If you are talking about a tree, then the RDDs are nodes, and the dependencies
are the edges.
If you are talking about a DAG, then the partitions in the RDDs are the nodes,
and the dependencies between the partitions are the edges.
On Thu, Apr 16, 2020 at 4:02 PM, Mania Abdi < abdi...@husky.neu
Is it correct to say, the nodes in the DAG are RDDs and the edges are
computations?
On Thu, Apr 16, 2020 at 6:21 PM Reynold Xin wrote:
> The RDD is the DAG.
>
>
> On Thu, Apr 16, 2020 at 3:16 PM, Mania Abdi wrote:
>
>> Hello everyone,
>>
>> I am implementing a caching mechanism for analytic wor
The RDD is the DAG.
On Thu, Apr 16, 2020 at 3:16 PM, Mania Abdi < abdi...@husky.neu.edu > wrote:
>
> Hello everyone,
>
> I am implementing a caching mechanism for analytic workloads running on
> top of Spark and I need to retrieve the Spark DAG right after it is
> generated and the DAG schedule
Hello everyone,
I am implementing a caching mechanism for analytic workloads running on top
of Spark and I need to retrieve the Spark DAG right after it is generated
and the DAG scheduler. I would appreciate it if you could give me some
hints or reference me to some documents about where the DAG i
Hi again,
Does anyone have thoughts on either the idea or the implementation?
Thanks,
Andrew
On Thu, Apr 9, 2020 at 11:32 PM Andrew Melo wrote:
>
> Hi all,
>
> I've opened a WIP PR here https://github.com/apache/spark/pull/28159
> I'm a novice at Scala, so I'm sure the code isn't idiomatic, but
Hi,
While trying to understand the relationship of BlockManager
and ShuffleManager I found that ShuffleManager is used for shuffle block
data [1] (and that makes sense).
What I found quite surprising is that BlockManager can call getLocalBytes
for non-shuffle blocks that in turn does...fetching s
Hi Jungtaek,
Thanks a lot for your answer. What you're saying reflects my understanding
perfectly. There's a small change, but makes understanding where rules are
used much simpler (= less confusing). I'll propose a PR and see where it
goes from there. Thanks!
Pozdrawiam,
Jacek Laskowski
htt