Re: Performance impact with ALLOW FILTERING clause.
Hi Asad, Seems to me that your development team will need to remodel the tables sooner than later. This problem can't be left unattended for long once it starts hitting severely. The way Cassandra is, you may want to have them replicate the same table with different PK / structure to suitably embed a WHERE clause in the base query if nothing else works out. Allow filtering is best avoided for routine queries or at max good for ad-hoc analysis not involving arithmetic operation (like count/sum) . Regards Devopam On Thu, Jul 25, 2019, 7:19 PM ZAIDI, ASAD A wrote: > Hello Folks, > > > > I was going thru documentation and saw at many places saying ALLOW > FILTERING causes performance unpredictability. Our developers says ALLOW > FILTERING clause is implicitly added on bunch of queries by spark-Cassandra > connector and they cannot control it; however at the same time we see > unpredictability in application performance – just as documentation says. > > > > I’m trying to understand why would a connector add a clause in query when > this can cause negative impact on database/application performance. Is that > data model that is driving connector make its decision and add allow > filtering to query automatically or if there are other reason this clause > is added to the code. I’m not a developer though I want to know why > developer don’t have any control on this to happen. > > > > I’ll appreciate your guidance here. > > > > Thanks > > Asad > > > > >
Re: Performance impact with ALLOW FILTERING clause.
Spark connector doesn't do the "select * from table;" - it does reads by token ranges, reading the data (see https://github.com/datastax/spark-cassandra-connector/blob/master/spark-cassandra-connector/src/main/scala/com/datastax/spark/connector/rdd/partitioner/CassandraPartition.scala#L14) Jacques-Henri Berthemet at "Thu, 25 Jul 2019 14:18:57 +" wrote: JB> Hi Asad, JB> That’s because of the way Spark works. Essentially, when you execute a Spark job, it pulls the full content of the datastore (Cassandra JB> in your case) in it RDDs and works with it “in memory”. While Spark uses “data locality” to read data from the nodes that have the JB> required data on its local disks, it’s still reading all data from Cassandra tables. To do so it’s sending ‘select * from Table ALLOW JB> FILTERING’ query to Cassandra. JB> From Spark you don’t have much control on the initial query to fill the RDDs, sometimes you’ll read the whole table even if you only JB> need one row. JB> Regards, JB> Jacques-Henri Berthemet JB> From: "ZAIDI, ASAD A" JB> Reply to: "user@cassandra.apache.org" JB> Date: Thursday 25 July 2019 at 15:49 JB> To: "user@cassandra.apache.org" JB> Subject: Performance impact with ALLOW FILTERING clause. JB> Hello Folks, JB> I was going thru documentation and saw at many places saying ALLOW FILTERING causes performance unpredictability. Our developers says JB> ALLOW FILTERING clause is implicitly added on bunch of queries by spark-Cassandra connector and they cannot control it; however at the JB> same time we see unpredictability in application performance – just as documentation says. JB> I’m trying to understand why would a connector add a clause in query when this can cause negative impact on database/application JB> performance. Is that data model that is driving connector make its decision and add allow filtering to query automatically or if there JB> are other reason this clause is added to the code. I’m not a developer though I want to know why developer don’t have any control on JB> this to happen. JB> I’ll appreciate your guidance here. JB> Thanks JB> Asad -- With best wishes,Alex Ott Solutions Architect EMEA, DataStax http://datastax.com/ - To unsubscribe, e-mail: user-unsubscr...@cassandra.apache.org For additional commands, e-mail: user-h...@cassandra.apache.org
Re: Performance impact with ALLOW FILTERING clause.
