… significant simplification of the design process. However, two major roadblocks have thus
far … presented at several annual conferences, known as FMPC, ICPP, IPPS, SPAA, SPDP (now …

… Specifically, two parallel instantiations of walk may attempt to update the shared global
variable output list in parallel at line 10. The traditional solution to fixing this kind of data race is to …

… SPAA 98 Puerto Vallarta Mexico Copyright ACM 1998 O-89791-989-0/98/ 6...$5.00 … the
“bounded tags” algorithm [28]. We claim that the deque implementation presented above is 122 …

… Because multithreaded computations with arbitrary dependencies can be impossible to
schedule efficiently [Blumofe and Leiserson 1998], we study subclasses of general …

… , ample “parallel slackness” [28] … two clones of each procedure--a fast clone and a slow
clone. The fast clone operates much as does the C elision and has little support for parallelism. …

… stealing algorithm, define seriesparallel and nested-parallel computations … 6, 9] we represent
a multithreaded computation as a directed acyclic graph, a dag, of instructions (see Figure 2…

… To help introduce the reader to our deterministic locking algorithm, we present two versions.
The first, presented in Section 3.2.1, is a simplified algorithm that does not support nested …

… SPAA 98 Puerto Vallarta Mexico Copyright ACM 1998 O-89791-989-0/98/ 6...$5.00 … is
about three times the overhead of a C function call [28]. In Figure 7 we show the times for union …

… -theoretic algorithms for illustrative purposes, we show experimental results on two shared-…
parallel algorithms with efficient fine-grained synchronization may yield good performance. …

… SPAA 98 Puerto Vallarta Mexico Copyright ACM 1998 O-89791-989-0/98/ … of Netzer and
Miller [28], upparent data races-those that appear to occur in a computation according to the …