1 Rajat Mehrotra Curriculum Vitae CONTACT INFORMATION RESEARCH INTERESTS #204, CRVB, Desert Research Institute Phone: , Raggio Parkway, Reno, NV https://sites.google.com/site/rajatmeh/ Autonomic Computing Systems, Cloud Computing Systems, Model Integrated Computing, Componentbased Management Frameworks, Cyber Physical Systems, Sensor Networks, Predictive Modeling, Big Data Systems, Energy-Aware HPC systems, and Machine Learning. EDUCATION Mississippi State University, Mississippi State, MS USA Ph.D., Electrical and Computer Engineering, Dec 2013 Motilal Nehru National Institute of Technology, Allahabad, India B.Tech., Computer Science and Engineering, July 2005 EXPERIENCE 1. Applied Innovation Center, Desert Research Institute, Reno, NV Innovation Fellow Applied Innovation Center for Advanced Analytics. 2. Department of Computer Science, University of Virginia, Charlottesville, VA July 2014 Present Research Associate Jan 2014 July 2014 PERFECT: Power Efficiency Revolution For Embedded Computing Technologies. 3. NSF Center for Cloud and Autonomic Computing, Mississippi State University, Mississippi State, MS Research Assistant August 2008 Dec 2013 Performance Modeling of Distributed Enterprise Multi-Tier Enterprise Systems. A Model-Based Framework for Autonomic Performance Management of Cloud Computing Systems. Power-Aware Autonomic Approach for High Performance Computing Systems. Automated Model-based Security Management of Web Services. 4. Institute for Software Integrated Systems, Vanderbilt University, Nashville, TN Visiting Research Scholar May 2011 Aug 2011 RFDMon: A Real-Time and Fault-Tolerant Distributed System Monitoring Approach. 5. Alcatel-Lucent Wireless R&D Division, Noida, India Senior Software Engineer May 2005 July 2008 Subject Matter Expert (SME) for my work area: Communication Assistance for Law Enforcement Act (CALEA). Lead a team of 3 engineers that was involved in debugging traffic related issues of live customer networks. Primary contact point of Call Processing group for customers that have live installation of Alcatel NGN product A5020. Worked as a Module Owner for Subscriber Profile Manager across the all global development sites. Worked on Client Site at Orvault France for 3 months in Real Network environment for Orange France, Orange UK to support the project till acceptance. page 1 of 5
2 Rajat, page 2 of 5 REFEREED JOURNAL PUBLICATIONS Mehrotra Rajat, Srivastava Srishti, Banicescu Ioana, and Abdelwahed Sherif. Towards an Autonomic Performance Management Approach for Cloud Brokers Using an Interaction Balance Based Methodology, The International Journal of Grid Computing and escience (Future Generation Computer Systems), Elsevier. (in review). Mehrotra Rajat, and Abdelwahed Sherif, Towards Autonomic Performance Management of Large Scale Data Centers using Interaction Balance Principles, Journal of Cluster Computing, Springer Publishers, pp.1-21, Dubey Abhishek, Mehrotra Rajat, Abdelwahed Sherif, and Tantawi Asser. Performance Modeling of Distributed MultiTier Enterprise Systems, ACM Performance Evaluation Review, vol. 37, no. 2, pp.9-11, September, BOOK CHAPTERS Mehrotra Rajat, Banicescu Ioana, Srivastava Srishti, and Abdelwahed Sherif. A Power-Aware Autonomic Approach for Managing Scientific Applications in a Data Center Environment in Handbook on Data Centers. Springer (in publication). Mehrotra Rajat, Dubey Abhishek, Abdelwahed Sherif, and Tantawi Asser. A Power-Aware Modeling and Autonomic Management Framework for Distributed Computing Systems in Handbook of Energy-Aware and Green Computing. Chapman and Hall/CRC Press Taylor and Francis Group LLC. Dec CONFERENCE PUBLICATIONS Mehrotra Rajat and Abdelwahed Sherif, A Component-Based Framework for Autonomic Performance Management in a Distributed Computing Environment, The 11th ACS/IEEE International Conference on Computer Systems and Applications (AICCSA 2014) November 10-13, 2014, Doha, Qatar (in publication). Mehrotra Rajat, Srivastava Srishti, Banicescu Ioana, and Abdelwahed Sherif. An Interaction Balance Based Approach for Autonomic Performance Management in a Cloud Computing Environment Workshop on Adaptive Resource Management and Scheduling for Cloud Computing, (ARMS- CC), July 15-19, 2014, Paris, France (in publication). Mehrotra Rajat, Abdelwahed Sherif, and Erradi Abdelkarim, A Distributed Control Approach for Autonomic Performance Management in Cloud Computing Environment, 6th IEEE/ACM International Conference on Utility and Cloud Computing (UCC-13), pp , 2013, Dresden, Germany. Mehrotra Rajat, Dubey Abhishek, Abdelwahed Sherif, and Rowland Krisa, RFDMon: A Real- Time and Fault-Tolerant Distributed System Monitoring Approach. The Eighth International Conference on Autonomic and Autonomous Systems (ICAS-2012). Chen Qian, Mehrotra Rajat, Dubey Abhishek, Abdelwahed Sherif, and Rowland Krisa, On State of The Art in Virtual Machine Security. IEEE South East Conference (SouthEastCon-2012). Mehrotra Rajat, Banicescu Ioana, and Srivastava Srishti, A Utility Based Power-Aware Autonomic Approach for Running Scientific Applications. 13th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing,2012. Monceaux Weston, Deland E. Evans, Keith N. Rappold, Cary D. Butler, Abdelwahed Sherif, Mehrotra Rajat, and Dubey Abhishek, Implementing Autonomic Computing Methods to Improve Attack Resilience in Web Services DoD High Performance Computing Modernization Program Users Group Conference, pp , Mehrotra Rajat, Dubey Abhishek, Abdelwahed Sherif, and Monceaux Weston, Large Scale Monitoring and Online Analysis in a Distributed Virtualized Environment. IEEE International Conference and Workshop on Engineering of Autonomic and Autonomous Systems, 0:1-9, Mehrotra Rajat, Dubey Abhishek, Abdelwahed Sherif, and Tantawi Asser, Integrated Monitoring and Control for Performance Management of Distributed Enterprise Systems 18th Annual IEEE/ACM International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems, pp , 2010.
