Amazon.com, Inc.
Predictive caching of identical starting sequences in content

Abstract:

Disclosed are various embodiments for predictive caching of identical starting sequences in content. A content item library is scanned to identify an initial portion shared by multiple content items. The initial portion is extracted from a first content item. It is determined that a second content item is to be predictively cached by a client. The initial portion of the first content item is sent to the client in place of the initial portion of the second content item.