[SPARK-732] Recomputation of RDDs may result in duplicated accumulator updates

• Type: Bug
• Status: Closed
• Priority: Blocker
• Resolution: Won't Fix
• Affects Version/s: 0.6.2, 0.7.0, 0.7.1, 0.7.2, 0.7.3, 0.8.0, 0.8.1, 0.8.2, 0.9.0, 1.0.1, 1.1.0
• Component/s: Spark Core
• Assignee: Nan Zhu
• Reporter: Josh Rosen
• Created: 14/Apr/13 22:50
• Updated: 24/Feb/16 22:23
• Resolved: 27/Nov/14 01:32

Description

Currently, Spark doesn't guard against duplicated updates to the same accumulator due to recomputations of an RDD. For example:

    val acc = sc.accumulator(0)
    data.map(x => acc += 1; f(x))
    data.count()
    // acc should equal data.count() here
    data.foreach{...}
    // Now, acc = 2 * data.count() because the map() was recomputed.

I think that this behavior is incorrect, especially because this behavior allows the additon or removal of a cache() call to affect the outcome of a computation.

There's an old TODO to fix this duplicate update issue in the DAGScheduler code.

I haven't tested whether recomputation due to blocks being dropped from the cache can trigger duplicate accumulator updates.

Hypothetically someone could be relying on the current behavior to implement performance counters that track the actual number of computations performed (including recomputations). To be safe, we could add an explicit warning in the release notes that documents the change in behavior when we fix this.

Ignoring duplicate updates shouldn't be too hard, but there are a few subtleties. Currently, we allow accumulators to be used in multiple transformations, so we'd need to detect duplicate updates at the per-transformation level. I haven't dug too deeply into the scheduler internals, but we might also run into problems where pipelining causes what is logically one set of accumulator updates to show up in two different tasks (e.g. rdd.map(accum += x; ...) and rdd.map(accum += x; ...).count() may cause what's logically the same accumulator update to be applied from two different contexts, complicating the detection of duplicate updates).

Issue Links

• is related to: SPARK-3628 Don't apply accumulator updates multiple times for tasks in result stages
• is related to: SPARK-5490 KMeans costs can be incorrect if tasks need to be rerun
• is related to: SPARK-3642 Better document the nuances of shared variables
• links to: [Github] Pull Request #2524 (CodingCat)