I'm Mingliang LIU, 刘明亮 in Chinese. I love computer programming as well as research.
I'm working on compilers, high performance computing (HPC) and distributed systems.
This webpage has been finalized after my graduation from Tsinghua University.
No news is good news.
Elected as Apache Hadoop Committer. See my Hadoop blog. Aug 2016
APPRIME was accepted by SIGMETRICS as a regular paper (32/239). Feb 2015
ACIC (extended) was published by IEEE TPDS journal. Dec 2014
Graduated from Tsinghua University with PhD degree. Jul 2014
ACIC is accepted by SC 2013 as a technical paper (92/457). Jul 2013
The idea of the IIO is to obtain a program slice through static analysis, and to generate a compilable and human-readable benchmark from it. We generate the benchmark from original parallel application by reducing those irrelevant statements, while preserving all the variables and statements in the original program relevant to the spatial and volume attributes.
It is very challenging and costly to obtain high-fidelity benchmarks reflecting the scale and complexity of state-of-the-art parallel applications. Taking as input standard communication-I/O traces of an application’s execution, APPRIME couples accurate automatic phase identification with statistical regeneration of event parameters to create compact, portable, and to some degree reconfigurable parallel application benchmarks, which retain the original applications’ performance characteristics, in particular the relative performance across platforms.
Dynamic program slicing is a technique that can precisely determine which instructions affected a particular value in a single execution of a program. This project was firstly developed by Sahoo, Swarup Kumar and John Criswell under Vikram S. Adve from UIUC. It was selected by the Google Summer of Code (GSoC) 2013, under its umbrella project LLVM.
ACIC takes a given application running on a given cloud platform, and automatically searches for optimized I/O system configurations. ACIC utilizes machine learning models to perform black-box performance/cost predictions. To tackle the high-dimensional parameter exploration space unique to cloud platforms, we enable affordable, reusable, and incremental training guided by Plackett and Burman Matrices. Results with four representative applications indicate that ACIC consistently identifies near-optimal configurations among a large group of candidate settings.
PhD in Computer Science and Technology
Aug 2007 - Jul 2014
B.S. in Information Systems
Sep 2002 - Jul 2006
Scholarship for overall excellence -
1st Class, Tsinghua University
Scholarship for overall excellence - 3rd Class, Tsinghua University Oct 2012
Scholarship for overall excellence - 1st Class, Tsinghua University Oct 2011
Scholarship for overall excellence - 2nd Class, Tsinghua University Oct 2008
Outstanding student cadres - Tsinghua University May 2008
National scholarship - 2nd class, Three Gorges University Sep 2004