Ningfang Mi

I joined Department of Electrical and Computer Engineering at Northeastern University as an Assistant Professor in Fall 2009. I am looking for self-motivated PhD or Master's students to work in the area of capacity planning, resource management, energy/power management, performance evaluation, system modeling, simulation, virtulization, and cloud computing. Contact me if you are interested.

• Ph.D. in Computer Science, College of William and Mary, 2009.
• M.S. in Computer Science, University of Texas at Dallas, 2004.
• B.S. in Computer Science, Nanjing Univerity, 2000.

• 2012 The NSF Grant, CNS-1251129 ($272,351), PI, "An Integrated Framework for Performance and Reliability in Large-scaled Computing Systems"
• 2010 The IBM Faculty Award ($20,000)
• 2010 The AWS (Amazon Web Services) in Education Research Grant ($7,500)
• 2010 The Best Student Paper Award at the 22nd International Teletraffic Congres (ITC-22) for the paper titled "Fastrack for Taming Burstiness and Saving Power in Multi-Tiered Systems"
• 2009 The Computer Management Group (CMG) Graduate Fellowship
• 2008 The Best Paper Award at the ACM/IFIP/USENIX 9th International Middleware Conference for the paper titled "Burstiness in Multi-Tier Applications: Symptoms, Causes, and New Models"

[ps] [pdf] (updated by August 2012)

Refereed Journals

1. Giuliano Casale, Ningfang Mi, Ludmila Cherkasova, and Evgenia Smirni, "Dealing with Burstiness in Multi-Tier Applications: New Models and Their Parameterization", to appear, IEEE Transactions on Software Engineering (TSE), 2012.
2. Ningfang Mi, Giuliano Casale, and Evgenia Smirni, "ASIdE: Using Autocorrelation- Based Size Estimation for Scheduling Bursty Workloads", IEEE Transactions on Network and Service Management (TNSM), 9(2): 198-212, 2012.
3. Ningfang Mi, Giuliano Casale, Ludmila Cherkasova, and Evgenia Smirni, "Sizing Multi-Tier Systems with Temporal Dependence: Benchmarks and Analytic Models", Journal of Internet Services and Application (JISA), 1(2): 117-134, 2010.
4. Giuliano Casale, Ningfang Mi, and Evgenia Smirni, "Model-Driven System Capacity Planning Under Workload Burstiness", IEEE Transactions on Computers (TC), Vol. 59, No. 1, pp. 66-80, January 2010.
5. Ludmila Cherkasova, Kivanc Ozonat, Ningfang Mi, Julie Symons, and Evgenia Smirni, "Automated Anomaly Detection and Performance Modeling of Enterprise Applications", ACM Transactions on Computer Systems (TOCS), Vol. 27, Issue 3, November 2009.
6. Ningfang Mi, Alma Riska, Qi Zhang, Evgenia Smirni, and Eric Riedel, "Efficient Management of Idleness in Systems",  ACM Transactions on Storage (TOS), Vol. 5, No. 2, Article 4, June 2009.
7. Qi Zhang, Ludmila Cherkasova, Ningfang Mi, and Evgenia Smirni, "A Regression-Based Analytic Model for Capacity Planning of Multi-Tier Applications",  in Journal of Cluster Computing, Vol 11, No. 3, pp. 197-211, 2008.
8. Qi Zhang, Ningfang Mi, Alma Riska, and Evgenia Smirni, ''Performance-Guided Load (Un)Balancing Under Autocorrelated Flows'', in IEEE Transactions on Parallel and Distributed Systems (TPDS), Vol. 19, No. 2, pp. 652-665, 2008.
9. Ningfang Mi, Qi Zhang, Alma Riska, Evgenia  Smirni, and Eric Riedel, "Performance Impacts of Autocorrelated Flows in Multi-tiered Systems",  in Performance Evaluation, Vol 64, No. 9-12 pp. 1082-1101, Oct. 2007. Also presented at the 26th International Symposium on Computer Performance, Modeling, Measurements, and Evaluation (Performance'07), Cologne, Germany, 2007. Acceptance Rate: 20%.
10. Ovidiu Daescu, Ningfang Mi, Chan-Su Shin, and Alexander Wolff,  "Farthest-point Queries with Geometric and Combinatorial Constraints", in Computational Geometry: Theory & Applications, Vol 33, No. 3, pp.174-185, 2006.
11. Danny Z. Chen, Ovidiu Daescu, John Hershberger, Peter M. Kogge, Ningfang Mi,  and Jack Snoeyink, "Polygonal Path Approximation with Angle Constraints", in Computational Geometry: Theory & Applications, Vol. 32, No. 3, pp. 173-187, 2005.
12. Ovidiu Daescu, and Ningfang Mi, " Polygonal Path Approximation: a Query Based Approach", in Computational Geometry: Theory & Applications, Vol. 30, No. 1, pp. 41-58, 2005.

