|
From: Bryan T. <br...@sy...> - 2015-08-31 19:36:08
|
That is IO Wait (or sometimes GC) associated with flushing high level nodes in the B+Tree to the disk. Given the data I see above, it looks like IO Wait was probably spiking. Are you running against local SSD? If you are on EC2, have you exceeded your write rate throughput for the instances? There are faster ways to drop an entire triple store or quad store. But dropping a large named graph out of a set of named graphs in a quad store will incur a lot of IO. Bryan ---- Bryan Thompson Chief Scientist & Founder SYSTAP, LLC 4501 Tower Road Greensboro, NC 27410 br...@sy... http://blazegraph.com http://blog.bigdata.com <http://bigdata.com> http://mapgraph.io Blazegraph™ <http://www.blazegraph.com/> is our ultra high-performance graph database that supports both RDF/SPARQL and Tinkerpop/Blueprints APIs. Blazegraph is now available with GPU acceleration using our disruptive technology to accelerate data-parallel graph analytics and graph query. CONFIDENTIALITY NOTICE: This email and its contents and attachments are for the sole use of the intended recipient(s) and are confidential or proprietary to SYSTAP. Any unauthorized review, use, disclosure, dissemination or copying of this email or its contents or attachments is prohibited. If you have received this communication in error, please notify the sender by reply email and permanently delete all copies of the email and its contents and attachments. On Mon, Aug 31, 2015 at 1:56 PM, Jeremy J Carroll <jj...@sy...> wrote: > We have had big problems dropping a large graph, I have just been looking > at the logs and we seem to have many entries like > > WARN : 199265281 com.bigdata.journal.Journal.executorService942 > com.bigdata.btree.AbstractBTree.writeNodeRecursive(AbstractBTree.java:3716): > wrote: name=.spo.OCSP, 2 records (#nodes=1, #leaves=1) in 8048ms : > addrRoot=-1052669065178378143 > WARN : 199301423 com.bigdata.journal.Journal.executorService942 > com.bigdata.btree.AbstractBTree.writeNodeRecursive(AbstractBTree.java:3716): > wrote: name=.spo.OCSP, 2 records (#nodes=1, #leaves=1) in 36141ms : > addrRoot=-1052669099538116489 > WARN : 199817295 com.bigdata.journal.Journal.executorService1068 > com.bigdata.btree.AbstractBTree.writeNodeRecursive(AbstractBTree.java:3716): > wrote: name=.spo.OCSP, 1 records (#nodes=1, #leaves=0) in 21813ms : > addrRoot=-1052358706546595628 > WARN : 200111575 com.bigdata.journal.Journal.executorService18 > com.bigdata.btree.AbstractBTree.writeNodeRecursive(AbstractBTree.java:3716): > wrote: name=.spo.OCSP, 2 records (#nodes=1, #leaves=1) in 5953ms : > addrRoot=-1052338112178413386 > WARN : 200197937 com.bigdata.journal.Journal.executorService1070 > com.bigdata.btree.AbstractBTree.writeNodeRecursive(AbstractBTree.java:3716): > wrote: name=.spo.OCSP, 1 records (#nodes=1, #leaves=0) in 10652ms : > addrRoot=-1053867623046964134 > WARN : 200763920 com.bigdata.journal.Journal.executorService942 > com.bigdata.btree.AbstractBTree.writeNodeRecursive(AbstractBTree.java:3716): > wrote: name=.spo.OCSP, 2 records (#nodes=1, #leaves=1) in 6214ms : > addrRoot=-1052630260148849987 > > can someone offer an interpretation? > > thanks > > Jeremy > > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > Bigdata-developers mailing list > Big...@li... > https://lists.sourceforge.net/lists/listinfo/bigdata-developers > > |