Mitch Gusat

Pronouns

He/Him/His

Title

Research Scientist
Mitch Gusat

Bio

Mitch Gusat is a research scientist at IBM Research Zurich. His current focus is on hi-dimension timeseries data mining (Cloud KPIs, logs, u-services), eXplainability, Causality, forecasting and ML DL/NN meta-models for on/off-line Anomaly Detection for Performance and Security apps in the Cloud, e.g., high-D streaming systems (large graphs with 100s-1000s KPIs per node) with causality constraints / root cause analysis. 

Until recently, Mitch was busy with performance analysis, quantitative modeling and optimization for large distributed systems - ie., stability conditions, saturation scenarios, congestive bottlenecks - detection, prediction and control. Here he worked on the feedback control and performance modeling of large distributed systems and lossless datacenter networks beyond 100Gbps, including their flow and congestion control, adaptive routing, workload optimization and monitoring. In this area he has contributed to the standardization of Converged Enhanced Ethernet/802 DCB, InfiniBand and RapidIO. Mitch's other research interests include switching, Software Defined Networking, HPC interconnection networks, shared (virtual) memory, real-time scheduling, high performance protocols and IO acceleration. Prior to this, Mitch has worked on cache coherent HPC systems, packet switching, message passing networks, real-time distributed systems, interconnection networks for cache-coherent snooping SMP and cc-NUMA systems. He was a Research Associate at the University of Toronto where he contributed to the design and construction of NUMAchine, a 64-way cache-coherent computer.

Mitch holds Masters in Computer Engineering and Electrical Engineering from the University of Toronto and Politechnical University of Timisoara, respectively, where he was also a researcher designig multiprocessor systems, video interfaces, algorithms and image processors for Nuclear Cardiology. He is member of ACM, IEEE, and holds dozens of patents. In addition to his Technical Program Committee activities for IEEE/ACM conferences, Mitch is reviewing technical papers for a number of IEEE and ACM journals, as well as networking and computer architecture books. He is also advising Master and PhD students from several European universities and has participated in several European Horizon projects with consortia partners from academia and industry.

Projects

FoundationModels_Blog_Leadspace_@2x.jpg

AI for IT Infrastructure

AI to improve infrastructure efficiency and user productivity, and to extract valuable insights from business data.

Top collaborators