Refereed Conference Publications

Acceptance rates are provided when known.

1. Yi Yao, Jiayin Wang, Bo Sheng and Ningfang Mi, "Using a Tunable Knob for Reducing Makespan of MapReduce Jobs in a Hadoop Cluster", to appear, the IEEE International Conference on Cloud Computing (Cloud'13), Santa Clara Marriott, CA, June 2013. Acceptance Rate: 19.0%.
2. Yi Yao, Jianzhe Tai, Bo Sheng, and Ningfang Mi, "Scheduling Heterogeneous MapReduce Jobs for Efficiency Improvement in Enterprise Clusters", to appear, the IFIP/IEEE Integrated Network Management Symposium (IM'13), Ghent, Belgium, May 2013. (Short paper)
3. Zhen Li, Jianzhe Tai, Jiahui Chen, and Ningfang Mi, "ADUS:Adaptive Resource Allocation in Cluster Systems under Heavy-Tailed and Bursty Workloads", in the Proceedings of the IEEE International Conference on Communications (ICC'12), Ottawa, Canada, June, 2012. Acceptance Rate: 37.0%.
4. Jianzhe Tai, Juemin Zhang, Jun Li, Waleed Meleis and Ningfang Mi, "ArA: Adaptive Resource Allocation for Cloud Computing Environments under Bursty Workloads", in the Proceedings of IEEE International Performance Computing and Communications Conference (IPCCC'11), Orlando, Florida, Nov. 17-19, 2011. Acceptance Rate: 27.9%.
5. Yi Yao, Bo Sheng and Ningfang Mi, "DAT: An AP Scheduler using Dynamically Adjusted Time Windows for Crowded WLANs", in the Proceedings of IEEE International Performance Computing and Communications Conference (IPCCC'11), Orlando, Florida, Nov. 17-19, 2011. Acceptance Rate: 34.8%.
6. Juemin Zhang, Ningfang Mi, Jianzhe Tai and Waleed Meleis, "Decentralized Scheduling of Bursty Workload on Computing Grids", in the Proceedings of IEEE International Conference on Communications (ICC'11), Kyoto, Japan, 2011. Acceptance Rate: 38.5%.,
7. Andrew Caniff, Lei Lu, Ningfang Mi, Ludmila Cherkasova, and Evgenia Smirni, "Fastrack for Taming Burstiness and Saving Power in Multi-Tiered Systems", in the Proceedings of the 22nd International Teletraffic Congress (ITC'10), Amsterdam, The Netherlands, Sept, 2010. Acceptance Rate: 30%. (Best Student Paper Award)
8. Lei Lu, Ludmila Cherkasova, V. de Nitto Person, Ningfang Mi, and Evgenia Smirni, "AWAIT: Efficient Overload Management for Busy Multi-tier Web Services under Bursty Workloads", in the Proceedings of the 10th International Conference onWeb Engineering (ICWE'10), Vienna, Austria, July, 2010. Acceptance Rate: 20%.
9. Giuliano Casale, Ningfang Mi, and Evgenia Smirni, "CWS: a Model-Driven Scheduling Policy for Correlated Workloads", in the Proceedings of the 2010 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS'10), New York, NY, June, 2010. Acceptance Rate: 16%.
10. Ningfang Mi, Giuliano Casale, Alma Riska, Qi Zhang, and Evgenia Smirni, "Autocorrelation-Driven Load Control in Distributed Systems", in the Proceedings of 17th Annual Meeting of the IEEE/ACM International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS'09), London, U.K., September 2009. Acceptance Rate: 20%.
11. Ningfang Mi, Giuliano Casale, Ludmila Cherkasova, and Evgenia Smirni, "Injecting Realistic Burstiness to a Traditional Client-Server Benchmark", in the Proceedings of International Conference on Autonomic Computing and Communications (ICAC'09), pp. 149-158, Barcelona, Spain, 2009. Acceptance Rate: 16%.
12. Alma Riska, Ningfang Mi, Giuliano Casale, and Evgenia Smirni, "Feasibility Regions: Exploiting Trade-offs between Power and Performance in Disk Drives", in the Proceedings of second Workshop on Hot Topics in Measurement & Modeling of Computer Systems (HotMetrics'09), ACM Perf. Eval. Rev, Vol. 37, Issue 3, pp. 49-54, Seattle, WA, 2009.
