http://rdf.ncbi.nlm.nih.gov/pubchem/patent/KR-101724247-B1
Outgoing Links
Predicate | Object |
---|---|
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F9-522 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F9-30087 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F9-3851 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F9-3887 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F9-38 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F9-30 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F9-52 |
filingDate | 2014-08-07-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2017-04-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2017-04-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | KR-101724247-B1 |
titleOfInvention | Barrier synchronization with dynamic width calculation |
abstract | The sequencer of the processing unit determines, at execution time, a barrier width of a barrier operation for group threads, wherein the barrier width is less than the total number of threads in a group of threads, Executes the data parallel code on the calculation units. In response to each thread executing a barrier operation in a subgroup of a group of threads, including the same number of threads as the barrier width, the sequencer causes the subgroup of groups of threads to be moved to another thread May be enabled to run on one or more processors past the barrier operation, without waiting to execute barrier operations, and the subgroup of the group of threads is less than the total number of threads in the group of threads. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/KR-102501187-B1 http://rdf.ncbi.nlm.nih.gov/pubchem/patent/KR-20210124081-A |
priorityDate | 2013-08-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 21.