Jiangling Zhang

Huazhong University of Science and Technology, Wu-han-shih, Hubei, China

Are you Jiangling Zhang?

Claim your profile

Publications (17)0.8 Total impact

  • Zhan Shi · Jiangling Zhang · Xinrong Zhou ·
    [Show abstract] [Hide abstract]
    ABSTRACT: While the storage market grows rapidly, software RAID, as a low-cost solution, becomes more and more important nowadays. However the performance of software RAID is greatly constrained by its implementation. Varies methods have been taken to improve its performance. By integrating a novel buffer mechanism – DMA aligned buffer (DAB) into software RAID kernel driver, we achieved a significant performance improvement, especially on small I/O requests.
    Computational Science - ICCS 2004, 4th International Conference, Kraków, Poland, June 6-9, 2004, Proceedings, Part III; 01/2004
  • Song Wu · Hai Jin · Jiangling Zhang · Dan Feng ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a device level approach to provide high availability of storage systems to an existing cluster computing system with minimal cost both in hardware and software. The parity-distribution scheme is used to place parity information across the network. After adopting the method, an existing system can tolerate and recover from disk and cluster node failures by enhanced disk drivers. Compared with previous work, this approach provides an attractive shortcut to realize higher availability, lower cost of design and implementation, better compatibility for applications and higher expandability
    High Performance Computing in the Asia-Pacific Region, 2000. Proceedings. The Fourth International Conference/Exhibition on; 02/2000
  • Yongfeng Huang · Dan Feng · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A voice gateway is a complex piece of networking communication equipment, which belongs to an MIMD multiprocessor system. This paper designs a voice gateway based on TMS320C6201 DSP introduces the architecture of a voice gateway, and analyses the data switching of the voice gateway. Some concepts, such as cluster components and resources, etc., are suggested
    High Performance Computing in the Asia-Pacific Region, 2000. Proceedings. The Fourth International Conference/Exhibition on; 02/2000
  • Yongfeng Huang · Yang Juan · Sifa Zhang · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper describes how to implement the G.723.1 recommendation in the IP telephony gateway and studies in detail the programming of the TMS320C6201 DSP and optimization methods for reducing the speech processing delay of the G.723.1 codec. As a result of adopting these optimization and programming methods, we have implemented a high-speed speech codec which can process concurrently 18 voice channels with a single TMS320C6201 chip in the IP telephony gateway. Finally, the paper summarizes the performance of the resulting ITU-T G.723.1 speech codec
    Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on; 02/2000
  • Yongfeng Huang · Sifa Zhang · Jiangling Zhang ·