13. Ningfang Mi, Alma Riska, Xin Li, Evgenia Smirni, and Erik Riedel, "Restraint Utilization of Idleness for Transparent Scheduling of Background Tasks", in the Proceedings of the 2009 ACM SIGMETRICS international conference on Measurement and modeling of computer systems (SIGMETRICS'09), pp. 205-216, Seattle, WA, 2009. Acceptance Rate: 15%.
14. Ningfang Mi, Giuliano Casale, Ludmila Cherkasova, and Evgenia Smirni, "Burstiness in Multi-Tier Applications: Symptoms, Causes, and New Models", in ACM/IFIP/USENIX 9th International Middleware Conference (Middleware'08),  pp. 265-286, Leuven, Belgium, 2008. Acceptance Rate: 18%. (Best Paper Award)
15. Ningfang Mi, Giuliano Casale, and Evgenia Smirni, "Scheduling for Performance and Availability in Systems with Temporal Dependent Workloads", in the International Conference on Dependable Systems and Networks (DSN'08),  pp. 336-345, Anchorage, AK, 2008. Acceptance Rate: 25%.
16. Ningfang Mi, Alma Riska, Evgenia Smirni, and Erik Riedel, "Enhancing Data Availability through Background Activities", in the International Conference on Dependable Systems and Networks (DSN'08),  pp. 492-501, Anchorage, AK, 2008. Acceptance Rate: 25%.
17. Ludmila Cherkasova, Kivanc Ozonat, Ningfang Mi, Julie Symons, and Evgenia Smirni, "Anomaly? Application Change? or Workload Change?", in the International Conference on Dependable Systems and Networks (DSN'08),  pp. 452-461, Anchorage, AK, 2008. Acceptance Rate: 25%.
18. Giuliano Casale, Ningfang Mi, and Evgenia Smirni, "Bound Analysis of Closed Queueing Networks with Workload Burstiness", in the Proceedings of the 2008 ACM SIGMETRICS international conference on Measurement and modeling of computer systems (SIGMETRICS'08), pp. 13-24, Annapolis, Maryland,   2008. Acceptance Rate: 18%.
19. Giuliano Casale, Ningfang Mi, Ludmila Cherkasova, and Evgenia Smirni, "How to Parameterize Models with Bursty Workloads", in the Proceedings of First Workshop on Hot Topics in Measurement & Modeling of Computer Systems (HotMetrics'08), Annapolis, Maryland,  2008. Acceptance Rate: 27%.
20. Ningfang Mi, Ludmila Cherkasova, Kivanc Ozonat, Julie Symons, and Evgenia Smirni, "Analysis of Application Performance and Its Change via Representative Application Signatures", accepted in IEEE/IFIP Network Operations and Management Symposium (NOMS'08), Salvador, Brazil, pp. 216-223, 2008. Acceptance Rate: 27%.
21. Ningfang Mi, Qi Zhang, Alma Riska, and Evgenia Smirni, "Load Balancing for Performance Differentiation in Dual-Priority Clustered Servers", in the 3rd International Conference on the Quantitative Evaluation of Systems (QEST'06), Riverside, CA, pp. 385-394, 2006.
22. Qi Zhang, Ningfang Mi, Alma Riska, and Evgenia Smirni, "Load Unbalancing to Improve Performance under Autocorrelated Traffic", in the 26th International Conference on Distributed Computing Systems (ICDCS'06), Lisboa, Portugal, pp. 20, Jul. 2006. Acceptance Rate: 14%.
23. Qi Zhang, Alma Riska, Ningfang Mi, Erik Riedel, and Evgenia Smirni, "Evaluating the Performability of Systems with Background Jobs", in the International Conference on Dependable Systems and Networks (DSN'06), Philadelphia, PA, pp. 495-504, 2006. Acceptance Rate: 18%.
24. Ovidiu Daescu, Ningfang Mi, Chan-Su Shin, and Alexander Wolff,  "Farthest-Point Queries with Geometric and Combinatorial Constraints", in Proceedings of the Japan Conference on Discrete and Computational Geometry (JCDCG'04), pp. 62-75, 2004.
25. Ovidiu Daescu, and Ningfang Mi, "Polygonal Path Approximation: a Query Based Approach", in Proceedings of the 14th Annual International Symposium on Algorithms and Computation (ISAAC'03), pp. 36-46, 2003.

