So, as Andy pointed, transients would do the trick for you. And maybe type hints could also help but, as Stuart said, benchmark it all the way down.
Regards Plinio On Wednesday, April 20, 2016, JvJ <kfjwhee...@gmail.com> wrote: > I don't think I'll go with primitive arrays. A big part of the reason I'm > using Clojure is for the immutable persistence. I just want to use them in > the most efficient way possible. I know I'm not going to get hyper-blazing > C-level performance. > > On Wednesday, 20 April 2016 15:38:11 UTC-7, Stuart Sierra wrote: >> >> The first answer: test & measure. Benchmark your code, use a JVM profiler >> to find hotspots, etc. Test every change you make to see if it has a >> measurable improvement. Any assumptions about what “should” be >> faster/slower are likely to be wrong. >> >> The long answer: >> >> The JVM does not give you much control over how objects are arranged in >> memory. In Java and Clojure, almost everything is a pointer to an object on >> the heap. Java collection classes and Clojure collections store pointers to >> objects; they do not store values “in-line” like an array of structs in C. >> The JVM *may* have optimizations that try to arrange objects “near” >> other objects, but you have no control over this. >> >> So my (untested) expectation is that all Clojure collection types are >> more-or-less equal in terms of memory locality. >> >> The only built-in data structure that offers the possibility of >> contiguous allocation in Java — without dropping down to native code — is >> an array of primitives, such as integers or doubles. Clojure has built-in >> functions to create and manipulate Java primitive arrays, if that works for >> your use case. >> >> –S >> >> >> On Wednesday, April 20, 2016 at 2:03:10 PM UTC-4, JvJ wrote: >>> >>> I'm writing some code that I would like to perform as quickly as >>> possible. Currently, I am iterating over large hash maps and performing >>> assocs and dissocs. >>> >>> I don't know much about performance optimization, but I am told that >>> memory locality is a big factor. I would like to know how Persistent Maps, >>> Persistent Vectors, Transient Maps, and Transient Vectors compare to one >>> another in this respect. >>> >>> Also, the objects in the collection that I'm iterating over will >>> themselves be maps. So, if I had a vector with good memory locality, but >>> it stored what are effectively pointers to maps allocated elsewhere, will >>> that nullify the benefits of memory locality? >>> >>> Thanks >>> >> -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > <javascript:_e(%7B%7D,'cvml','clojure@googlegroups.com');> > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > <javascript:_e(%7B%7D,'cvml','clojure%2bunsubscr...@googlegroups.com');> > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > --- > You received this message because you are subscribed to the Google Groups > "Clojure" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to clojure+unsubscr...@googlegroups.com > <javascript:_e(%7B%7D,'cvml','clojure%2bunsubscr...@googlegroups.com');>. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
