Below is a list of recent research publications and brief summaries. A complete list can be found in my CV, on DBLP and Google scholar.
- J. Wen, L. Ren, F. Yan, D. J. Dubois, G. Casale, and E. Smirni. How to Supercharge the Amazon T2: Observations and Suggestions, IEEE CLOUD 2017, to appear in June 2017.
- The paper proposes a methodology to increase the performance benefits of T2 instances on Amazon EC2.
- G. Casale. Accelerating Performance Inference over Closed Systems by Asymptotic Methods. ACM SIGMETRICS, Best paper award, June 2017.
- Novel exact and asymptotic expressions for the normalising constant of closed multiclass queueing networks enabling optimisation-based statistical inference.
- An extended abstract is available here.
- Presentation available here.
- J. F. Pérez and G. Casale. LINE: Evaluating Software Applications in Unreliable Environments. IEEE Transactions on Reliability, 2017.
- A tool for layered queueing network reliability analysis in stochastically-evolving environments
- K. Molka, G. Casale. Energy-Efficient Resource Allocation and Provisioning for In-Memory Database Clusters. IFIP/IEEE IM, May 2017. Best student paper award.
- An energy-aware consolidation algorithm for SAP HANA in-memory database clusters
- D. Dubois, G. Casale. Performance Prediction for Burstable Cloud Resources. VALUETOOLS 2016 (tool paper).
- A tool for predicting performance metrics such as resource utilizations, request response times, and credit usage in burstable resource.
- S. Dipietro, G. Casale, G. Serazzi. A Queueing Network Model for Performance Prediction of Apache Cassandra. ICST VALUETOOLS 2016.
- A multiclass simulation model for Apache Cassandra capacity planning.
- G. Casale, A. Sansottera, P. Cremonesi. Compact Markov-Modulated Models for Multiclass Trace Fitting, INFORMS European J. of Operational Research (EJOR), full paper, 2016.
- A fitting methodology for marked Markovian arrival processes (MMAPs) and a method to avoid state-space explosion in their superposition.
- W. Wang, G. Casale, C. Sutton. A Bayesian Approach to Parameter Inference in Queueing Networks, ACM Trans. on Modeling and Computer Simulation (TOMACS), 2016.
- A method to obtain service demands in closed product-form queueing networks based on observations of unfinished work at resources (queue-length samples).
- D. J. Dubois, G. Casale. OptiSpot: Minimizing Application Deployment Cost using Spot Cloud Resources, Cluster Computing, Springer, 2016
- An heuristic to minimize deployment costs on Amazon EC2 spot VMs, taking into account the workload and the topology of the application and quality-of-service requirements.
- G. Casale, V. de Nitto-Personé, E. Smirni. QRF: An Optimization-Based Framework for Evaluating Complex Stochastic Networks, ACM Trans. on Modeling and Computer Simulation (TOMACS), 26(3):15, Jan 2016.
- Optimization-based algorithm to bound performance measures in queueing networks with bursty workloads and finite capacity buffers.
- R. Osman, J. F. Peréz, G. Casale. Quantifying the Impact of Replication on the Quality-of-Service in Cloud Databases, in Proc. of IEEE QRS, Aug 2016
- A method to characterize the performance of Amazon Relational Database Service (RDS) deployments when scaling the underpinning compute resources.
- D. Dubois, C. Trubiani, G. Casale. Model-driven Application Refactoring to Minimize Deployment Costs in Preemptible Cloud Resources, in Proc. of IEEE CLOUD, Jun 2016.
- A method to improve the deployment of a cloud application on Amazon EC2 spot instances through atomic refactoring actions.
- W. Wang, G. Casale, A. Kattepur, M. Nambiar. Maximum Likelihood Estimation of Closed Queueing Network Demands from Queue Length Data, in Proc. of ACM/SPEC ICPE, Mar 2016. Best Paper Award.
- Maximum-likelihood estimators for service demands in closed queueing network models, including those with load-dependent stations.
- J.F. Pérez, G. Casale, S. Pacheco-Sanchez. Estimating Computational Requirements in Multi-Threaded Applications, IEEE Trans. on Software Engineering, 41(3):264–278, Mar 2015
- A method to estimate CPU requirements of requests using response time data.
- S. Spinner, G. Casale, F. Brosig, S. Kounev. Evaluating Approaches to Resource Demand Estimation, Performance Evaluation, Elsevier 92:51-71, Oct 2015.
- A survey and experimental comparison on demand estimation methods used in capacity planning, resource management and performance modelling.
- G. Casale, J. F. Pérez, W. Wang. QD-AMVA: Evaluating Systems with QueueDependent Service Requirements, Performance Evaluation, Elsevier, 91:80-98, Sep 2015. Also presented at IFIP Performance 2015.
- An efficient heuristic for the analysis of closed queueing networks where stations have service rates that depend on the current number or mix of requests.
- K. Molka, G. Casale. Experiments or Simulation? A Characterization of Evaluation Methods for In-Memory Databases, in Proc. of IFIP/IEEE CNSM, Nov 2015.
- A comparison of the relative merits of design-of-experiments and simulation in the performance evaluation of SAP HANA.
- D. J. Dubois, G. Casale. Autonomic Provisioning and Application Mapping on Spot Cloud Resources, in Proc. of IEEE ICCAC, Sep 2015. Best Paper Award.
- An heuristic to minimize deployment costs on Amazon EC2 spot VMs, taking into account the workload and topology of the application and subject to quality of services requirements.
- J. Wen, L. Lu, G. Casale, E. Smirni. Less can be More: micro-Managing VMs in Amazon EC2, in Proc. of IEEE CLOUD, Jun 2015. Best paper award.
- A characterization of micro-VMs on Amazon EC2 and a proposal for CPU throttling methods.
2003-2014: see CV