    Proceedings of the ISCA 15th International Conference Computers and Their Applications, March 29-31, 2000, New Orleans, Louisiana, USA; 01/2000
  • Yongfeng Huang · Dan Feng · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A new idea of design on an Integrated IP Telephone Gateway is stated in this paper. Some key techniques are analyzed and discussed in detail. A Centered Echo Chancellor based on TMS320C6201 DSP is implemented, and many optimization methods for voice compression code are concluded to enhance the performance of the gateway. Finally, the performance parameters of the Integrated IP Telephony Gateway are summarized in this paper.
    Proceedings of SPIE - The International Society for Optical Engineering 08/1999; DOI:10.1117/12.360364 · 0.20 Impact Factor
  • Peng Cheng · Hai Jin · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we design a high performance RAID (Redundant Arrays of Independent Disks), called HUST_RAID, in real-time system. First, we give the architecture and the real-time process model of HUST_RAID. On the study of real-time disk I/O scheduling algorithm, we design a multiple queues real-time I/O scheduling algorithm. A new buffer management policy used in HUST_RAID is also given. It is the combination of some advanced technologies. We also introduce the design of software of the control system, which implements request receiving and decomposing, buffer management, data reconstruction, data rebuilding etc. A last, the performance of HUST_RAID is measured. From the testing results, the performance of HUST_RAID system is very excellent
    ACM SIGARCH Computer Architecture News 06/1999; 27(3):10-17. DOI:10.1145/333680.333688
  • Peng Cheng · Hai Jin · Xinrong Zhou · Qiong Chen · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: We discuss the design issues of a high performance RAID prototype in a real-time system, called HUST RAID. Based on the architecture and the real-time process model of HUST RAID, a multiple queue real-time I/O scheduling algorithm and new buffer management policy are proposed. We also introduce the software design issues of the control system. From the performance measurement results of HUST RAID, we can see that the performance of the HUST RAID system is excellent
    Communications, Computers and Signal Processing, 1999 IEEE Pacific Rim Conference on; 02/1999
  • Hai Jin · Jin He · Dan Feng · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: With the scale of disk array, more and more disk arrays are becoming stand alone. There are many advantages of using stand alone disk array, such as easy to install, upgrade and maintain, scalable. But the communication between host (called master) and disk array controller (called slave) is the main bottleneck problem in the system. The requirements for communication between master and slave are much higher. The bandwidth needed is well beyond that interface can supported and new ideas and methods are needed. In the communication between master and slave, higher performance is required for optimal utilization of the bandwidth. One of the promising technology to meet these requirements is to use the high speed data channel between master and slave. In this paper, a new method of using high speed data channel in the system of disk array is proposed. Some design issues of disk array using high speed data channel are also analyzed.
    SYBEN-Broadband European Networks and Electronic Image Capture and Publishing; 09/1998
  • Hai Jin · Xinrong Zhou · Dan Feng · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Redundant Array of Independent Disks (RAID) is now the most efficient way to solve the bottleneck problem between CPU and storage devices. In RAID 5, for the write operation, disk array controller first reads the old data from the disks which the new data will write into and the old parity information. Then, it calculates the new parity information using the old data, the new data and the old parity information to get the new parity. At last, disk array controller writes the new data and the new parity information altogether into the corresponding disks in the array. Thus, in order to write the new data into the disk array, it should perform the two extra read operations and one extra write operation compared with RAID 0. Hence, improving the stripe write operation is the key factor in improving the write performance of RAID 5. The basic idea is the reduce the stripe read/write operations to as few as possible. This paper discusses the different situations of stripe write operation in RAID 5, and proposes an adaptive control algorithm to improve the partial stripe write performance. The basic idea is to reduce the stripe read/write operation as much as possible
    Devices, Circuits and Systems, 1998. Proceedings of the 1998 Second IEEE International Caracas Conference on; 04/1998
  • Hai Jin · Zhichun Zhu · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In high performance scientific computing, operations such as searching, sorting, filtering are very frequent. Due to limited access ability of storage system and limited capacity of main memory, it is difficult to archive high performance of data access using traditional methods, and it is also impossible to carry all the data into main memory and operate in the memory. A new kind of storage system architecture, called Parallel Functional Disk Array (PFDA), is proposed in this paper. The study of the PFDA in this paper is mainly focused on the following two aspects: one is the architecture study of PFDA, including the storage requirement of high performance scientific computing environment and the operation frequency for different function, the other is the algorithm of dynamic analysis and scheduling of function units to perform in parallel, so as to solve the conflict problem among function units and the data dependent problem caused by the data distribution in the redundant disk array
    Information, Communications and Signal Processing, 1997. ICICS., Proceedings of 1997 International Conference on; 10/1997
  • Hai Jin · Peng Cheng · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The optical storage device has been put into practical use as an ultimate high capacity, low cost per bit, removable and reliable information recording device. But it has the disadvantage of slow speed compared with magnetic disks. In order to improve the data transfer rate of the optical storage subsystem, and to enhance the overall data throughput of the system, the authors propose a new optical storage system architecture, which is a combination of an array of optical storage devices and a plurality of buffers coupled to a separate optical storage device. For scheduling of the buffer, they use the technique of the Ping-Pong buffer and the tagged elevator write back as the destage algorithm. The performance of the optical storage system is dramatically enhanced, especially for small writes
    Communications, Computers and Signal Processing, 1997. '10 Years PACRIM 1987-1997 - Networking the Pacific Rim'. 1997 IEEE Pacific Rim Conference on; 09/1997
  • Dan Feng · Xinrong Zhou · Hai Jin · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A stochastic Petri nets (SPN) model of RAID-5 is constructed. With the model and its isomorphic Markov chain, the average utilization of disk drives in RAID for small write and large I/O request can be calculated. It provides us a good method to evaluate the performance of RAID in the paper
    Advances in Parallel and Distributed Computing, 1997. Proceedings; 04/1997
  • Dan Feng · Zhichun Zhu · Hai Jin · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A fast disk array is designed for the large continuous image storage. It includes a high speed data architecture and the technology of data striping and organization on the disk array. The high speed data path which is constructed by two dual port RAM and some control circuit is configured to transfer data between a host system and a plurality of disk drives. The bandwidth can be more than 100 MB/s if the data path based on PCI (peripheral component interconnect). The organization of data stored on the disk array is similar to RAID 4. Data are striped on a plurality of disk, and each striping unit is equal to a track. I/O instructions are performed in parallel on the disk drives. An independent disk is used to store the parity information in the fast disk array architecture. By placing the parity generation circuit directly on the SCSI (or SCSI 2) bus, the parity information can be generated on the fly. It will affect little on the data writing in parallel on the other disks. The fast disk array architecture designed in the paper can meet the demands of the image storage.
    Proceedings of SPIE - The International Society for Optical Engineering 01/1997; DOI:10.1117/12.263519 · 0.20 Impact Factor
  • Dan Feng · Hai Jin · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The EVENODD code is a kind of code for tolerating double disk failures in a RAID (redundant arrays of independent disks) architecture. Compared with Reed-Solomon (RS) error-correcting codes, it has a simple computational complexity. However, it also has a disadvantage, the small write performance is worse than that of RS codes. An improved scheme of EVENODD code is presented. Compared with EVENODD, the scheme can improve the small write performance without increasing the computational complexity of the code
  • Xinrong Zhou · Peng Cheng · Hai Jin · Jiangling Zhang ·

    01/1997; 21(S_2_PMRC_97_2):S2_343-346. DOI:10.3379/jmsjmag.21.S2_343
  • Hai Jin · Dan Feng · Jiangling Zhang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Disk shadowing technology is the best way to enhance the availability and reliability of I/O subsystem. The most important metric of on-line performance of disk shadowing subsystem is the utilization of CPU. This paper first employs the way to calculate the CPU utilization of system connected with shadowed disk using statistic average method system. From the simulation results of CPU utilization of system connected with disk shadowing subsystem can we se that in most cases the on-line performance of shadowed disk are better than single disk, thus using disk shadowing technology can have enhanced on-line performance combined with highest availability and highest reliability.
    Proceedings of SPIE - The International Society for Optical Engineering 05/1996; DOI:10.1117/12.241175 · 0.20 Impact Factor