| | |
Summary: Deadlock Avoidance for Streaming Computations
with Filtering
Peng Li Kunal Agrawal Jeremy Buhler Roger D. Chamberlain
Dept. of Computer Science and Engineering
Washington University in St. Louis
{pengli, kunal, jbuhler, roger}@wustl.edu
ABSTRACT
The paradigm of computation on streaming data has re-
ceived considerable recent attention. Streaming computa-
tions can be efficiently parallelized using systems of com-
puting nodes organized in dataflow-like architectures. How-
ever, when these nodes have the ability to filter, or dis-
card, some of their inputs, a system with finite buffering
is vulnerable to deadlock. In this paper, we formalize a
model of streaming computation systems with filtering, de-
scribe precisely the conditions under which such systems
may deadlock, and propose provably correct mechanisms
to avoid deadlock. Our approach relies on adding extra
"dummy" tokens to the data streams and does not require
global run-time coordination among nodes or dynamic resiz-
|