3 Rajat, page 3 of 5 TECHNICAL REPORTS Mehrotra Rajat and Abdelwahed Sherif, Application of Interaction Balance Principle for Optimal Control in the Distributed Web Service Deployment. Technical Report, MSU-ECE-13, Mississippi State University, May Jia Rui, Mehrotra Rajat, Abdelwahed Sherif, and Erradi Abdelkarim, An Implementation of a Real- Time Monitoring Framework to Support Fault Management in Distributed Systems. Technical Report, MSU-ECE , Mississippi State University, May Mehrotra Rajat, Dubey Abhishek, and Abdelwahed Sherif, RFDMon: A Real-Time and Fault- Tolerant Distributed System Monitoring Approach. Technical report for Institute for Software Integrated Systems (ISIS), Vanderbilt University, Mehrotra Rajat, Dubey Abhishek., Abdelwahed Sherif, and Tantawi Asser. Model Identification for Performance Management of Distributed Enterprise Systems. Technical report for Institute for Software Integrated Systems (ISIS), Vanderbilt University, POSTERS Srivastava Srishti, Mehrotra Rajat, Banicescu Ioana, and Abdelwahed Sherif, A Model-based Framework for Autonomic Performance Management of Cloud Computing Systems. 15th SIAM Conference on Parallel Processing for Scientific Computing, Savannah, GA, Feb Mehrotra Rajat, Qian Chen, Dubey Abhishek, Abdelwahed Sherif, and Rowland Krisa Automated Model-based Security Management of Web Services. 20th USENIX Security Symposium, San Francisco, CA, Aug Mehrotra Rajat, Dubey Abhishek, Abdelwahed Sherif, and Tantawi Asser, Integrated Monitoring and Control for Performance Management of Distributed Enterprise Systems 18th Annual IEEE/ACM International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems, pp , HARDWARE AND SOFTWARE SKILLS Operating Systems: Windows-95/98/NT/2000/XP, Linux, Sun Solaris 8/10, Unix. Programming Methodologies/Framework: OOP, SOA, REST, MVC, and UML. Languages: C, C++, Java, Python, UNIX Shell scripts, Java Script, Ruby on Rails, and PHP. Big Data Technologies: Hadoop (MapReduce), NoSQL (MongoDB), and Predictive Modeling. Middleware Technologies: Data Distribution Services (DDS) OpenSplice, CORBA, SUN RPC, and MPI. Tools and Utilities: Matlab / Simulink, Sun Studio, Workshop, Sun Memory Monitor, and DTrace. Analysis Tools: UPPAAL, SMV, and Spin. Configuration Management: CVS, Clear Case, SVN, and GIT. Virtualization Technologies: Xen, VMWare vsphere, and VMWare ESX Server. PROJECTS 1. Dept. of Computer Science, University of Virginia. PERFECT: Power Efficiency Revolution For Embedded Computing Technologies: This project is aimed for developing future generation extremely power efficient and fault tolerant embedded systems that can be utilized in mobile and aerial systems. It develops an integrated resilience framework that consists of design options from all layers: from hardware level to algorithm level. In this project, I am responsible for developing an analytical modeling tool that explores the possible combination of various resilient techniques with respect to improvement in soft error rate of the system. This tool performs design state exploration of system design while choosing multiple control options of resilience techniques across the different system abstraction layers. These layers include algorithms based fault tolerance, recovery methods, micro-architecture level logic parity, and circuit level hardening.
