News for hackers. Stuff that matters.
Re:Speeding up SQL queries by orders of magnitude using UNION by slt2021 on Saturday March 20, 2021 @10:15PM (#26527624)
all queries with JOINs can be rewritten as cartesians: select * from a join B on col1=col2 will be rewritten by query optimizer into:

select * from A, B /* this is cartesian that slows down your query */ WHERE a.col1=B.col2

in fact both queries produce the same execution plan if you check yourself

Re:Speeding up SQL queries by orders of magnitude using UNION by gfody on Saturday March 20, 2021 @10:45PM (#26527730)
I think the point is rather that a join with a condition (which is the norm) is almost never actually executed as a cartesian product. take for example from tfa

  select * from customer_order_items
the grain is one row per customer order item, and the next two joins don't change that

  select * from customer_order_items join customer_orders on order_id
the grain is still one row per customer order item

  select * from customer_order_items join customer_orders on order_id join customers on customer_id
the grain is still one row per customer order item

..etc..

of course later on they screw it up and take the cartesian product of customer_order_items by employ_markouts, but it's just 2 big factors not 7 - their query did finish after a few seconds. usually mistakes involving cartesian products with 7 factors just run for hours and eventually throw out of memory.

Re:Speeding up SQL queries by orders of magnitude using UNION by ineedasername on Saturday March 20, 2021 @11:15PM (#26527844)
All queries could technically be expressed as a cartesian product but that is not necessary and not what happens in practice. Both of the above might produce the same plan because they are both treated as joins. One is expressed as an explicit join, the other as an implicit join, but neither requires the query engine to produce a cartesian product on execution. If it did, queries I run in seconds would require more storage or RAM than probably exists in my entire workplace, and I'm not using anything that would usually be considered "big data".
Re:Speeding up SQL queries by orders of magnitude using UNION by slt2021 on Saturday March 20, 2021 @11:25PM (#26527889)
>>queries I run in seconds would require more storage or RAM than probably exists in my city

my explanation to this: CPUs ave very fast. My CPU is 4Ghz so a single core can do a lot of computations in one second, and a SQL engine is smart enough to make cartesian computation (as part of a query plan) and discard the result if row does not meet predicate condition.

in fact I agree that not entire cartesian is being computed, if you specify enough predicates. But the query still multiplies rows. In the author's article he is joining employees when customer_id columns in NULL so this is technically a cartesian, because NULL predicate is not very selective (=there are a lot of rows with value of NULL)

Re:Speeding up SQL queries by orders of magnitude using UNION by ineedasername on Sunday March 21, 2021 @10:20AM (#26530654)
You are not using the term cartesian product correctly as applied to how database engines execute queries. Multiple people have detailed the problems with your ideas on how such things work. A fast processor cannot overcome the the need for multiple terabytes of RAM that would be required to process common queries if databases worked as you describe. Databases are significantly more likely to get bottlenecked by RAM than CPU, and your incorrect understanding of how databases work would require as much RAM as a large "big data" cluster to run queries on moderately sized data sets. Then even if it had that RAM, the bandwidth of the system's bus channels would be incapable of transporting it fast enough to return results in 500ms. Certainly not on my system with a few dozen other people working at the same time, and especially not when I have to query the live production system with a few thousand end users instead of the data warehouse. Databases do not work this way.
Re:Speeding up SQL queries by orders of magnitude using UNION by slt2021 on Sunday March 21, 2021 @12:59PM (#26531844)
multiple people can be wrong, so it is not a valid argument. I can write 10 nested loops and overwrite value of a single cpu register with a single value, so many times, that it will exceed any big data cluster capacity. This is what I meant by CPUs are fast.

if you study engine internals, especially MergeJoin, HashJoin, NestedLoopJoin - they all do comoute cartesian while simultaneously applying predicates. Some operations are faster because of sorting and hashing, but they still do cartesian.

Re:Speeding up SQL queries by orders of magnitude using UNION by ineedasername on Sunday March 21, 2021 @07:03PM (#26534968)
If they are applying predicates to reduce the number of rows processed then they are not performing cartesian joins. You don't need to take the word of people here, you merely need to read any source about how database engines process queries. I am of course open to the possibility of being wrong should you find authoritative sources that show cartesian joins produced for all queries. However, nearly two decades of working with a variety of engines tells me you are unlikely to find such a source. Your comments betray a fundamental lack of understanding on the topic. Your unwillingness to recognize the exponential resources required for your understanding to be correct also betrays a fundamental misunderstanding. You even contradict yourself by in one comment insisting that such queries would be "SLOW" while in another comment stating that you obtain speedy results due to a fast CPU. Which in itself betrays yet another fundamental misunderstanding in how databases utilize resources and where bottlenecks arise.

Last, you have offered no significant support for your claims, and as the initiator of the discussion making the claim of cartesian joins, the burden of proof is on you to provide evidence for your claims.

In any case, your persistence in digging your hole deeper and deeper on this issue is embarrassing, and I won't be cruel enough to enable that further. Feel free to reply, but I am done, with a final recommendation that you continue your learning and not stop at the point of your current understanding. You at least seem earnest in your interests in this topic, so you should pursue it further.

Re:Speeding up SQL queries by orders of magnitude using UNION by gfody on Monday March 22, 2021 @06:15PM (#26548061)
> if you study engine internals, especially MergeJoin, HashJoin, NestedLoopJoin

it's called merge join because it employs the merge algorithm (linear time constant space)[1] - this can only be used when the inputs are sorted. likewise the hash join entails building a hash table and using it (linear time linear space)[2] but the inputs don't have to be sorted. the point of these is to avoid the O(N*M) nested loop

[1] https://en.wikipedia.org/wiki/Merge_algorithm

[2] https://en.wikipedia.org/wiki/Hash_join

Re:Speeding up SQL queries by orders of magnitude using UNION by zeroimpl on Sunday March 21, 2021 @11:43AM (#26531255)
Umm.. virtually no join on tables with more than a handful of rows is done as a cartesian product. A suitable set of columns is used to sort or hash each side of the join, such that the actual join logic is basically linear performance (each row in table A is probably only ever paired with approximately 1 row in table B when evaluating the predicates). A cartesian product would involve testing the predicates on every combination of A and B (each row in table A being paired with each row in table B)

(Note when I say linear performance, I mean the join predicate is executed a linear amount of times, but the initial sort/hash operation was probably more like O(n log n))

Re:Speeding up SQL queries by orders of magnitude using UNION by whimsicalism on Sunday March 21, 2021 @02:03PM (#26532438)
Cartesian product means a very specific thing - testing the predicates on every possible combination. This is not what occurs in practice
Re:Speeding up SQL queries by orders of magnitude using UNION by whimsicalism on Saturday March 20, 2021 @11:09PM (#26527822)
? not how it works behind the scenes at all