Hi,
I have to populate a triple store with a big number of data (~38k records x
12) and there is a deadly narrow bottleneck - IO operations speed. To fix
it I did:
1. To avoid threads overflow I put all compute results into channel.
2. Loading data in chunks is better than single transaction for single
record
I tried to do by creating channel with poputale-all traversal but it seems
doesn't work properly. In the following mock example it works when the
chunk size is equal the data vector (i.e. 6): "value: [of made is fruit
soup Berry]" - for now I do not care the order.
(let [q (a/chan 500 (partition-all 6))
in ["Berry" "soup" "is" "made" "of" "fruit"]]
(a/go-loop [j (a/<! q)]
(when j
(println "value: " j)
(recur (a/<! q))))
(doseq [itm in]
(a/go (a/>! q itm))))
I cannot see any problem. How can I solve it? In the following example
chunk size should be max 6? I expected partition-all will work the same way
as itself:
(partition-all 5 ["Berry" "soup" "is" "made" "of" "fruit"]) ==>
(("Berry" "soup" "is" "made" "of") ("fruit"))
Thanks a lot,
Jacek
--
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.