Hi, did you also consider to “tame” your spark job by reducing it’s executors? Probably the Job will have a longer runtime in exchange to reducing the stress on the Cassandra cluster. Regards Christian Von: "ZAIDI, ASAD A" Antworten an: "user@cassandra.apache.org" Datum: Donnerstag, 25. Juli 2019 um 20:05 An: "user@cassandra.apache.org" Betreff: RE: Performance impact with ALLOW FILTERING clause. Thank you all for your insights. When spark-connector adds allows filtering to a query, it makes the query to just ‘run’ no matter if it is expensive for larger table OR not so expensive for table with fewer rows. In my particular case, nodes are reaching 2TB/per node load in 50 node cluster. When bunch of such queries run , causes impact on server resources. Since allow filtering is an expensive operation - I’m trying find knobs which if I turn, mitigate the impact. What I think , correct me if I am wrong , is – it is query design itself which is not optimized per table design - that in turn causing connector to add allow filtering implicitly. I’m not thinking to add secondary indexes on tables because they’ve their own overheads. kindly share if there are other means which we can use to influence connector not to use allow filtering. Thanks again. Asad From: Jeff Jirsa [mailto:jji...@gmail.com] Sent: Thursday, July 25, 2019 10:24 AM To: cassandra Subject: Re: Performance impact with ALLOW FILTERING clause. "unpredictable" is such a loaded word. It's quite predictable, but it's often mispredicted by users. "ALLOW FILTERING" basically tells the database you're going to do a query that will require scanning a bunch of data to return some subset of it, and you're not able to provide a WHERE clause that's sufficiently fine grained to avoid the scan. It's a loose equivalent of doing a full table scan in SQL databases - sometimes it's a valid use case, but it's expensive, you're ignoring all of the indexes, and you're going to do a lot more work. It's predictable, though - you're probably going to walk over some range of data. Spark is grabbing all of the data to load into RDDs, and it probably does it by slicing up the range, doing a bunch of range scans. It's doing that so it can get ALL of the data and do the filtering / joining / searching in-memory in spark, rather than relying on cassandra to do the scanning/searching on disk. On Thu, Jul 25, 2019 at 6:49 AM ZAIDI, ASAD A mailto:az1...@att.com>> wrote: Hello Folks, I was going thru documentation and saw at many places saying ALLOW FILTERING causes performance unpredictability. Our developers says ALLOW FILTERING clause is implicitly added on bunch of queries by spark-Cassandra connector and they cannot control it; however at the same time we see unpredictability in application performance – just as documentation says. I’m trying to understand why would a connector add a clause in query when this can cause negative impact on database/application performance. Is that data model that is driving connector make its decision and add allow filtering to query automatically or if there are other reason this clause is added to the code. I’m not a developer though I want to know why developer don’t have any control on this to happen. I’ll appreciate your guidance here. Thanks Asad
Re: Performance impact with ALLOW FILTERING clause.
If you're thinking about rewriting your data to be more performant when doing analytics, you might as well go the distance and put it in an analytics friendly format like Parquet. My 2 cents. On Thu, Jul 25, 2019 at 11:01 AM ZAIDI, ASAD A wrote: > Thank you all for your insights. > > > > When spark-connector adds allows filtering to a query, it makes the query > to just ‘run’ no matter if it is expensive for larger table OR not so > expensive for table with fewer rows. > > In my particular case, nodes are reaching 2TB/per node load in 50 node > cluster. When bunch of such queries run , causes impact on server > resources. > > > > Since allow filtering is an expensive operation - I’m trying find knobs > which if I turn, mitigate the impact. > > > > What I think , correct me if I am wrong , is – it is query design itself > which is not optimized per table design - that in turn causing connector > to add allow filtering implicitly. I’m not thinking to add secondary > indexes on tables because they’ve their own overheads. kindly share if > there are other means which we can use to influence connector not to use > allow filtering. > > > > Thanks again. > > Asad > > > > > > > > *From:* Jeff Jirsa [mailto:jji...