Mengjou Lin

Performance of a Mass-Storage System for Video-on-Demand

Advancements in storage technology along with the fast deployment of high-speed networks has allowed the storage, transmission, and manipulation of multimedia information such as text, graphics, still images, video, and audio to be feasible. Our study focused on the performance of the mass storage system for a large-scale video-on-demand server. Different video file striping schemes, such as application level striping and device driver level striping, were examined in order to study scalability and performance issues. To study the impact of different concurrent access patterns on the performance of a server, experimental results were obtained on group access on a single video file and multiple group accesses on multiple video files. All of our experiments were conducted on a fully configured Silicon Graphics Inc. Onyx computer system. The Onyx machine was connected to 31 SCSI-2 channels which have 496 GBytes disk storage, 20 MIPS R4400 processors, and 768 MBytes main memory. From the experimental results, the storage system of Onyx machine can potentially provide about 360 concurrent video accesses with guaranteed quality of service.

Performance of high-speed network I/O subsystems: case study of a Fibre Channel network

Emerging high-speed networks provide several hundred megabits per second to several gigabits per second of raw communication bandwidth. However, the maximum achievable throughput available to the end-user or application is quite limited. In order to fully utilize the network bandwidth and to improve the performance at the application level, a careful examination of I/O subsystems is essential. In this paper, we study one emerging high-speed network, the Fibre Channel network. The objectives of this study are: to understand how the I/O subsystem relates to network operations; to evaluate and analyze the performance of such a subsystem; and to propose possible approaches for improving the maximum achievable bandwidth. We show (by simply modifying device driver code) a 75% maximum achievable bandwidth improvement. Other ways of improving network performance are also discussed.<<ETX>>

Performance of a mass storage system for video-on-demand

Advancements in storage technology along with the fast deployment of high-speed networks has allowed the storage, transmission and manipulation of multimedia information such as text, graphics, still images, video and audio to be feasible. Our study focused on, the performance of the mass storage system for a large-scale video-on-demand server. Different video file striping schemes, such as application level striping and device driver level striping, were examined in order to study scalability and performance issues. To study the impact of different concurrent access patterns on the performance of a server, experimental results were obtained on group access on a single video file and multiple group accesses on multiple video files.

Performance evaluation of the CM-5 interconnection network

The authors present performance characteristics of the CM-5 (Connection Machine-5) data network with respect to bandwidth and latency. They present the maximum effective network bandwidth as a function of massage size, varying data path through different levels of the network hierarchy, and system load. They also present latency as a function of message size and network hierarchy. The maximum effective network bandwidth is given as a function of data flow pattern (2-D and 3-D mesh, stencils, ring, tree, and hypercube) using several variant embedding schemes. Measurements were obtained using CMMD V 1.3.1, Thinking Machine Corporations message-passing library, and CMNF V 1.1, a fast message-passing library. Using CMNF, measurements which were very close to actual hardware capabilities were observed.<<ETX>>

Performance of a storage system for supporting different video types and qualities

Future video-on-demand (VOD) servers will need to support many existing and emerging video data types. These data types include 15-fps (frames per second) animation, 30-fps NTSC (National Television Systems Committee) quality video and 80-fps HDTV (High Definition Television) video. The different display speeds and frame sizes of these various video types impose a major constraint on the design of VOD storage systems. This paper presents the results of an experimental study, conducted on a Silicon Graphics Inc. Onyx computer system, that investigated the impact of these video types on the design of a VOD storage system. The key issues involved in supporting these different video types in a VOD environment are: (1) the video allocation method, and (2) the proper block size (a block is a basic unit of several contiguous video frames that will be accessed from several disks each time a request is made) to use for data striping and retrieval. Two allocation schemes, logical volume striping and application level striping, along with varying frame and block sizes for each of the three different video data types are examined. The focus of our study is to determine the maximum number of concurrent accesses that can be supported with a guaranteed quality of service. The degree of scalability (i.e., striping data over more disk arrays) of the experimental VOD system used is also studied.

Key to the success of asynchronous transfer mode: an application programming interface

Asynchronous transfer mode (ATM) networks are scalable via a wide variety of physical link speeds and can be run in both the local and wide areas. Bandwidth reservation and quality of service guarantees make ATM attractive to deliver multimedia information in real-time. Existing protocol suites can be extended. However, they do not give access to those characteristics. Therefore, applications demand an ATM application programming interface (API) to take advantage of those features. This paper covers a special syntax mapping of the ATM API services on top of X/Open Transport Interface (XTI) which is able to give applications access directly to the special quality and traffic services of an ATM network. A particular realization of the API on the XTI environment is proposed and implemented on the MacOS networking platform. It provides a flexible and modular design of the ATM networking. Finally, the paper concludes with one ATM-aware application (video on demand) which requires real-time streaming data.

Performance characteristics of the Connection Machine hypertree network

Abstract This paper presents performance characteristics of the CM-5 data network with respect to bandwidth and latency. We present the maximum effective network bandwidth as a function of (1) message size, (2) varying data path through different levels of the network hierarchy, and (3) system load. We also present latency as a function of message size and network hierarchy. In the second part of our evaluation, we present the maximum effective network bandwidth as a function of data flow pattern (2-d and 3-d mesh, stencils, ring, tree, and hypercube) using several variant embedding schemes. Measurements were obtained using CMMD V 1.3.1, Thinking Machine Corporation′s message-passing library, and CMNF V 1.1, a fast message-passing library developed by the Minnesota Supercomputer Center (MSC) for the Army High Performance Research Center (AHPCRC). Using CMNF we were able to observe measurements which were very close to actual hardware capabilities.

Virtual Permanent Connection: network computing over switch-based high-speed networks

Recent progress in switch based high speed local area networks (LANs) makes distributed network computing promising. Three evolving switch based high speed networks are the High Performance Parallel Interface (HIPPI), Fiber Channel (FC), and Asynchronous Transfer Mode (ATM) standards. We study how high performance computing can be carried out over such networks. High performance computing can be characterized as follows: it includes multiple modules and each module is executed in a processor; its communication data flow forms a special application topology and usually such application topologies are regular; and it requires frequent communication between adjacent modules in the application topology. In order to reduce the amount of time required for a processor to set up a connection during the execution of an application, we propose a new communication protocol called the Virtual Permanent Connection (VPC). For a given application topology, a set of connections are set up and permanently maintained during the execution of the application. Communication between processors are via this group of connections. We study how a set of VPCs are chosen based on a given application topology (this process is called application topology embedding).

High-performance computing over switch-based high-speed local networks

Communication between processors has long been the bottleneck of distributed network computing. However, recent progress in switch-based high-speed networks including ATM, HIPPI and Fibre Channel may be changing this situation. To utilize the aggregate computational power of clusters of workstations, We have especially concentrated our effort on maximizing the achievable throughput and minimizing the delay for end-users. In order to achieve better performance, we have investigated the following issues: i) ways to improve the performance of I/O subsystems and ii) porting PVM over high-speed network APIs. The typical communications between processors are going through a stack of network protocols, (i) represents our effort to improve the lower layer communication performance. However, the application (or user) level of performance may not be improved too much even if the lower layer performance is improved. Therefore, to improve user level performance it is required to bypass most of the network protocols in the communication stack and, hence, the overheads generated by them by porting applications (like PVM) directly over lower layer network protocols.