Skip to content
This repository has been archived by the owner on Oct 31, 2018. It is now read-only.

Issues: NetSys/spark-monotasks

22 Open 11 Closed
22 Open 11 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

NoSuchFileException when HdfsDiskMonotask tries to access local disk
#34 opened Nov 21, 2017 by hzhua
NetworkUtilization is not correctly decoded from JSON
#33 opened May 3, 2016 by kayousterhout
Add off-heap memory use to the UI
#32 opened Nov 13, 2015 by kayousterhout
DiskScheduler sometimes hangs on shutdown
#31 opened Sep 22, 2015 by kayousterhout
Fix Java.util.concurrentModificationException in Hadoop code path
#30 opened Sep 21, 2015 by kayousterhout
@ccanel
2
Clean up temporary data when a monotask fails.
#29 opened Sep 20, 2015 by ccanel
Write unit test(s) for new code to read/write all shuffle data in a single file enhancement
#28 opened Sep 17, 2015 by kayousterhout
@kayousterhout
Handle scenario where the entire input file isn't needed enhancement
#27 opened Aug 6, 2015 by kayousterhout
Bug in BlockManager's concurrency control
#26 opened Jul 17, 2015 by kayousterhout
@ccanel
2
Remove all support for external shuffle and associated dynamic executor pool sizing low-priority inconsistency with spark
#21 opened Apr 20, 2015 by kayousterhout
All types of ExecutorBackend should implement updateFreeCores enhancement
#19 opened Mar 21, 2015 by kayousterhout
JsonProtcol.scala should properly decode BlockDeviceUtilizations bug
#18 opened Mar 21, 2015 by kayousterhout
Throw exception in HadoopRDD.compute() enhancement
#17 opened Mar 17, 2015 by kayousterhout
@ccanel
4
Implement a control mechanism to prevent two macrotasks from loading or computing the same block at the same time. enhancement
#14 opened Mar 7, 2015 by ccanel
Reimplement the BlockManager's ability to drop blocks from memory to disk if the MemoryStore becomes full. enhancement
#13 opened Feb 22, 2015 by ccanel
Remove the BlockManager's ability to fetch remote blocks without using a NetworkMonotask. enhancement
#12 opened Feb 22, 2015 by ccanel
Fail all dependent monotasks when a monotask fails enhancement
#10 opened Feb 3, 2015 by kayousterhout
1
Support checkpoint RDDs bug
#9 opened Jan 28, 2015 by kayousterhout
Read Broadcast variables using a Network Monotask
#7 opened Jan 26, 2015 by kayousterhout
Add back support to periodically update the master about metrics for running tasks enhancement
#6 opened Jan 22, 2015 by kayousterhout
Support killing tasks enhancement
#4 opened Jan 22, 2015 by kayousterhout
Flushing to disk throws a ClassCastException when value is a single value (not an array or serialized buffer) bug
#2 opened Dec 5, 2014 by kayousterhout
@kayousterhout
1
ProTip! Follow long discussions with comments:>50.