Building Next-generation Cloud Systems
ABC-Lab aims to design, analyze, and evaluate large-scale distributed cloud systems with a focus on easier management, better QoS, higher efficiency, and more predictable service.
This project implement distributed Deep Learning on Big Data stacks enabling easy integration with other Big Data components and guaranting local data access.
A hybrid NVM based in-memory computing architecture is proposed with effective data sharing and communication strategy for ML applications.
Store, manage, and process data records in batch based streaming systems to provide a high service availability and low latency are main concerns in this research domian.
This project develops resource provisioning approaches and workload placement policies, which aim to maximize system performance by prioritizing power budget of workloads.