Short bio:I am a final-year Ph.D. student advised by Prof. Daniel Sanchez at MIT CSAIL. My research focuses on computer system and architecture. Specifically, I work on memory hierarchy design, resource management in multi-core systems, and software/hardware co-optimization.
I will be joining Nvidia Research as a Research Scientist in 2019 summer.
About my research:Modern processors are much faster than their main memory. A multiply-and-add in a processor is >10x faster and takes >100x less energy than an access to its main memory. This makes applications stall and wait for data most of the time and spend a huge amount of energy on accessing it. Without a drastic reduction in data movement, memory accesses will limit the scalability of future systems.
My research focuses on two directions to reduce data movement:
1. object-based memory hierarchies that better suited for modern languages, and
2. specialized memory hierarchies that automatically adapt to workloads
My research spans across hardware and software, including architecture, automatic memory management, system runtime, and optimization algorithms.
Apart from my main research, I have a broad interest in computer system performance, from warehouse-scale datacenters (workload scheduling in datacenters) to nano-scale devices (emerging memory technologies).
Last update: April 2019
Compress Objects, Not Cache Lines: An Object-Based Compressed Memory Hierarchy
Po-An Tsai and Daniel Sanchez.
The 24th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-24), April 2019.
[paper] [talk] [poster] [lightning] [MIT news]
Rethinking the Memory Hierarchy for Modern Languages
Adaptive Scheduling for Systems with Asymmetric Memory Hierarchies
KPart: A Hybrid Cache Partitioning-Sharing Technique for Commodity Multicores
Nosayba El-Sayed, Anurag Mukkara, Po-An Tsai, Harshad Kasture, Xiaosong Ma, and Daniel Sanchez.
The 24th Intl' Symposium on High Performance Computer Architecture (HPCA-24), February 2018.
[paper] [talk] [code]
Nexus: A New Approach to Replication in Distributed Shared Caches
Jenga: Software-Defined Cache Hierarchies
Scaling Distributed Cache Hierarchies through Computation and Data Co-Scheduling
Hybrid Path-Diversity-Aware Adaptive Routing with Latency Prediction Model in Network-on-Chip Systems
Po-An Tsai, Yu-Hsin Kuo, En-Jui Chang, and An-Yeu Wu.
2013 International Symposium on VLSI Design, Automation and Test, (VLSI-DAT), March 2013.
Path-Diversity-Aware Adaptive Routing in Network-on-Chip Systems
Yu-Hsin Kuo, Po-An Tsai, Hao-Ping Ho, En-Jui Chang, Hsien-Kai Hsin, and An-Yeu Wu.
The 6th International Symposium on Embedded Multicore SoCs (MCSoC), September 2012.
Office: 32 vassar street G888, Cambridge, MA, 02139