Anthony Kougkas
Research Assistant Professor
Distributed and Parallel Storage - I/O
Distributed and Parallel Storage - I/O
High-Performance Storage and Exascale I/O.
Director of I/O research at the SCS Lab @Illinois Tech.
Technical Lead on the NSF-funded Hermes project (NSF OCI-1835764).
Co-Principal Investigator on the NSF-funded ChronoLog project (NSF CSSI-2104013).
"Accelerating I/O Using Data Labels: A Contention-aware, Multi-tiered, Scalable, and Distributed I/O Platform", May 2019.
I/O is now the biggest challenge in modern supercomputers!
Storage systems are called to efficiently support both compute- and data-intensive workloads. As we move towards the exascale era, most of these storage systems face significant challenges in performance, scalability, complexity, and limited metadata services, creating an I/O bottleneck that leads to less scientific productivity.
Unification of HPC and Data Analytics is essential to address a wide spectrum of major research domains.
A new memory and storage hierarchy is a reality and modern storage needs to support advanced I/O buffering.
A new data representation and runtime to support a wide variety of of confliicting I/O workloads.
A scalable, parallel, and high-performance solution that can
support a diverse set of log store requirements.
10 West 31st Street
Chicago, IL, 60616
Email: akougkas@iit.edu
Work: +1 312 567 5023
Phone: +1 312 493 9389
Website was created with Mobirise site theme
FOLLOW ME!