Push to storage network enabling fast start

• US 20090106392A1
• Filed: 12/20/2008
• Published: 04/23/2009
• Est. Priority Date: 06/20/2006
• Status: Active Grant

• Current Assignee: Patentvc Limited
• Original Assignee: Patentvc Limited
• Inventors: Thieberger, Gil, Zuckerman, Gal
• Granted Patent: US 7,895,291 B2
• Time in Patent Office: Days
• US Class Current: 709/217
• CPC Class Codes: G06F 16/1834 implemented based on peer-to-peer networks, e.g. gnutella p2p communication protocols H04L67/104
• CPC Class Codes: G06F 16/70 of video data
• CPC Class Codes: H04L 67/104 Peer-to-peer [P2P] networks
• CPC Class Codes: H04L 67/1044 Group management mechanisms management of multicast group membership H04L12/185; reconfiguring of node membership in a computing system to eliminate errors G06F11/1425
• CPC Class Codes: H04L 67/1046 Joining mechanisms

Abstract

Methods and systems for distributed storage and retrieval systems comprising a plurality of storage-and-computing elements, each storage-and-computing element storing a plurality of pushed content fragments and able to simultaneously retrieve pushed content fragments from a plurality of storage-and-computing elements. Wherein many storage-and-computing elements, within a short duration of receiving a request, can retrieve and play streaming content.

Claims

1. A distributed storage and retrieval system comprising:

a plurality of storage-and-computing elements, each storage-and-computing element storing a plurality of pushed content fragments and able to simultaneously retrieve pushed content fragments from a plurality of storage-and-computing elements; and

wherein up to about ‘

N’

storage-and-computing elements, within ‘

T’

seconds of receiving a request, can retrieve and provide streams that can be played;

wherein ‘

T’

is shorter than about BUF/(3×

R) seconds, ‘

N’

roughly equals the available aggregated uplink bandwidth of the storage-and-computing elements divided by the average streaming bitrate, BUF is the buffered data needed to start playing the content (expressed in Kbits), and ‘

R’

(expressed in Kbps) is the average available uplink bandwidth of a storage-and-computing element.

2. The distributed storage and retrieval system of claim 1, wherein the ‘

N’

storage-and-computing elements can provide the streams starting from approximately arbitrary positions within the contents.

3. The distributed storage and retrieval system of claim 2, wherein the average streaming bitrate is greater than three times ‘

R’ and

greater than 0.5 Mbps.

4. The distributed storage and retrieval system of claim 2, wherein most of the streaming is retrieved from storage-and-computing elements having uplink bandwidths of the same order of magnitude and downlink bandwidths of the same order of magnitude.

5. The distributed storage and retrieval system of claim 1, wherein the plurality of pushed content fragments are distributed between the storage-and-computing elements according to a distribution function and the distribution function features a low autocorrelation and a low correlation with other distribution functions used by the distributed storage and retrieval system.

6. The distributed storage and retrieval system of claim 1, wherein the plurality of pushed content fragments are distributed approximately homogeneously between the storage-and-computing elements that are relevant to the retrieving.

7. The distributed storage and retrieval system of claim 6, wherein ‘

T’

is shorter than about BUF/(8×

R) seconds.

8. The distributed storage and retrieval system of claim 1, wherein the plurality of pushed content fragments are distributed between the storage-and-computing elements according to a pseudo random distribution function.

9. A distributed storage and retrieval system comprising:

a plurality of storage-and-computing elements, each storage-and-computing element storing a plurality of content fragments wherein the plurality of content fragments are distributed a priori approximately homogeneously between the storage-and-computing elements; and

up to about ‘

N’

storage-and-computing elements, within ‘

T’

seconds of receiving a request, can retrieve and provide streaming contents that can be played;

wherein ‘

T’

is shorter than about BUF/(3×

R) seconds, ‘

N’

roughly equals the available aggregated uplink bandwidth of the storage-and-computing elements divided by an average streaming bitrate, BUF is the buffered data needed to start playing the content (expressed in Kbits), and ‘

R’

(expressed in Kbps) is calculated by dividing the total available upload bandwidth of the storage-and-computing elements by the number of the storage-and-computing elements.

10. The distributed storage and retrieval system of claim 9, wherein the content fragments are distributed a priori utilizing fragment distribution information featuring low autocorrelation and low correlation, and the fragment distribution information are also utilized for retrieving the content fragments by the storage-and-computing elements.