@gmail.com] > *Sent:* Thursday, July 25, 2019 10:24 AM > *To:* cassandra > *Subject:* Re: Performance impact with ALLOW FILTERING clause. > > > > "unpredictable" is such a loaded word. It's quite predictable, but it's > often mispredicted by users. > > > > "ALLOW FILTERING" basically tells the database you're going to do a query > that will require scanning a bunch of data to return some subset of it, and > you're not able to provide a WHERE clause that's sufficiently fine grained > to avoid the scan. It's a loose equivalent of doing a full table scan in > SQL databases - sometimes it's a valid use case, but it's expensive, you're > ignoring all of the indexes, and you're going to do a lot more work. > > > > It's predictable, though - you're probably going to walk over some range > of data. Spark is grabbing all of the data to load into RDDs, and it > probably does it by slicing up the range, doing a bunch of range scans. > > > > It's doing that so it can get ALL of the data and do the filtering / > joining / searching in-memory in spark, rather than relying on cassandra to > do the scanning/searching on disk. > > > > On Thu, Jul 25, 2019 at 6:49 AM ZAIDI, ASAD A wrote: > > Hello Folks, > > > > I was going thru documentation and saw at many places saying ALLOW > FILTERING causes performance unpredictability. Our developers says ALLOW > FILTERING clause is implicitly added on bunch of queries by spark-Cassandra > connector and they cannot control it; however at the same time we see > unpredictability in application performance – just as documentation says. > > > > I’m trying to understand why would a connector add a clause in query when > this can cause negative impact on database/application performance. Is that > data model that is driving connector make its decision and add allow > filtering to query automatically or if there are other reason this clause > is added to the code. I’m not a developer though I want to know why > developer don’t have any control on this to happen. > > > > I’ll appreciate your guidance here. > > > > Thanks > > Asad > > > > > >
RE: Performance impact with ALLOW FILTERING clause.
Thank you all for your insights. When spark-connector adds allows filtering to a query, it makes the query to just ‘run’ no matter if it is expensive for larger table OR not so expensive for table with fewer rows. In my particular case, nodes are reaching 2TB/per node load in 50 node cluster. When bunch of such queries run , causes impact on server resources. Since allow filtering is an expensive operation - I’m trying find knobs which if I turn, mitigate the impact. What I think , correct me if I am wrong , is – it is query design itself which is not optimized per table design - that in turn causing connector to add allow filtering implicitly. I’m not thinking to add secondary indexes on tables because they’ve their own overheads. kindly share if there are other means which we can use to influence connector not to use allow filtering. Thanks again. Asad From: Jeff Jirsa [mailto:jji...@gmail.com] Sent: Thursday, July 25, 2019 10:24 AM To: cassandra Subject: Re: Performance impact with ALLOW FILTERING clause. "unpredictable" is such a loaded word. It's quite predictable, but it's often mispredicted by users. "ALLOW FILTERING" basically tells the database you're going to do a query that will require scanning a bunch of data to return some subset of it, and you're not able to provide a WHERE clause that's sufficiently fine grained to avoid the scan. It's a loose equivalent of doing a full table scan in SQL databases - sometimes it's a valid use case, but it's expensive, you're ignoring all of the indexes, and you're going to do a lot more work. It's predictable, though - you're probably going to walk over some range of data. Spark is grabbing all of the data to load into RDDs, and it probably does it by slicing up the range, doing a bunch of range scans. It's doing that so it can get ALL of the data and do the filtering / joining / searching in-memory in spark, rather than relying on cassandra to do the scanning/searching on disk. On Thu, Jul 25, 2019 at 6:49 AM ZAIDI, ASAD A mailto:az1...@att.com>> wrote: Hello Folks, I was going thru documentation and saw at many places saying ALLOW FILTERING causes performance unpredictability. Our developers says ALLOW FILTERING clause is implicitly added on bunch of queries by spark-Cassandra connector and they cannot control it; however at the same time we see unpredictability in application performance – just as documentation says. I’m trying to understand why would a connector add a clause in query when this can cause negative impact on database/application performance. Is that data model that is driving connector make its decision and add allow filtering to query automatically or if there are other reason this clause is added to the code. I’m not a developer though I want to know why developer don’t have any control on this to happen. I’ll appreciate your guidance here. Thanks Asad
Re: Performance impact with ALLOW FILTERING clause.
