David Hung-Chang Du

Path sensitization in critical path problem

Since the delay of a circuit is determined by the delay of its longest sensitizable paths (such paths are called critical paths), the problem of estimating the delay of a circuit is called critical path problem. One important aspect of the critical path problem is to decide whether a path is sensitizable. A framework which allows various previously proposed path sensitization criteria to be compared with each other in a unified way is presented. An exact path sensitization criterion and a looser path sensitization criterion based on the framework are also proposed.<<ETX>>

Video staging: a proxy-server-based approach to end-to-end video delivery over wide-area networks

Real-time distribution of stored video over wide-area networks (WANs) is a crucial component of many emerging distributed multimedia applications. The heterogeneity in the underlying network environments is an important factor that must be taken into consideration when designing an end-to-end video delivery system. We present a novel approach to the problem of end-to-end video delivery over WANs using proxy servers situated between local-area networks (LANs) and a backbone WAN. A major objective of our approach is to reduce the backbone WAN bandwidth requirement. Toward this end, we develop an effective video delivery technique called video staging via intelligent utilization of the disk bandwidth and storage space available at proxy servers. Using this video staging technique, only part of a video stream is retrieved directly from the central video server across the backbone WAN whereas the rest of the video stream is delivered to users locally from proxy servers attached to the LANs. In this manner, the WAN bandwidth requirement can be significantly reduced, particularly when a large number of users from the same LAN access the video data. We design several video staging methods and evaluate their effectiveness in trading the disk bandwidth of a proxy server for the backbone WAN bandwidth. We also develop two heuristic algorithms to solve the problem of designing a multiple video staging scheme for a proxy server with a given video access profile of a LAN. Our results demonstrate that the proposed proxy-server-based approach provides an effective and scalable solution to the problem of the end-to-end video delivery over WANs.

A network-conscious approach to end-to-end video delivery over wide area networks using proxy servers

In this paper we present a novel network-conscious approach to the problem of end-to-end video delivery over wide-area networks using proxy servers situated between local-area networks (LANs) and a backbone wide-area network (WAN). We develop a novel and effective video delivery technique called video staging via intelligent utilization of the disk bandwidth and storage space available at proxy servers. We also design several video staging methods and evaluate their effectiveness in reducing the backbone WAN bandwidth requirement. Our results demonstrate that the proposed proxy-server-based, network-conscious approach provides an effective and scalable solution to the problem of the end-to-end video delivery over wide-area networks.

Adaptive proportional routing: a localized QoS routing approach

Most of the QoS routing schemes proposed so far require periodic exchange of QoS state information among routers, imposing both communication overhead on the network and processing overhead on core routers. Furthermore, stale QoS state information causes the performance of these QoS routing schemes to degrade drastically. In order to circumvent these problems, we focus on localized QoS routing schemes where the edge routers make routing decisions using only local information and thus reducing the overhead at core routers. We first describe virtual capacity based routing (vcr), a theoretical scheme based on the notion of virtual capacity of a route. We then propose proportional sticky routing, an easily realizable approximation of vcr and analyze its performance. We demonstrate through extensive simulations that adaptive proportional routing is indeed a viable alternative to the global QoS routing approach.

Real-Time Detection of Clone Attacks in Wireless Sensor Networks

A central problem in sensor network security is that sensors are susceptible to physical capture attacks. Once a sensor is compromised, the adversary can easily launch clone attacks by replicating the compromised node, distributing the clones throughout the network, and starting a variety of insider attacks. Previous works against clone attacks suffer from either a high communication/storage overhead or a poor detection accuracy. In this paper, we propose a novel scheme for detecting clone attacks in sensor networks, which computes for each sensor a social fingerprint by extracting the neighborhood characteristics, and verifies the legitimacy of the originator for each message by checking the enclosed fingerprint. The fingerprint generation is based on the superimposed s-disjunct code, which incurs a very light communication and computation overhead. The fingerprint verification is conducted at both the base station and the neighboring sensors, which ensures a high detection probability. The security and performance analysis indicate that our algorithm can identify clone attacks with a high detection probability at the cost of a low computation/communication/storage overhead. To our best knowledge, our scheme is the first to provide realtime detection of clone attacks in an effective and efficient way.

