Browsing by Author "Wen, Charles H-P."

Now showing 1 - 3 of 3

CASTA: CUDA-Accelerated Static Timing Analysis for VLSI Designs
(2014)
Wang, Hunta H. -W.
;
Lin, Louis Y. -Z.
;
Huang, Ryan H-M.
;
Wen, Charles H-P.
;
資訊工程學系
General-purpose computing on graphics processing unit (GPGPU) enables the possibility of parallel computing for Static Timing Analysis (STA) of VLSI designs. However, memory access and synchronization between massively many cores become challenges to parallelizing STA. In this work, we developed a fast CUDA-Accelerated STA engine (named CASTA) that incorporates four novel techniques including Table-Index Remapping (TIR), Texture-Accelerated Rendering (TAR), Cell Levelization & Type Sorting (CLTS) and Timing-Table Restructuring (TTR) to enable high parallelism. Cell Levelization & Type Sorting (CLTS) levelizes cells and sort their types in order to efficiently access the same timing library. Timing-Table Restructuring (TTR) modifies the data structure for timing signals of cells to increase memory throughput. Table-Index Remapping (TIR) re-maps the axes of timing tables to retrieve data more efficiently while Texture-Accelerated Rendering (TAR) expands look-up tables (LUTs) to avoid extrapolation and stores LUTs in the texture for speed. As a result, our experimental result indicates that CASTA successfully enables high parallelism and outperforms a commercial tool by a three-order speedup on average over several benchmark circuits.
Flow-and-VM Migration for Optmizing Throughput and Energy in SDN-based Cloud Datacenter
(2013-01-01)
Lin, Wei-Chu
;
Liao, Chien-Hui
;
Kuo, Kuan-Tsen
;
Wen, Charles H-P.
;
電機資訊學士班
Minimizing energy consumption and improving performance in datacenters are critical to cost-saving for cloud operators, but traditionally, these two optimization objectives are treated separately. Therefore, this paper presents an unified solution combining two strategies, flow migration and VM migration, to maximize throughput and minimize energy, simultaneously. Traffic-aware flow migration (FM) is first incorporated in dynamic reroute (DENDIST), evolving into DENDIST-FM, in a software-defined network (SDN) for improving throughput and avoiding congestion. Second, given energy and topology information, VM migration (ETA-VMM) can help reduce traffic loads and meanwhile save energy. Our experimental result indicates that compared to previous works, the proposed method can improve throughput by 42.5% on average with only 2.2% energy overhead. Accordingly, the unified flow-and-VM migration solution has been proven effective for optimizing throughput and energy in SDN-based cloud datacenters.
Reducing Network Cost of Minimal-Migration Based VM Management in Cloud Datacenters
(2016)
Li, Kuan-Wei
;
Huang, Po-Han
;
Wen, Charles H-P.
;
電機工程學系
;
Department of Electrical and Computer Engineering
Nowadays, Cloud Service Providers (CSPs) provide flexible architectures and reliable internet accessories in cloud datacenters, where different types of applications are hosted. In order to achieve higher resource utilization and better quality of service, virtualization is widely used and implemented with virtual machines (VMs). However, most studies focus on VM management for utilization and energy, not yet considering the cost for emerging internet applications. Therefore, to further reduce network cost, we propose utilization-and-network-driven (UND) and only-network-driven (OND) strategies on top of the common minimal-migration (MM) based VM management as UND-MM and OND-MM. According to simulation results, both UND-MM and OND-MM strategies effectively reduce network cost by averagely 7% and 37%, respectively, of the original MM while retaining comparable resource utilization in CPU and memory for normally-loaded and overloaded datacenters. Moreover, the OND-MM, which is simpler, outperforms the UND-MM, and thus becomes more preferable for VM management in cloud datacenters.