"unpredictable" is such a loaded word. It's quite predictable, but it's often mispredicted by users. "ALLOW FILTERING" basically tells the database you're going to do a query that will require scanning a bunch of data to return some subset of it, and you're not able to provide a WHERE clause that's sufficiently fine grained to avoid the scan. It's a loose equivalent of doing a full table scan in SQL databases - sometimes it's a valid use case, but it's expensive, you're ignoring all of the indexes, and you're going to do a lot more work. It's predictable, though - you're probably going to walk over some range of data. Spark is grabbing all of the data to load into RDDs, and it probably does it by slicing up the range, doing a bunch of range scans. It's doing that so it can get ALL of the data and do the filtering / joining / searching in-memory in spark, rather than relying on cassandra to do the scanning/searching on disk. On Thu, Jul 25, 2019 at 6:49 AM ZAIDI, ASAD A wrote: > Hello Folks, > > > > I was going thru documentation and saw at many places saying ALLOW > FILTERING causes performance unpredictability. Our developers says ALLOW > FILTERING clause is implicitly added on bunch of queries by spark-Cassandra > connector and they cannot control it; however at the same time we see > unpredictability in application performance – just as documentation says. > > > > I’m trying to understand why would a connector add a clause in query when > this can cause negative impact on database/application performance. Is that > data model that is driving connector make its decision and add allow > filtering to query automatically or if there are other reason this clause > is added to the code. I’m not a developer though I want to know why > developer don’t have any control on this to happen. > > > > I’ll appreciate your guidance here. > > > > Thanks > > Asad > > > > >
Re: Performance impact with ALLOW FILTERING clause.
Hi Asad, That’s because of the way Spark works. Essentially, when you execute a Spark job, it pulls the full content of the datastore (Cassandra in your case) in it RDDs and works with it “in memory”. While Spark uses “data locality” to read data from the nodes that have the required data on its local disks, it’s still reading all data from Cassandra tables. To do so it’s sending ‘select * from Table ALLOW FILTERING’ query to Cassandra. From Spark you don’t have much control on the initial query to fill the RDDs, sometimes you’ll read the whole table even if you only need one row. Regards, Jacques-Henri Berthemet From: "ZAIDI, ASAD A" Reply to: "user@cassandra.apache.org" Date: Thursday 25 July 2019 at 15:49 To: "user@cassandra.apache.org" Subject: Performance impact with ALLOW FILTERING clause. Hello Folks, I was going thru documentation and saw at many places saying ALLOW FILTERING causes performance unpredictability. Our developers says ALLOW FILTERING clause is implicitly added on bunch of queries by spark-Cassandra connector and they cannot control it; however at the same time we see unpredictability in application performance – just as documentation says. I’m trying to understand why would a connector add a clause in query when this can cause negative impact on database/application performance. Is that data model that is driving connector make its decision and add allow filtering to query automatically or if there are other reason this clause is added to the code. I’m not a developer though I want to know why developer don’t have any control on this to happen. I’ll appreciate your guidance here. Thanks Asad
Performance impact with ALLOW FILTERING clause.
Hello Folks, I was going thru documentation and saw at many places saying ALLOW FILTERING causes performance unpredictability. Our developers says ALLOW FILTERING clause is implicitly added on bunch of queries by spark-Cassandra connector and they cannot control it; however at the same time we see unpredictability in application performance – just as documentation says. I’m trying to understand why would a connector add a clause in query when this can cause negative impact on database/application performance. Is that data model that is driving connector make its decision and add allow filtering to query automatically or if there are other reason this clause is added to the code. I’m not a developer though I want to know why developer don’t have any control on this to happen. I’ll appreciate your guidance here. Thanks Asad