Invited Publications

Giuliano Casale, Ningfang Mi, and Evgenia Smirni, "Versatile Models of Systems Using MAP Queueing Networks", in the IEEE International Parallel and Distributed Processing Symposium (IPDPS), Next Generation Software (NGS) Workshop, 2008.

Evgenia Smirni, Qi Zhang, Ningfang Mi, Alma Riska, and Giuliano Casale, " New Results on the Performance Effects of Autocorrelated Flows in Systems ", in the IEEE International Parallel and Distributed Processing Symposium (IPDPS), Next Generation Software (NGS) Workshop, Long Beach, CA, pp. 1-6, 2007.

[Back to Top]

• ARA: Adaptive Resource Allocation for Cloud Computing Environment

  Cloud computing platforms provide mechanisms that allow multiple instances of an application to run simultaneously across the cloud which could be aggressively aggregated together as the pattern of burst. We argue that the presence of burstiness can cause load unbalancing in clouds and consequently degrade the overall system performance. Motivated by this problem, we are developing novel methodologies for resource allocation in cloud systems, which attempt to counteract the deleterious effect of burstiness, improving overall system performance and availability and maintaining the cloud usersa€? Service Level Agreement (SLA)We expect that our new resource allocation methodologies are suitable for cloud systems (e.g., Amazon EC2) which attempt to reduce the effect of burstiness on system performance and optimize resource utilizations.

  This work is currently supported by AWS (Amazon Web Services) in Education Research Grant and has been presented in ICC'11. Project Link.

  ---

• Impact of Autocorrelation in Multi-tiered Systems

  Multi-tiered systems, a prevalent architecture of today's web sites, is an example of closed systems because the hardware imposes a limit on the number of simultaneous connections. In collaboration with researchers at Seagate Research, we have built an e-commerce server according to the TPC-W benchmark to identify the presence of autocorrelation in different tiers of the system. We observed that autocorrelated flows severely degrade overall system performance even under medium loads. It implies that if autocorrelation is ignored, the throughput and utilization of specific devices - metrics often used in capacity planning and admission control - may give a distorted view of system performance.

  This work has been presented in Performance'07.

  ---

• General Autocorrelation-driven Scheduling Policies

  Temporal dependence in workloads creates conditions in which a server, in order to remain available, should quickly process bursts of requests with large service requirements. We have studied on how to counteract the resulting peak congestion and maintain high availability by delaying/dropping selected requests that contribute to temporal locality. SWAP was proposed to approximate the shortest job first (SJF) scheduling without requiring any knowledge of job service times. We also designed ALoC, an autocorrelation-driven load control policy, that drops a percentage of the requests in order to meet pre-defined quality-of-service levels. To the best of our knowledge, this is the first direct application of autocorrelation of service times to autonomic load control.

  This work has been presented in DSN'08.

  ---

• Background Job Scheduling Methodologies in Storage Systems

  Background activities are scheduled with low priority and served during system idle times. We presented that idle waiting, i.e., delay scheduling of a background job, is insufficient as a ``standalone'' technique to manage the trade-off between the performance of foreground and background tasks. We complemented ``idle-waiting'' with the ``estimation'' of background work to be served in every idle interval and then proposed a methodology that determines the schedulability of background work in storage systems, i.e., when and for how long idle times can be used for serving background tasks. An extensive set of trace-driven simulation experiments and measurements in a prototype on the Linux 2.6.22 kernel, show that the new approach meets the performance targets by finding a solution that is among the best. Finally, the effectiveness of two known as background activities, namely scrubbing and intra-disk data redundancy, is evaluated to detect and/or recover from latent sector errors.

