[
https://issues.apache.org/jira/browse/FLINK-1536?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17503714#comment-17503714
]
Amit Srivastava edited comment on FLINK-1536 at 3/9/22, 7:32 PM:
-----------------------------------------------------------------
Hi [~vkalavri] and [~ivan.mushketyk]
Just wanted to know what is the current state of this JIRA? I am working on a
problem that has a big graph constructed using geospatial entities. The
vertices denote the geospatial entities _(like building, roads)_ and edges
describes the relationship _(connection between building and road-name)_
between them. Partition using the spatial hash for me will increase the system
performance significantly.
Is it possible to provide a custom partition to Gelly instead of using the
default hash-based partition? In my scenario, vertices Ids has
[S2CellId|https://github.com/google/s2-geometry-library-java/blob/master/src/com/google/common/geometry/S2CellId.java#L192]
as a prefix to Id. Is it possible in Gelly to use
[partitionByRange|https://github.com/apache/flink/blob/master/flink-java/src/main/java/org/apache/flink/api/java/DataSet.java#L1363]
to partition based on Id prefix?
was (Author: JIRAUSER286363):
Hi [~vkalavri] and [~ivan.mushketyk]
Just wanted to know what is the current state of this JIRA? I am working on a
problem that has a big graph constructed using geospatial entities. The
vertices denote the geospatial entities and edges describes the relationship
between them. Partition using the spatial hash for me will increase the system
performance significantly. Is it possible to provide a custom partition to
Gelly instead of using the default hash-based partition?
Additionally, the vertices Ids of my project has
[S2CellId|https://github.com/google/s2-geometry-library-java/blob/master/src/com/google/common/geometry/S2CellId.java#L192]
as a prefix. Is it possible in Gelly to use
[partitionByRange|https://github.com/apache/flink/blob/master/flink-java/src/main/java/org/apache/flink/api/java/DataSet.java#L1363]
easily?
> Graph partitioning operators for Gelly
> --------------------------------------
>
> Key: FLINK-1536
> URL: https://issues.apache.org/jira/browse/FLINK-1536
> Project: Flink
> Issue Type: New Feature
> Components: Library / Graph Processing (Gelly)
> Reporter: Vasia Kalavri
> Assignee: Ivan Mushketyk
> Priority: Not a Priority
> Labels: stale-assigned
>
> Smart graph partitioning can significantly improve the performance and
> scalability of graph analysis applications. Depending on the computation
> pattern, a graph partitioning algorithm divides the graph into (maybe
> overlapping) subgraphs, optimizing some objective. For example, if
> communication is performed across graph edges, one might want to minimize the
> edges that cross from one partition to another.
> The problem of graph partitioning is a well studied problem and several
> algorithms have been proposed in the literature. The goal of this project
> would be to choose a few existing partitioning techniques and implement the
> corresponding graph partitioning operators for Gelly.
> Some related literature can be found [here|
> http://www.citeulike.org/user/vasiakalavri/tag/graph-partitioning].
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