4 Rajat, page 4 of 5 2. NSF Center for Cloud and Autonomic Computing, Mississippi State University. A. Model-Based Autonomic Performance Management of Distributed Enterprise Systems and Applications: This project has developed a component-based framework that can be utilized to deploy various key elements of a performance management system for distributed multi-tier enterprise application management in normal and off-normal operating conditions. The developed framework provides flexibility to the administrators to define multi-dimensional QoS objectives with their relative priorities, the initial settings of the distributed application, monitoring components placement in a distributed manner, interaction among the components, and fault prediction/isolation in case of component failure. B. Performance Modelling of Distributed Multi-Tier Enterprise Systems: This project was an effort to accurately capture and model the system dynamics of a distributed multi-tier enterprise system in a highly dynamic and time-varying environment. This model was developed using dynamic regression and queuing modelling techniques. This model was used later within a control structure which effectively optimized the system parameters of a multi-tier application for a given set of objectives (eg. response time & power consumption). In this project, Daytrader application hosted on IBM Websphere Application Server Community Edition was used with MySQL to deploy enterprise systems. C. A Model-Based Framework for Autonomic Performance Management of Cloud Computing Systems: This project is an effort to employ a model-based management approach for enabling self-management attributes in a cloud management system such that the overall performance of the system (in terms of QoS, availability, and reliability) meets the SLA specifications. The proposed management approach will guarantee a certain level of application performance for applications hosted in cloud environment by developing the behavioural model of the hosted application with respect to available resources and control decisions. Additionally, it will identify and project the possible faults in the system and provide recommendations for placement of the applications and number of instances to be allocated based on the application QoS requirements. D. Power-Aware Autonomic Approach for HPC Systems: This project was an effort to develop a utility based power-aware approach that uses a modelbased control theoretic framework for executing scientific applications. The approach and related results indicate that the performance and the power requirements of the system can dynamically be adjusted, while maintaining the predefined quality of service (QoS) goals in terms of deadline of execution and power consumption of the HPC system, even in the presence of computational resource related perturbations. This approach is autonomic, performance directed, dynamically controlled, and independent of (do not interfere with) the execution of the scientific application. E. Automated Model-based Security Management of Web Services: This project will accurately develop a model-based limited look-ahead predictive control framework with added security monitoring features and multi dimensional QoS objectives for distributed Multi- Tier enterprise systems from outside security risks. This model will be developed by experimentally identifying the various system and network parameters impacting the performance and security of the system, defining the dependency relationship among these parameters, and then using that relationship to develop the intrusion detection and performance model structure of the system. 3. Institute for Software Integrated Systems, Vanderbilt University. A Real-Time and Fault-Tolerant Distributed System Monitoring Approach:
5 Rajat, page 5 of 5 This project is an extension of work performed at Vanderbilt University for Fermi Batavia, IL. During this project, a distributed monitoring system was developed for monitoring of clusters hosted at Fermi Lab with respect to system resources and hardware health. This monitoring system utilizes data distribution services (DDS) for communication among distributed nodes and avionics operating system guidelines for fault isolation in the framework. As an extension of this work, fault-tolerant and self-configuration features were added to the monitoring system. In addition, scientific application and web application performance monitoring capabilities were added to the existing developed monitoring system. 4. Alcatel-Lucent Wireless R&D Division, Noida, India. A. H.248 (MEGACO) Implementation for Wireless Call Server: This project covered all the changes required in the existing system to adapt H.248 MEGACO standards on the interface between Alcatel Wireless Call Server (WCS A5020) and Alcatel Media Gateway (A7570). All the expected modifications were related to the Supplementary Services, CALEA, Handover, COT, MPTY and their interaction among themselves. This project was meant to remove some proprietary features of previous Tekelec MGW (A7540). I have also worked for stabilization of different previous releases, which were already delivered to customers across the Globe (AT&T, T-Mobile USA). Moreover, I worked on memory management aspects and real time traffic support. B. Subscriber Profile Manager: I have worked as a Global Module Owner of Subscriber Profile Manager module of Wireless Call Server, which keeps track of all subscribed services, related to subscriber and interaction of WCS with out-side world entities for CAMEL/IN services. In this module, I worked on MAP and CAP interfaces to develop core features, such as CLIP-COLP, Location Number and some enhanced features like Supercharger to make WCS complied with the 3GPP specifications. HONORS AND ACTIVITIES Phi kappa Phi, Member, Present Eta Kappa Nu, Member, 2009 Present IEEE, Student Member, 2008 Present President of MSU Cricket Club, Mississippi State University for academic year Vice-President of MSU Cricket Club, Mississippi State University for academic year PROFESSIONAL SERVICES Reviewer for Journal of Cluster Computing, Springer Publications. Reviewer for IEEE Transactions on Cloud Computing. Technical Program Committee Member of IEEE International Workshop on Management Issues in SDN, SDI and NFV (MISSION 2015), London, UK, Technical Program Committee Member of IEEE International Workshop on Manageability and Security of Network Function Virtualization and Software Defined Network (MASONS), Technical Program Committee Member of International Conference on Computing in Mechanical Engineering (ICCME), Aug Technical Program Committee Member of International Conference on Information and Communication Technologies for Disaster Management (ICT-DM2015). Technical Program Committee Member of ACS/IEEE International Conference on Computer Systems and Applications (AICCSA 2014). Reviewer for International Conference on Autonomic and Autonomous Systems (ICAS). Reviewer for Cluster Computing and the Grid (CCGRID) REFERENCES AVAILABLE TO CONTACT Available upon request.