Microsoft Technology Licensing, LLC

Ashit Gosalia

Qingqing Zhou

Srikumar Rangarajan

Aleksandras Surna

Artem A Oks

Wey Guy

Cipri Clinciu

Oleksandr Gololobov

Architecture that includes an index creation algorithm that utilizes available resources and dynamically adjusts to successfully scale with increased resources and be able to do so for any data distribution. The resources can be processing resources, memory, and/or input/output, for example. A finer level of granularity, called a segment, is utilized to process tuples in a partition while creating an index. The segment also aligns with compression techniques for the index. By choosing an appropriate size for a segment and using load balancing the overall time for index creation can be reduced. Each segment can then be processed by a single thread thereby limiting segment skew. Skew is further limited by breaking down the work done by a thread into parallelizable stages.

Scalable index build techniques for column stores

Various technologies and techniques are disclosed for improving performance of parallel scans. Disk head randomization that occurs when performing a parallel scan is minimized by assigning a worker entity to each disk involved in the parallel scan, and by ensuring data is only accessed on a respective disk by the worker entity assigned to the disk. A parallel scan can be performed that is NUMA aware by ensuring a particular sub-set of data is resident in the same memory node during each parallel scan, and by ensuring the particular sub-set of data is processed by a worker entity assigned to a node in which the sub-set of data is resident. A process for performing a parallel scan involves breaking up work into sub-sets, assigning work to each worker entity that corresponds to a respective disk, and having the worker entities process the assigned work to complete the parallel scan.

Techniques for improving parallel scan operations

Growing community of inventors

Sammamish, WA, United States of America

Ashit Gosalia