Google Inc.

Microsoft Technology Licensing, LLC

Matthew B Tolton

Hossein Ahmadi

Michael Entin

Matthew A Armstrong

David L Detlefs

Timothy Lawrence Harris

Michael M Magruder

Yosseff Levanoni

Narayanan Shivakumar

Weirong Zhu

Sergey Melnik

Theodore Vassilakis

Andrey Gubarev

Geoffrey M Romer

This disclosure generally relates to data delivery in distributed applications. One example method includes identifying a data source associated with a shuffle operation, the data source configured provide data from a data set associated with the shuffle operation; identifying a data sink associated with the shuffle operation, the data sink configured to receive data provided by the data source; associating a shuffler component with the shuffle operation, the shuffler component configured to receive data from the data source and provide the data to the data sink; receiving, by the shuffler component, a first data portion from the data source; providing, by the shuffler component, the first data portion to the data sink; receiving, by the shuffler component, a second data portion from the data source, the second data portion being received from the data source prior to or concurrent with providing the first data portion to the data sink.

Data delivery

Methods, systems, and apparatus for allocating, by a source of one or more sources, a segment of a data file of a transient memory for exclusive access by the source, the transient memory being a distributed in-memory file system that supports remote direct memory access; writing, by the source, data from an initial partition to one or more blocks within the allocated segment of the data file, wherein a portion of the initial partition is written to a first block of the one or more blocks; publishing, by the source, the segment of the data file of the transient memory to be accessible for reading by one or more sinks; and reading by a particular sink of the one or more sinks, a particular block of the published segment of the data file of the transient memory, wherein the particular block is associated with the particular sink.

Repartitioning data in a distributed computing system

A software transactional memory system is provided with overflow handling. The system includes a global version counter with an epoch number and a version number. The system accesses the global version counter prior to and subsequent to memory accesses of transactions to validate read accesses of the transaction. The system includes mechanisms to detect global version number overflow and may allow some or all transactions to execute to completion subsequent to the global version number overflowing. The system also provides publication, privatization, and granular safety properties.

STM with global version overflow handling

Growing community of inventors

Seattle, WA, United States of America

Matthew B Tolton