systemml-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Matthias Boehm (JIRA)" <>
Subject [jira] [Created] (SYSTEMML-2169) Spark nary cbind/rbind with broadcasts
Date Wed, 07 Mar 2018 05:44:00 GMT
Matthias Boehm created SYSTEMML-2169:

             Summary: Spark nary cbind/rbind with broadcasts
                 Key: SYSTEMML-2169
             Project: SystemML
          Issue Type: Task
            Reporter: Matthias Boehm

The introduction of nary cbind and rbinds in SYSTEMML-1986 added support for operations like
{{E = cbind(A,B,C,D)}} which concatenates the matrices A, B, C, D column-wise without the
need for intermediates as requires by traditional binary cbind operations ({{cbind(cbind(cbind(A,B),C),D)}}).
SystemML also provides rewrites to automatically collapse chains of cbind or rbind operations
into their nary counter-parts. 

However, for distributed spark operations, the binary cbind is still much better optimized
than the nary operation, which only provides a general case operation based on repartition

This tasks aims to address this by extending {{BuiltinNarySPInstruction}} at runtime level.
Given the unlimited number of inputs, this runtime approach seems more appropriate than dedicated
physical operations at compiler level. In detail, we need to evaluate if a subset of input
fits into the broadcast budget, and if so provide alternative code path for nary cbind/rbind
operations with broadcast joins.

Note that distributed codegen operations have a similar characteristics of unlimited inputs
and already leverage broadcast variables when possible. Hence, we can probably use a similar
approach as done in {{SpoofSPInstruction}}.

This message was sent by Atlassian JIRA

View raw message