Stream-based data deduplication with peer node prediction

Patent metadata

Field	Value
Publication number	US-10778760-B2
Application number	US-201916267549-A
Country	US
Kind code	B2
Filing date	Feb 5, 2019
Priority date	Dec 27, 2012
Publication date	Sep 15, 2020
Grant date	Sep 15, 2020

Abstract

Stream-based data deduplication is provided in a multi-tenant shared infrastructure but without requiring “paired” endpoints having synchronized data dictionaries. Data objects processed by the dedupe functionality are treated as objects that can be fetched as needed. As such, a decoding peer does not need to maintain a symmetric library for the origin. Rather, if the peer does not have the chunks in cache that it needs, it follows a conventional content delivery network procedure to retrieve them. In this way, if dictionaries between pairs of sending and receiving peers are out-of-sync, relevant sections are then re-synchronized on-demand. The approach does not require that libraries maintained at a particular pair of sender and receiving peers are the same. Rather, the technique enables a peer, in effect, to “backfill” its dictionary on-the-fly. On-the-wire compression techniques are provided to reduce the amount of data transmitted between the peers.

First claim

What is claimed is as follows:

1. A method operative in an overlay network, comprising;
associating a receiving peer with an edge cache, the receiving peer having an associated sending peer associated with an origin server;
maintaining a directed cyclic graph in association with each of the receiving and sending peers, wherein the directed cyclic graph comprises a set of nodes and edges that collectively represent temporal and ordered relationships among blocks of data that have been seen in a data stream by the respective peer, wherein each node represents a chunk of data, and an edge between nodes represents a transition that the respective peer has seen in the data stream with respect to that chunk, the directed cyclic graph being annotated with information from which the respective peer can generate a prediction about blocks of data that are subject to a stream-based data deduplication;
using the directed cyclic graphs to predict future occurrences of data based upon patterns of access in prior-seen data; and
using results of the prediction to pre-warm the edge cache to reduce round trips and latency between the sending and receiving peers.

2. The method as described in claim 1 wherein the information is a page node annotation.

3. The method as described in claim 1 wherein the information is a composed-of-edge annotation.

4. The method as described in claim 1 wherein the information is a leads-to-edge annotation.

5. The method as described in claim 4 wherein the leads-to-edge annotation includes an associated edge heat index.

6. The method as described in claim 1 wherein the edge cache is a cache in the overlay network that from a network perspective is located near a requesting end user.

7. The method as described in claim 6 wherein the edge cache is pre-warmed with content anticipated to be requested by the requesting end user.

8. The method as described in claim 7 wherein the content is no-store content.

9. The method as described in claim 7 wherein the content is one of: an embedded page object, and a linked page comprising an HTML container and one or more embedded objects.

10. The method as described in claim 7 wherein the content is an HTML page and its embedded object, and wherein the content is delivered from the origin to the requesting end user in a one round-trip operation.

11. The method as described in claim 1 wherein the overlay network is a content delivery network (CDN).

12. The method as described in claim 11 wherein the CDN comprises a server infrastructure shared by a set of participating content providers.

Assignees

• Akamai Tech Inc

Inventors

• Leighton F Thomson
• Champagne Andrew F
• Gero Charles E
• Podjarny Guy

Classifications

• H04L67/108 Primary

  characterised by resources being split in blocks or fragments · CPC title

• H04L67/5681 Primary

  Pre-fetching or pre-delivering data based on network characteristics · CPC title

• H04L67/01

  Protocols · CPC title

• H04L67/42

  Electricity · mapped topic

• H04L67/2847

  Electricity · mapped topic