Rejuvenator: A static wear leveling algorithm for NAND flash memory with minimized overhead

NAND flash memory is fast replacing traditional magnetic storage media due to its better performance and low power requirements. However the endurance of flash memory is still a critical issue in using it for large scale enterprise applications. Rethinking the basic design of NAND flash memory is essential to realize its maximum potential in large scale storage. NAND flash memory is organized as blocks and blocks in turn have pages. A block can be erased reliably only for a limited number of times and frequent block erase operations to a few blocks reduce the lifetime of the flash memory. Wear leveling helps to prevent the early wear out of blocks in the flash memory. In order to achieve efficient wear leveling, data is moved around throughout the flash memory. The existing wear leveling algorithms do not scale for large scale NAND flash based SSDs. In this paper we propose a static wear leveling algorithm, named as Rejuvenator, for large scale NAND flash memory. Rejuvenator is adaptive to the changes in workloads and minimizes the cost of expensive data migrations. Our evaluation of Rejuvenator is based on detailed simulations with large scale enterprise workloads and synthetic micro benchmarks.

Performance study of iSCSI-based storage subsystems

iISCSI is emerging as an end-to-end protocol for transporting storage I/O block data over IP networks. By exploiting the ubiquitous Internet infrastructure, iSCSI greatly facilitates remote storage, remote backup, and data mirroring. This article evaluates the performance of two typical iSCSI storage subsystems by measuring and analyzing block-level I/O access performance and file-level access performance. In the file-level performance study, we compare file access performance in an NAS scheme with that in an iSCSI-based SAN scheme. Our test results show that Gigabit Ethernet-based iSCSI can reach very high bandwidth, close to that of a direct FC disk access in block I/O access. However, when the iSCSI traverses through longer distance, throughput relies heavily on the available bandwidth between the initiator and the target. On the other hand, the file-level performance shows that iSCSI-based file access (SAN scheme) provides higher performance than using NFS protocol in Linux and SMB protocol in Windows (NAS scheme). However, the advantage of using iSCSI-based file accesses decreases as the file size increases. The obtained experimental results shed some light on the performance of applications based on iSCSI storage.

A Forest-structured Bloom Filter with flash memory

A Bloom Filter (BF) is a data structure based on probability to compactly represent/record a set of elements (keys). It has wide applications on efficiently identifying a key that has been seen before with minimum amount of recording space used. BF is heavily used in chunking based data de-duplication. Traditionally, a BF is implemented as in-RAM data structure; hence its size is limited by the available RAM space on the machine. For certain applications like data de-duplication that require a big BF beyond the size of available RAM space, it becomes necessary to store a BF into a secondary storage device. Since BF operations are inherently random in nature, magnetic disk provides worse performance for the random read and write operations. It will not be a good fit for storing the large BF. Flash memory based Solid State Drive (SSD) has been considered as an emerging storage device that has superior performance and can potentially replace disks as the preferred secondary storage devices. However, several special characteristics of flash memory make designing a flash memory based BF very challenging. In this paper, our goal is to design an efficient flash memory based BF that is fully aware of these physical characteristics. To this end, we propose a Forest-structured BF design (FBF). FBF uses a combination of RAM and flash memory to design a BF. BF is stored on the flash, while RAM helps to mitigate the impact of slow write performance of flash memory. In addition, in-flash BF is organized in a forest-like structure in order to improve the lookup performance. Our experimental results show that FBF design achieves 2 times faster processing speed with 50% less number of flash write operations when compared with the existing flash memory based BF designs.

Efficient Algorithms for Extracting the K Most Critical Paths in Timing Analysis

Path extracting algorithms are a very important part of timing analysis approach. In this paper we designed and developed several algorithms which can generate the K most critical paths in a non-increasing order of their delays. The effectiveness of these algorithms is shown by some experimental results.

Path sensitization in critical path problem (logic circuit design)

An important aspect of the critical path problem is deciding whether a path is sensitizable. Three new path sensitization criteria are proposed in a general framework. Other path sensitization criteria can be presented in the same framework, enabling them to be compared with each other. An approximate criterion is also proposed and used to develop an efficient critical path algorithm for combinational circuits. >