I am currently an associate professor
Department of Electrical and Computer
Engineering and a cooperating faculty of
the Department of Computer Science
at University of
I am now leading the Extreme Storage &
Computer Architecture Laboratory.
interested in diverse research topics that allow
applications and programmers to more
efficiently use modern heterogeneous
hardware components. Together with my
students, our most recent work has demonstrated the potential of using emerging AI/ML accelerators (e.g., Google's Edge TPUs) in improving the performance of
non-AI/ML workloads through our lastest GPTPU framework [GitHub]. We also showed
how intelligent storage devices can help improve
performance, power and energy for
Our serious of work on intelligent storage systems has been recognized by two best paper nominations from IEEE/ACM International Symposium on Microarchitecture in 2021 and 2019, IEEE Micro "Top Picks from the 2019 Computer Architecture
Conferences" (IEEE MICRO Top Picks 2020) and Facebook
Research Award, 2018. In addition, we also applied our knowledge in optimizing storage systems to wireless network system stacks and developed the OpenUVR project [GitHub] that enables high-quality, untethered VR experience on commodity hardware components and won the outstanding paper award in RTAS 2021.
Prior to joining UCR, I served as an assistant professor for the Department of Computer Science and the Department of Electrical and Computer Engineering at NC State University . I was a PostDoc of the Non-volatile Systems Laboratory and a lecturer of the Department of Computer Science and Engineering at University of California, San Diego with Professor Steven Swanson. My thesis work with Professor Dean Tullsen is data-triggered threads, also selected by IEEE MICRO Top Picks in 2012.
We built a full-stack system and revised the algorithms of general-purpose applications to demonstrate the potential of using emerging AI/ML accelerators, essentially matrix processors to improve performance. The resulting systems demonstrate significant speedups and energy savings with edge TPUs that are just about USD 25. (More)
We built Intelligent Data Storage Systems, including Summarizer and Morpheus, to demonstrate the potential of exposing the processing power inside the controllers of modern non-volatile storage devices. The resulting systems demonstrate significant speedups in GPGPU applications, database systems, and the potential of machine learning applications. (More)
As the slowdown of Moore's Law and the discontinuation of Dennard Scailing, big-data applications (e.g. machine learning, data analytics, scientific computing, and etc.,.) must rely on heterogeneous computing units, including GPUs, FPGAs or ASICs as well as heterogeneous data storage, including DRAM, NVRAM, flash memory, to complete their own tasks. We built systems that make more efficient use of heterogeneous system components to accelerate applications. The resulting system can accelerate (More)
With hardware accelerators improving the latency in computation, the system software stack that were traditionally underrated in designing applications becomes more critical. In ESCAL, we focus on those underrated bottlenecks to achieve significant performance improvement without using new hardware. The most recent example is the OpenUVR system, where we eliminate unnecessary memory copies and allow the VR system loop to complete within 14 ms latency with just modern desktop PC, existing WiFi network links, raspberry Pi 4b+ and an HDMI compatible head mount display.(More)
The explosive demand on AI/ML workloads drive the emergence of AI/ML accelerators, including commercialized NVIDIA Tensor Cores and Google TPUs. These AI/ML accelerators are essentially matrix processors and are theoretically helpful to any application with matrix operations. This project bridges the missing system/architecture/programming language support in democratizing AI/ML accelerators. As matrix operations are conventionally inefficient, this project also revises the core algorithm in compute kernels to better utilize operators of AI/ML accelerators.
As parallel computer architectures significantly shrinking the execution time in compute kernels, the performance bottlenecks of applications shift to the rest of part of execution, including data movement, object deserialization/serialization as well as other software overheads in managing data storage. To address this new bottleneck, the best approach is to not move data and endow storage devices with new roles.
Morpheus is one of the very first research project that implements this concept in real systems. We utilize existing, commercially available hardware components to build the Morpheus-SSD. The Morpheus model not only speeds up a set of heterogeneous computing applications by 1.32x, but also allows these applications to better utilize emerging data transfer methods that can send data directly to the GPU via peer-to-peer to further achieve 1.39x speedup. Summarizer further provides mechanisms to dynamically adjust the workload between the host and intelligent SSDs, making more efficient use of all computing units in a system and boost the performance of big data analytics. This line of research also helps Hung-Wei's team receive Facebook research award, 2018.
As the discontinuation of Dannard scaling and Moore's Law, computers become heterogeneous. However, moving data among heterogeneous computing units and storage devices becomes an emerging bottleneck in these systems.
My research proposes the "Hippogriff" system that revisits the programming model of moving data in heterogeneous computer systems. Instead of using the conventional CPU-centric, programmer-specified methods, the Hippogriff system simplifies the application interface and provide a middle layer to efficiently handle the data movement. We also implemented peer-to-peer data transfer between the GPU and the SSD in the Hippogriff system.
The preliminary result demonstrates 46% performance gain by applying Hippogriff to a set of rodinia GPU applications. For highly optimized GPU MapReduce framework, Hippogriff still demonstrates up to 27% performance gain.
With hardware accelerators improving the latency in computation, the system software stack that were traditionally underrated in designing applications becomes more critical. In ESCAL, we focus on those underrated bottlenecks to achieve significant performance improvement without using new hardware. The most recent example is the OpenUVR system, where we eliminate unnecessary memory copies and allow the VR system loop to complete within 14 ms latency with just modern desktop PC, existing WiFi network links, raspberry Pi 4b+ and an HDMI compatible head mount display.