commit | c90a2a2ce843e685da4fb8e32224be97c8bbc6ee | [log] [tgz] |
---|---|---|
author | Joseph Koshakow <jkosh44@users.noreply.github.com> | Wed Oct 30 09:30:19 2019 -0400 |
committer | Keith Turner <kturner@apache.org> | Wed Oct 30 09:30:19 2019 -0400 |
tree | 59f9f34f8032ae328aa9f527d5ed29a51edf623e | |
parent | 1b6b01b0dcfd09d3f006fad054c861425599de1c [diff] |
Parallelize TransactionImpl.readUnread() (#1080) When a transaction only writes to a row+col and then has a collision Fluo will read the row+col after the collision to look for orphaned locks. This commit parallelizes this behavior by reading all row+cols at once. This commit accomplishes this by using a ParallelSnapshotScanner instead of a SnapshotScanner. Fixes #948
Apache Fluo is a distributed processing system that lets users make incremental updates to large data sets. With Apache Fluo, users can set up workflows that execute cross node transactions when data changes. These workflows enable users to continuously join new data into large existing data sets without reprocessing all data. Apache Fluo is built on Apache Accumulo. Check out the Fluo project website for news and general information.