Changeset c9019ce
- Timestamp:
- Jun 30, 2023, 1:48:25 PM (20 months ago)
- Branches:
- master
- Children:
- 599dc6a
- Parents:
- 70e47fec
- Location:
- doc/theses/colby_parsons_MMAth
- Files:
-
- 2 edited
Legend:
- Unmodified
- Added
- Removed
-
doc/theses/colby_parsons_MMAth/local.bib
r70e47fec rc9019ce 170 170 } 171 171 172 @misc{linux:iouring, 173 author = "Linux man pages", 174 title = "io\_uring(7) - Linux manual page", 175 howpublished = {\href{https://man7.org/linux/man-pages/man7/io\_uring.7.html}}, 176 note = "[Online; accessed 23-May-2023]" 177 } 178 172 179 @article{Ichbiah79, 173 180 title={Preliminary Ada reference manual}, … … 188 195 } 189 196 190 191 197 @misc{openmp, 198 author = "OpenMP", 199 title = "OPENMP API Specification", 200 howpublished = {\href{https://www.openmp.org/spec-html/5.0/openmpch1.html}}, 201 note = "[Online; accessed 23-May-2023]" 202 } 203 204 @techreport{wilson94, 205 title={The suif compiler system: a parallelizing and optimizing research compiler}, 206 author={Wilson, Robert and French, Robert and Wilson, Christopher and Amarasinghe, Saman and Anderson, Jennifer and Tjiang, Steve and Liao, Shih-Wei and Tseng, Chau-Wen and Hall, Mary and Lam, Monica and others}, 207 year={1994}, 208 institution={Stanford University Technical Report No. CSL-TR-94-620} 209 } 210 211 @misc{haskell:parallel, 212 author = "Haskell Wiki", 213 title = "Parallel Haskell", 214 howpublished = {\href{https://wiki.haskell.org/Parallel}}, 215 note = "[Online; accessed 23-May-2023]" 216 } 217 218 @misc{gcc:atomics, 219 author = "GCC team", 220 title = "Built-in Functions for Memory Model Aware Atomic Operations", 221 howpublished = {\href{https://gcc.gnu.org/onlinedocs/gcc/_005f_005fatomic-Builtins.html}}, 222 note = "[Online; accessed 23-May-2023]" 223 } 224 -
doc/theses/colby_parsons_MMAth/text/conclusion.tex
r70e47fec rc9019ce 15 15 This thesis scratches the surface of implicit concurrency by providing an actor system. 16 16 There is room for more implicit concurrency tools in \CFA. 17 User-defined implicit concurrency in the form of annotated loops or recursive functions exists in many other languages ~\cite{} and could be implemented and expanded on in \CFA.18 Additionally, the problem of automatic parallelism of sequential programs via the compiler is an interesting research space that other languages have approached~\cite{ } that could also be explored in \CFA.17 User-defined implicit concurrency in the form of annotated loops or recursive functions exists in many other languages and libraries~\cite{uC++,openmp} and could be implemented and expanded on in \CFA. 18 Additionally, the problem of automatic parallelism of sequential programs via the compiler is an interesting research space that other languages have approached~\cite{wilson94,haskell:parallel} that could also be explored in \CFA. 19 19 20 20 21 21 \subsection{Synchronously Multiplexing System Calls} 22 There are many tools that try to sychronize on or asynchronously check I/O, since improvements in this area pay dividends in many areas of computer science~\cite{ }. %cite all the poll/iouring utilities22 There are many tools that try to sychronize on or asynchronously check I/O, since improvements in this area pay dividends in many areas of computer science~\cite{linux:select,linux:poll,linux:epoll,linux:iouring}. 23 23 Research on improving user-space tools to synchronize over I/O and other system calls is an interesting area to explore in the world of concurrent tooling. 24 24 25 \subsection{Better Atomics} 25 \subsection{Advanced Actor Stealing Heuristics} 26 In this thesis, two basic victim selection heuristics were chosen when implementing the work stealing actor system. Good victim selection is an active area of work stealing research, especially when taking into account NUMA effects and cache locality~\cite{barghi18,wolke17}. The actor system in \CFA is modular and exploration of other victim selection heuristics for queue stealing in \CFA could constitute future work. The other question that arises with work stealing is: When should a worker thread steal? Work stealing systems can often be too aggressive when stealing, causing the cost of the steal to be higher than what is saved by sharing the work. In the presented actor system, a worker thread steals work after it checks all its work queues for work twice and sees them all empty. Given that thief threads often have cycles to spare, there is room for a more nuanced approach to choosing when to steal. 27 28 \subsection{Better Atomic Operations} 26 29 When writing low level concurrent programs, expecially lock/wait-free programs, low level atomic instructions need to be used. 27 In C, the gcc-builtin atomics~\cite{ } are commonly used, but leave much to be desired.30 In C, the gcc-builtin atomics~\cite{gcc:atomics} are commonly used, but leave much to be desired. 28 31 Some of the problems include the following. 29 32 Archaic and opaque macros often have to be used to ensure that atomic assembly is generated instead of locks.
Note: See TracChangeset
for help on using the changeset viewer.