Shi-Wu Lo

Real-time garbage collection for flash-memory storage systems of real-time embedded systems

Flash-memory technology is becoming critical in building embedded systems applications because of its shock-resistant, power economic, and nonvolatile nature. With the recent technology breakthroughs in both capacity and reliability, flash-memory storage systems are now very popular in many types of embedded systems. However, because flash memory is a write-once and bulk-erase medium, we need a translation layer and a garbage-collection mechanism to provide applications a transparent storage service. In the past work, various techniques were introduced to improve the garbage-collection mechanism. These techniques aimed at both performance and endurance issues, but they all failed in providing applications a guaranteed performance. In this paper, we propose a real-time garbage-collection mechanism, which provides a guaranteed performance, for hard real-time systems. On the other hand, the proposed mechanism supports non-real-time tasks so that the potential bandwidth of the storage system can be fully utilized. A wear-leveling method, which is executed as a non-real-time service, is presented to resolve the endurance problem of flash memory. The capability of the proposed mechanism is demonstrated by a series of experiments over our system prototype.

A remote thin client system for real time multimedia streaming over VNC

This paper proposes a remote thin client system for real time multimedia streaming over VNC. A remote frame can be split as two parts, i.e. high motion part and low motion part, and transmitted through the Internet from servers to clients according to the proposed hybrid RTP protocol. A Dynamic Image Detection Scheme (DIDS) is proposed to automatically detecting the high motion part of a frame with only 1% of extra CPU loading. In addition, an Error Detection Scheme (EDS) and a Dynamic Bit-rate Control Scheme (DBCS) are also proposed to ensure good video streaming quality under bandwidth limited applications. This paper also proposes a linear time BU-level rate control algorithm to ensure the proposed DBCS can be finished in real time. The proposed algorithm reduces the computational complexity from O(n2)to be O(n). By using the proposed thin client system, we can achieve about 22 fps of real-time SIF video streaming with good video quality under 32 KByte/s of bandwidth limitation, which speeds up about 172 times in remote frame display when compared to pure VNC.

A dynamic-voltage-adjustment mechanism in reducing the power consumption of flash memory for portable devices

A dynamic-voltage-adjustment method is proposed to reduce the power consumption of a flash memory storage system, depending on the system workload. The usefulness of the proposed method is demonstrated by a series of experiments with very encouraging results.

Swap-before-hibernate: a time efficient method to suspend an OS to a flash drive

Unlike a magnetic disk, a flash drive needs no seek time when performing random access. In addition, the read speed of a flash drive is faster than the write speed, and the write speed depends on the I/O request size: the bigger the request size, the faster the write speed. System hibernation stores the machine status completely in non-volatile memory. When the system reboots, the data will be reloaded to the machine, and the system status will be recovered completely. The user can continue immediately where they stopped the last time. We use the characteristic that the flash drives requires no seek time when performing random access to write the data in the memory to swap space or a hibernation file, or simply discard it without any I/O. The write process will try to combine random small writes into continuous large writes, if possible, to optimize the write speed. When the system resumes, it only reloads the data that the user needs, and the resume technique is based on paging on demand. Paging on demand is actually a kind of random access; however, since the flash drive requires no seek time, it does not drag down the system response time due to reducing the number of I/O.

Location management in cellular mobile computing systems with dynamic hierarchical location databases

