Other

Microsoft Technology Licensing, LLC

The Trustees of Princeton

Ironport System, Inc.

Vivek Pai

Anirudh Badam

Ming Zhang

Ratul Mahajan

Sunghwan Ihm

KyoungSoo Park

Willy Zwaenepoel

Alan L Cox

Lindsey Philip Poole

Y Charlie Hu

Vijay S Pai

Larry L Peterson

A virtual address pager and method for use with a bulk erase memory is disclosed. The virtual address pager includes a page protection controller configured with a heap manager interface configured to receive only bulk erase memory-backed page requests for a plurality of memory pages. A RAM object cache controller is configured to store and bulk write data for a portion of the bulk erase memory. The page protection controller may have an operating system interface configured to generate a page memory access permission for each of the plurality of memory pages. The page protection controller may be configured to receive a virtual memory allocation request and generate the page memory access permission based on the virtual memory allocation request.

Virtual address pager and method for use with a bulk erase memory

The illustrative embodiments described herein are directed to a system, methods, and computer program product, for network acceleration and efficient indexing for caching file systems. For example, the disclosed embodiments achieve (1) high compression rate, (2) low disk seeks, and (3) low memory pressure using by introducing novel features such as, but not limited to, multi-resolution chunking and intelligent load shedding. In addition, the disclosed embodiments introduces a family of HashCache policies that enable large size web caching with little or no use of main memory.

Systems and methods for network acceleration and efficient indexing for caching file systems

The claimed subject matter provides systems and/or methods that discover comparative differences in performance of Internet Service Provider networks. The system can include devices, components, and/or processes that partitions measurement processes into cycles and measures at least one Internet Service Provider network every cycle, develops and distributes to interrogation nodes a list of destination Internet Protocol (IP) addresses to interrogate, and utilizes results delivered or retrieved from interrogation nodes to generate a report on the comparative performance of Internet Service Provider networks queried.

Uncovering the differences in backbone networks

The invention provides a storage and retrieval routine for Web objects. A Web page typically includes several objects such as text, images and hyper-links to other Web pages. Each Web page and its objects are usually stored in separate files. A storage and retrieval routine stores objects that correspond to the same Web page in co-located positions on a disk. Therefore, when the Web page and its embedded and hyper-linked Web objects are retrieved, the seek time is reduced, thereby reducing the object retrieval time as perceived by a computer user and increasing the number of requests per second that a Web content server can deliver.

Storage and retrieval system for WEB cache

Growing community of inventors

Princeton, NJ, United States of America

Vivek Pai