Motion video signal processing for recording or reproducing – Local trick play processing – With randomly accessible medium
Reexamination Certificate
1994-12-05
2001-05-29
Garber, Wendy R. (Department: 2612)
Motion video signal processing for recording or reproducing
Local trick play processing
With randomly accessible medium
C725S092000
Reexamination Certificate
active
06240243
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to data processing systems and more particularly to data processing systems including display of data and to the storage and retrieval of scalable video data in a disk array-based video server.
2. Prior Art
Recently, the promise of multimedia technologies to have a significant impact on both information providing service and entertainment business has created several new ventures. Given the extremely large data size, the major challenge to handling multimedia data is to support not only very high disk bandwidth for video retrieval but also very high network bandwidth for data transmission. ATM (Asynchronous Transfer Mode) has been proposed as a solution to meet the demand for high network bandwidth. On the other hand, disk-arrays are employed to provide the disk bandwidth required for a video server. For example, an HDTV (high definition television) quality image, will require video data at 2 Mbytes per second (even after compression). It is not desirable to store such video in a single disk for two reasons. First, a 100 minute HDTV movie will require 12 Gbyte of storage. Such a large disk is usually expensive. Second, playing a hot (i.e., frequently requested) movie by a single disk may cause a performance bottleneck. In fact, even for playing movies of ordinary quality, the need to support multiple video streams from a video server also calls for the use of disk-arrays. Consequently, it is highly desirable to use data striping in a disk-array to handle the storage and retrieval of video data, where data striping means dividing the video data into blocks according to their presentation order (i.e., time sequence) and storing these blocks on different disks. It is noted that with such a data striping method, better load balancing can be achieved by staggering the starting times of different video streams. This is called time staggering.
Although data striping improves system performance, a large buffer is required to store the data and loading between disks in the array can become unbalanced.
In order to serve users with different requirements and capabilities from a single copy of a video, it is necessary to develop a scalable video stream. Multi-resolution coding is able to provide scalable video.
SUMMARY OF THE INVENTION
It is an object of the present invention to store and retrieve scalable video data in a disk-array-based video server so as to minimize the buffer space required by the server and to improve the system throughput.
Accordingly, a proposed rate staggering technique is as follows: Let r be the number of different classes of video the server can provide; the whole video data is divided into r partitions, called rate 1 data, rate 2 data, . . . , and rate r data; the lowest quality video, referred to as class 1 video, requires only rate 1 data for playout; the second to the lowest quality video, referred to as class 2 video, requires both rate 1 and rate 2 data for playout; in general, class i video requires all rate j data, 1≦j≦i, for playout; let P (in byte/second) be the playout speed for the decoder to play out the full resolution video and T be the one round retrieval time by the disk array; using the double buffering method (i.e., the buffer space is chosen to be twice as that needed to accommodate the data retrieved in one round), the buffer space required by the server for a full resolution stream is equal to 2T P.
Let b (in bytes) be the size of each block and k be the displacement factor for staggering data blocks of different rates in the disk array. For example, the displacement factor in Table 2 is two.
⌈
TP
b
⌉
is the number of data blocks needed by a full resolution video stream within the time duration T, where ┐.┌ is a ceiling function. Hence, in order to achieve load balancing among disks, the displacement factor k is designed to be
⌈
TP
br
⌉
.
Such a placement by rate staggering can spread the workload of each stream evenly across disks.
Denote the buffer size of the end decoder as B
D
. It is noted that the maximal amount of data the end decoder can retrieve at a time is half of its total buffer size (i.e., the other half is being used for playout). B
D
thus has to be greater than or equal to
2
⌈
TP
b
⌉
b
.
The data placement for scalable video in a disk array of n disks can be determined by Procedure R below.
Procedure R: Data placement for scalable video in a disk array of n disks.
Step 1: Determine the displacement factor
k
=
⌈
TP
br
⌉
.
Step 2: Place block B
i,j
in disk d(i, j), where
d(i, j)=[(j−1)k+i]
n
[b]
a
means b mod a. B
i, j
is a block comprising rate j data of the ith clip. It can be seen from Table 2 that with n=8 and k=2, B
3, 2
resides in disk d(3, 2)=[2+3]
8
=5 and B
5, 4
resides in disk d(5, 4)=[6+5]
8
=3.
The rate staggering technique for efficient storage of video data is embodied in a data processing system including a disk array based video server, which includes a disk array for storing video data, a processor for processing data, a memory buffer for storing video data and a rate staggering controller. Data is transferred to a network from video server through a network interface and from the network to client stations employing devices having differing resolution capabilities. The processor executes tasks under control of the rate staggering controller. These tasks include a retrieval task which retrieves scalable video from the array of disks, a storage task which stores videos temporarily in the memory buffer, and a transmission task which transmits video to client stations through the network.
These, and other objects, features and advantages of this invention will become apparent from the following detailed description of the invention taken in conjunction with the accompanying drawings.
REFERENCES:
patent: 5027400 (1991-06-01), Baji et al.
patent: 5414455 (1995-05-01), Hooper et al.
patent: 5450139 (1995-09-01), Haraguchi et al.
patent: 5453779 (1995-09-01), Dan et al.
patent: 5461415 (1995-10-01), Wolf et al.
patent: 5510905 (1996-04-01), Birk
patent: 5528282 (1996-06-01), Voeten et al.
“Scalable Storage Servers For Digital Audio and Video”, P Lougher et al., IEE Conference on Storage and Recording Systems, Apr. 5-7, 1994, pp. 14-143, Publication No. 402.*
“Designing a Multiuser HDTV Storage Server”, Harrick M. Vin et al., IEEE Journal On Selected Areqas In Communications, vol. 11. No. 1. Jan. 1993, pp. 153-164.
Chen Ming-Syan
Kandlur Dilip Dinkar
Yu Philip Shi-Lung
Garber Wendy R.
International Business Machines - Corporation
Jordan Kevin M.
Moe Aung S.
LandOfFree
Method and apparatus for storing and retrieving scalable... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for storing and retrieving scalable..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for storing and retrieving scalable... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2460093