Woo-Cheol Kwon

Optimal voltage allocation techniques for dynamically variable voltage processors

This paper presents a set of new important results for the problem of task scheduling and voltage allocation in dynamically variable voltage processor for minimizing the total processor energy consumption. The contributions are two folds: (1) For given multiple discrete supply voltages and tasks with arbitrary arrival-time/deadline constraints, we propose a voltage allocation technique which produces a feasible task schedule with optimal processor energy consumption; (2) We then extend the problem to include the case in which tasks have non-uniform load (i.e., switched) capacitances, and solve it optimally.

In-network reorder buffer to improve overall NoC performance while resolving the in-order requirement problem

Data-intensive functions on chip, e.g., codec, 3D graphics, pixel processing, etc. need to make best use of the increased bandwidth of multiple memories enabled by 3D die stacking via accessing multiple memories in parallel. Parallel memory accesses with originally in-order requirements necessitate reorder buffers to avoid deadlock. Reorder buffers are expensive in terms of area and power consumption. In addition, conventional reorder buffers suffer from a problem of low resource utilization. In our work, we present a novel idea, called in-network reorder buffer, to increase the utilization of reorder buffer resource. In our method, we move the reorder buffer resource and related functions from network entry/exit points to network routers. Thus, the in-network reorder buffers can be better utilized in two ways. First, they can be utilized by other packets without in-order requirements while there are no in-order packets. Second, even in-order packets can benefit from in-network reorder buffers by enjoying more shares of reorder buffers than before. Such an increase in reorder buffer utilization enables NoC performance improvement while supporting the original in-order requirements. Experimental results with an industrial strength DTV SoC example show that the presented idea improves the total execution cycle by 16.9%.

A practical approach of memory access parallelization to exploit multiple off-chip DDR memories

3D stacked memory enables more off-chip DDR memories. Redesigning existing IPs to exploit the increased memory parallelism will be prohibitively costly. In our work, we propose a practical approach to exploit the increased bandwidth and reduced latency of multiple off-chip DDR memories while reusing existing IPs without modification. The proposed approach is based on two new concepts: transaction id renaming and distributed soft arbitration. We present two on-chip network components, request parallelizer and read data serializer, to realize the concepts. Experiments with synthetic test cases and an industrial strength DTV SoC design show that the proposed approach gives significant improvements in total execution cycle (21.6%) and average memory access latency (31.6%) in the DTV case with a small area overhead (30.1% in the on-chip network, and less than 1.4% in the entire chip).

An open-loop flow control scheme based on the accurate global information of on-chip communication

3D stacked memory is being adopted as a promising solution to offer high bandwidth and low latency in memory access. Compared with the on-chip network design with conventional off chip memory, it gives a new problem of minimizing communication conflicts since multiple concurrent high bandwidth data transfers will flow through the on-chip network. In order to tackle this problem, we propose applying an open-loop flow control scheme based on the accurate global information (destination and status) of on-chip communication. The proposed open-loop flow control scheme exploits the information and selectively buffers and arbitrates data transfers to remove conflicts at destinations in a preventive manner. As an implementation of the presented scheme, we present on-chip buffers called Buf3Ds that share the global information with each other to perform the selective buffering and arbitration of data transfers. Experiments with synthetic test cases and an industrial strength DTV design show that the proposed method improves aggregate memory bandwidth significantly (average 19.0 %~25.8 % in the synthetic cases and up to 18.4 % in the DTV case) with a small area overhead (15.2 % in the DTV case) of on-chip network.

A systematic IP and bus subsystem modeling for platform-based system design

The topic on platform-based system modeling has received a great deal of attention today. One of the important tasks that significantly affect the effectiveness and efficiency of the system modeling is the modeling of IP components and communication between IPs. To be effective, it is generally accepted that the system modeling should be performed in two steps; In the first step, a fast but some inaccurate system modeling is considered to facilitate the simultaneous development of software and hardware. The second step then refines the models of the software and hardware blocks (i.e., IPs) to increase the simulation accuracy for the system performance analysis. Here, one critical factor required for a successful system modeling is a systematic modeling of the IP blocks and bus subsystem connecting the IPs. In this respect, this work addresses the problem of systematic modeling of the IPs and bus subsystem in different levels of refinements. In the experiments, we found that by applying our proposed IP and bus modeling methods to the MPEG-4 application, we are able to achieve 4times performance improvement and at the same time, reduce the software development time by 35%, compared to that by conventional modeling methods