11. The distributed storage and retrieval system of claim 10, wherein the storage-and-computing elements are spanned by one network node, and the network node is selected from the group of:

DSLAM, B-RAS, IP services switch, IP services router, or a combination thereof.

12. The distributed storage and retrieval system of claim 9, wherein the ‘

N’

storage-and-computing elements can retrieve and provide the streaming contents starting from approximately arbitrary positions within the contents.

13. A distributed storage and retrieval system comprising:

a plurality of storage-and-computing elements, each storage-and-computing element storing a plurality of pushed content fragments according to its associated storage element type; and

at least a quarter of the storage-and-computing elements comprising means for simultaneously retrieving the pushed content fragments from the plurality of storage-and-computing elements;

wherein about ‘

N’

storage-and-computing elements, within ‘

T’

seconds of receiving a request, can retrieve and provide streaming contents that can be played;

wherein ‘

T’

is shorter than about BUF/(3×

R) seconds, ‘

N’

roughly equals the available aggregated uplink bandwidth of the storage-and-computing elements divided by an average streaming bitrate, BUF is the buffered data needed to start playing the content (expressed in Kbits), and ‘

R’

(expressed in Kbps) is calculated by dividing the total available upload bandwidth of the storage-and-computing elements by the number of the storage-and-computing elements.

14. The distributed storage and retrieval system of claim 13, wherein approximately all of the storage-and-computing elements comprise means for simultaneously retrieving the pushed content fragments.

15. The distributed storage and retrieval system of claim 13, wherein the ‘

N’

storage-and-computing elements can provide up to about ‘

N’

different streaming contents starting from approximately arbitrary positions within the contents.

16. The distributed storage and retrieval system of claim 13, wherein the plurality of pushed content fragments are distributed between the storage-and-computing elements according to a distribution function featuring a low autocorrelation and a low correlation with other distribution functions utilized by the distributed storage and retrieval system, and the storage-and-computing elements are accessed via Internet protocol.

17. A method comprising:

receiving, by ‘

N’

storage-and-computing elements, at least one a priori fragment distribution information;

utilizing the a priori fragment distribution information for retrieving approximately simultaneously ‘

N’

content streams comprised of a plurality of pushed content fragments from a plurality of storage-and-computing elements;

wherein ‘

N’

roughly equals the available aggregated uplink bandwidth of the plurality of storage-and-computing elements divided by an average streaming bitrate, and the amount of time ‘

T’

that passes between requesting a content stream and being able to play the content stream is shorter than about BUF/(3×

R) seconds;

whereby BUF is the buffered data needed to start playing the content (expressed in Kbits), and ‘

R’

(expressed in Kbps) is the average available uplink bandwidth of a storage-and-computing element.

18. The method of claim 17, wherein the a priori fragment distribution information features a low autocorrelation and a low correlation with other a priori fragment distribution information.

19. The method of claim 17, wherein the average streaming bitrate is higher than ‘

R’

.

20. A method comprising:

receiving a priori fragment distribution information;

utilizing the a priori fragment distribution information for retrieving, by a peer client, a plurality of pushed content fragments from a plurality of peer clients; and

providing a video stream that can be played within about ‘

T’

seconds of receiving a request, wherein the video stream can be provided from approximately an arbitrary starting position within the content, and ‘

T’

is at least two times shorter than what could be achieved if only one peer client storing all of the pushed content fragments was to transmit the streaming content.

21. The method of claim 20, wherein the fragment distribution information features a low autocorrelation and a low correlation with other fragment distribution information.

22. The method of claim 20, wherein the plurality of peer clients having uplink bandwidths approximately similar to the uplink bandwidth of the retrieving peer client.

23. The method of claim 20, further comprising retrieving, by ‘

N’

peer clients, a plurality of pushed content fragments, providing ‘

N’

video streams that can be played within about BUF/(3×

R) seconds of receiving a request;

wherein ‘

N’

roughly equals the available aggregated uplink bandwidth of the plurality of peer clients divided by an average streaming bitrate, BUF is the buffered data needed to start playing the content (expressed in Kbits), and ‘

R’

(expressed in Kbps) is calculated by dividing the total available upload bandwidth of the peer clients by the number of the peer clients.