Source author record

Donghong Wu

Donghong Wu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

astro-ph.IM astro-ph.SR Hardware Architecture Operating Systems Performance

Catalog footprint

What is connected

2works

5topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2015arXiv

Eclipsing Binaries From the CSTAR Project at Dome A, Antarctica

The Chinese Small Telescope ARray (CSTAR) has observed an area around the Celestial South Pole at Dome A since 2008. About $20,000$ light curves in the i band were obtained lasting from March to July, 2008. The photometric precision achieves about 4 mmag at i = 7.5 and 20 mmag at i = 12 within a 30 s exposure time. These light curves are analyzed using Lomb--Scargle, Phase Dispersion Minimization, and Box Least Squares methods to search for periodic signals. False positives may appear as a variable signature caused by contaminating stars and the observation mode of CSTAR. Therefore the period and position of each variable candidate are checked to eliminate false positives. Eclipsing binaries are removed by visual inspection, frequency spectrum analysis and locally linear embedding technique. We identify 53 eclipsing binaries in the field of view of CSTAR, containing 24 detached binaries, 8 semi-detached binaries, 18 contact binaries, and 3 ellipsoidal variables. To derive the parameters of these binaries, we use the Eclipsing Binaries via Artificial Intelligence (EBAI) method. The primary and the secondary eclipse timing variations (ETVs) for semi-detached and contact systems are analyzed. Correlated primary and secondary ETVs confirmed by false alarm tests may indicate an unseen perturbing companion. Through ETV analysis, we identify two triple systems (CSTAR J084612.64-883342.9 and CSTAR J220502.55-895206.7). The orbital parameters of the third body in CSTAR J220502.55-895206.7 are derived using a simple dynamical model.

preprint2014arXiv

Rank-Aware Dynamic Migrations and Adaptive Demotions for DRAM Power Management

Modern DRAM architectures allow a number of low-power states on individual memory ranks for advanced power management. Many previous studies have taken advantage of demotions on low-power states for energy saving. However, most of the demotion schemes are statically performed on a limited number of pre-selected low-power states, and are suboptimal for different workloads and memory architectures. Even worse, the idle periods are often too short for effective power state transitions, especially for memory intensive applications. Wrong decisions on power state transition incur significant energy and delay penalties. In this paper, we propose a novel memory system design named RAMZzz with rank-aware energy saving optimizations including dynamic page migrations and adaptive demotions. Specifically, we group the pages with similar access locality into the same rank with dynamic page migrations. Ranks have their hotness: hot ranks are kept busy for high utilization and cold ranks can have more lengthy idle periods for power state transitions. We further develop adaptive state demotions by considering all low-power states for each rank and a prediction model to estimate the power-down timeout among states. We experimentally compare our algorithm with other energy saving policies with cycle-accurate simulation. Experiments with benchmark workloads show that RAMZzz achieves significant improvement on energy-delay2 and energy consumption over other energy saving techniques.