Last update: June, 2022

Platforms for Big-data Workloads

1. Distributed Stream Processing Systems

   • [Topics] Streaming Systems, In-memory Computing, Admission Control, Query Planning, Query Scheduling, Heterogeneous Computing (CPU-GPU)

   • [Frameworks/Tools] Apache Spark, Apache Kafka, Apache Cassandra, Apache Hadoop File Systems, Network File Systems, NVIDIA CUDA, NVIDIA Rapids

   • [Results] 1 project /w industry, 5 publications

2. Cluster Resource Management

   • [Topics] Heterogeneous Computing (CPU-GPU), Container Scheduling

   • [Frameworks/Tools] Kubernetes, Docker Container, Remote Procedure Call, NVIDIA CUDA

   • [Results] 1 coursework project, 1 publication

3. Large-scale DBMS

   • [Topics] LSM-tree Based DBMS, SQL/noSQL Query Optimization, Spatial DBMS

   • [Frameworks/Tools] Apache Cassandra, RocksDB, LevelDB, Microsoft SQL Sever (Spatial Data)

   • [Results] 1 project /w industry, 1 coursework project

Distributed Systems & Architectures

1. Disaggregated Systems

   • [Topics] Disaggregated Memory Systems

   • [Frameworks/Tools] RDMA with Infiniband, DRAM Management, NVM Management, Extended Page Tables (Intel), KVM Hypervisor, QEMU, Linux Kernel Modules

   • [Results] 1 project /w industry, currently working on for a paper submission

2. Operating Systems

   • [Topics] Hypervisor, Memory Management, File Systems

   • [Frameworks/Tools] Linux Kernel Programming, Linux Kernel Modules

   • [Results] 1 project /w industry, 3 coursework projects, currently working on for a paper submission

3. Systems for Data Centers New! (PhD Research Area)

   • [Topics] Tiered Memory Management (Local-Remote/DRAM-NVM-HBM), High Performance Communication (RDMA/Programmable Switches/smartNICs), Emerging Memory Computing (In-memory/Near-memory/CXL)

   • [Frameworks/Tools] - (TBA)

   • [Results] - (TBA)