow) rate can be directly controlled by actuators  Queue length can be linearly controlled by controlling the flow  E.g., server utilization control loop

Absolute convergence guarantee Delay control  More difficult  Delay is inversely proportional to flow Queuing delay d = Q/r where Q is queue length & r is service rate Nonlinear

Relative guarantee For example, fix the delays of two traffic classes at a ratio 3:1 H i : measured perf of class i C i : weight of class i Relative guarantee specifies H 1 :H 2 = 1:3 Set point = 1/3 Error e = 1/3 – H 1 /H 2

Controlled variable: relative delay ratio Manipulated variable: #allocated processes per class to control connection delay HTTP protocol summary  A client, e.g., web browser establishes a TCP connection with a server process  The client submits an HTTP request to the sever over the TCP connection  The server sends the response back to the client  Keep open the TCP connection for the Keep Alive interval, e.g., 15s -> Claim connection delay dominates service response time -> Scheduling can also significantly relative delay ratio, but it is not considered Relative guarantee in Apache web server

System identification based on the ARMA model Randomly change per class process allocations Measure response time

Relative guarantee in Apache web server Perf settings  4 Linux machines run the Surge web workload generator  1 Linux machine runs the Apache web server  Suddenly increase #premium clients by 100 at time 870s

Relative guarantee in Apache web server Perf results Open Loop Closed Loop Stable?

Related work ControlWare CPU scheduling Storage management Network routers Power/heat management RTDB

Conclusions Feedback control is applicable to managing performance in SW systems Future work  Adaptive/robust control  Predictive control  Apply to other computational systems such as embedded systems

Adptive Control: Self-Tuning Regulator Dynamically estimate a model of the system via the Recursive Least Square method Controller will accordingly set the actuators to support the desired perf.

References (HP Storage Systems Lab) Designing controllable computer systems, Christos Karamanolis, Magnus Karlsson and Xiaoyun Zhu. USENIX Workshop on Hot Topics in Operating Systems (HotOS), June 2005, pp , Santa Fe, NM. Designing controllable computer systems Dynamic black-box performance model estimation for self-tuning regulators, Magnus Karlsson and Michele Covell. International Conference on Autonomic Computing (ICAC), pp , June 2005, Seattle, WA. Dynamic black-box performance model estimation for self-tuning regulators

IBM Autonomic Computing Lab ndex.html ndex.html General, broader research issues regarding self-tuning, self-managing systems Also, visit Joe Hellerstein’s Adaptive Systems DepartmentJoe Hellerstein’s Adaptive Systems Department

Some University Labs Tarek Abdelzaher: Chenyang Lu:

Announcement Programming Assignment 1 is posted on the course web page

Questions?