An important issue in the design of a mobile computing system is how to manage the location information of mobile clients. In the existing commercial cellular mobile computing systems, a two-tier architecture is adopted. However, the two-tier architecture is not scalable. In the literatures, a hierarchical database structure is proposed in which the location information of mobile clients within a cell is managed by the location database responsible for the cell. The location databases of different cells are organized into a tree-like structure to facilitate the search of mobile clients. Although this architecture can distribute the updates and the searching workload amongst the location databases in the system, location update overheads can be very expensive when the mobility of clients is high. In this paper, we study the issues on how to generate location updates under the distance-based method for systems using hierarchical location databases. A cost-based method is proposed for calculating the optimal distance threshold with the objective to minimize the total location management cost. Furthermore, under the existing hierarchical location database scheme, the tree structure of the location databases is static. It cannot adapt to the changes in mobility patterns of mobile clients. This will affect the total location management cost in the system. In the second part of the paper, we present a reorganization strategy to restructure the hierarchical tree of location databases according to the mobility patterns of the clients with the objective to minimize the location management cost. Extensive simulation experiments have been performed to investigate the reorganization strategy when our location update generation method is applied.

Real-time task scheduling for SMT systems

Although simultaneous multithreading (SMT) has been shown being an efficient technique to improve processor performance, little work has been done on real-time SMT scheduling. The objective of this paper is to explore realtime SMT scheduling for independent periodic task sets with schedulability guarantees. We propose emulation-based scheduling algorithms with and without task migration to emulate an adjustable SMT processor over a non-adjustable SMT processor. The schedulability tests for the proposed scheduling algorithms are presented. An approximation bound on the average number of tasks running in parallel is also shown. The performance of the proposed algorithms was evaluated by a series of simulation experiments.

Multi-layer bus minimization for SoC

The deployment of multiple processing elements such as a microprocessor or a Digital Signal Processor in embedded systems often results in significant communication overheads. The challenge lies in resolving the communication cost minimization problem, while simultaneously satisfying the timing constraints of job executions. In this paper, we explore bus-layer minimization problems by first identifying factors that contribute to the NP-hardness of these problems. Existing proposed algorithms and NP-hard problems are then identified and elucidated. A simulated annealing algorithm is proposed and compared with heuristics-based algorithms to provide further insights for system designers. Lastly, a series of extensive simulations is carried out and a case study is presented to show comparisons among different approaches and workloads.

Multi-disk scheduling for time-constrained requests in RAID-0 devices

In this paper, we study the scheduling problem of real-time disk requests in multi-disk systems, such as RAID-0. We first propose a multi-disk scheduling algorithm, called Least-Remaining-Request-Size-First (LRSF), to improve soft real-time performance of I/O systems. LRSF may be integrated with different real-time/non-real-time single-disk scheduling algorithms, such as SATF and SSEDV, adopted by the disks in a multi-disk system. We then extend LRSF by considering the serving requests on-the-way (OTW) to the target request to minimize the starvation problem for requests that need to retrieve a large amount of data. The pre-fetching issue in RAID-0 is also studied to further improve the I/O performance. The performance of the proposed algorithm and schemes is investigated and compared with other disk scheduling algorithms through a series of experiments using both randomly generated workload and realistic workload.

A signature-based Grid index design for main-memory RFID database applications

A large-scale RFID application often requires a highly efficient database system in data processing. This research is motivated by the strong demand for an efficient index structure design for main-memory database systems of RFID applications. In this paper, a signature-based Grid index structure is proposed for efficient data queries and storage. An efficient methodology is proposed to locate duplicates and to execute batch deletions and range queries based on application domain knowhow. The capability of the design is implemented in an open source main-memory database system H2 and evaluated by realistic workloads of RFID applications.

A priority assignment strategy of processing elements over an on-chip bus

The number of bus transactions in System-on-Chip (SoC) grows significantly in recent years. Because of different timing constraints for different applications, how to find a proper priority assignment for processing elements (PEs) of SoC becomes very challenging. In this paper, we first show that the priority assignment problem with one unique priority for each PE is NP-complete. When each bus transaction can have one unique priority, we propose an optimal priority assignment algorithm for a given workload. We then propose a priority assignment strategy based on Simulated Annealing (SA) for PEs, where bus arbitration is done in a priority-driven fashion. The objective is to minimize the number of priorities needed for each PE and to satisfy the timing constraints of applications. The experimental results show some encouraging results in priority assignment.