  This work has been presented in TOS, DSN'08.

  ---

• New Capacity Planning Models for Autocorrelated Workloads

  Building effective models of complex enterprise systems are central to capacity planning and resource provisioning. We found that if autocorrelated flows exist in the system, then classic queueing theory models such as MVA give wrong predictions. Indeed, We have devised a new class of queueing network models that overcome the weakness of classic models by capturing the performance effects of autocorrelation in the service process. However, there is a lack of understanding and of practical results on how to perform model parameterization, especially when this model parameterization must be derived from limited coarse measurements as is often encountered in practice. In collaboration with researchers at HP Labs, We have devised a new methodology to integrate workload autocorrelation in performance models, which well captures autocorrelation and variability of the true service process, despite inevitable inaccuracies that result from inexact and limited measurements.

  This work has been presented in HotMetrics'08, SIGMETRICS'08, Middleware'08.

  ---

• Automated Detection of Application Performance Anomaly and Change

  Application servers are a core component of a multi-tier architecture that has become the industry standard for building scalable client-server applications. A client communicates with a service deployed as a multi-tier application via request-reply transactions. A typical server reply consists of the web page dynamically generated by the application server. Then, the application server may issue multiple database calls while preparing the reply. Understanding the cascading effects of the various tasks that are sprung by a single request-reply transaction is a challenging task. we have addressed the problem of efficiently diagnosing essential performance changes in application behavior in order to provide timely feedback to application designers and service providers. A new approach based on an application signature has been proposed to enable a quick performance comparison of the new application signature against the old one, while the application continues its execution in the production environment. Application signatures provide a simple and powerful solution that are further used for efficient capacity planning, anomaly detection, and provisioning of multi-tier applications in rapidly evolving IT environments.

  This work has been presented in NOMS'08, DSN'08.

Paper Reading Lists

[Back to Top]

Present Graduate Students:

• Jianzhe Tai: Ph.D. student
• Yi Yao: Ph.D. student
• Wei Cai: M.S. student
• Anoop Raghunathan: M.S. student
• Yuqing Lin: M.S. student
• Jun Li: M.S. student (grauduated in Aug. 2012)
• Chen Mao: M.S. student (grauduated in Aug. 2012)
• Jiahui Chen: M.S. student (grauduated in Aug. 2011)
• Zhen Li: M.S. student (graduated in Aug. 2011)

[Back to Top]

Teaching

Fall 2009

• EECE3326 Optimization Methods

Covers the design and implementation of algorithms to solve engineering problems using a high-level programming language. Reviews elementary data structures, such as arrays, stacks, queues, and lists, and introduces more advanced structures, such as trees and graphs and the use of recursion. Covers both the algorithms to manipulate these data structures as well as their use in problem solving. Emphasizes the importance of software engineering principles. Introduces algorithm complexity analysis and its application to developing efficient algorithms. Prereq. CS 1500.

Syllabus

Course website on blackboard.

Spring 2010

• EECE7398 Simulation and Performance Evaluation

Covers topics on computer simulation and performance evaluation in computer systems. The course mainly covers both classic and timely techniques in the area of performance evaluation, including capacity planning to predict system performance, scheduling, and resource allocation in systems. The course also introduces some basic computational and mathematical techniques for modeling, simulating and analyzing the performance by using simulation, including models, random-number generation, statistics, and discrete event-driven simulation.

Syllabus

Fall 2010

• EECE3326 Optimization Methods

Spring 2011

• EECE5698 Simulation and Performance Evaluation

  Paper Presentation List

Fall 2011

• EECE7366 Simulation and Performance Evaluation

  Paper Presentation List

Editorial board member of "Simulation Modelling Practice and Theory"

Demo/poster chair for ACM International Conference on Performance Engineering (ICPE 2012)

Publicity Co-Chair for ACM/IFIP/USENIX International Middleware Conference (2010)

TPC Member

• CCGrid: IEEE International Symposium on Cluster, Cloud and Grid Computing (2013)
• MiPS: Workshop of Middleware for Pervasive Systems (2013)
• QEST: International Conference on the Quantitative Evaluation of Systems (2010, 2011)
• SIGMETRICS: International conference on Measurement and modeling of computer systems (2010) - Shadow Program Committee Member