Update of /cvsroot/monetdb/MonetDB4/src/modules/plain
In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv21408/src/modules/plain
Modified Files:
Tag: MonetDB_4-20
xtables.mx
Log Message:
made a more advanced version of the sample trick. Was needed as the old
solution could over estimate (leading to failed hash table creation)
Index: xtables.mx
===================================================================
RCS file: /cvsroot/monetdb/MonetDB4/src/modules/plain/xtables.mx,v
retrieving revision 1.4.2.1
retrieving revision 1.4.2.2
diff -u -d -r1.4.2.1 -r1.4.2.2
--- xtables.mx 22 Oct 2007 19:55:13 -0000 1.4.2.1
+++ xtables.mx 23 Oct 2007 17:29:12 -0000 1.4.2.2
@@ -703,22 +703,35 @@
@c
#define SAMPLE_SIZE 1024
hash_t
-derive_mask( BAT *ct_histo, BAT *b)
+derive_mask( BAT *ct_map, BAT *ct_histo, BAT *b)
{
+ size_t cnt = BATcount(b);
int n = bits(BATcount(ct_histo)), *N = &n;
if (BATcount(b) > (SAMPLE_SIZE<<3)) {
BAT *s = BATsample(b, SAMPLE_SIZE);
- BAT *u = BATkunique(BATmirror(s));
- size_t uc = BATcount(u);
+ grp d,o;
+ BUN p, q;
+ BAT *histo;
+ hash_t r = 0;
+ o.map = ct_map->batCacheid;
+ o.histo = ct_histo->batCacheid;
+ if (CTderive(&d, &o, s) != GDK_SUCCEED) {
+ BBPunfix(s->batCacheid);
+ return GDK_FAIL;
+ }
BBPunfix(s->batCacheid);
- BBPunfix(u->batCacheid);
- if (uc > 0) {
- size_t ratio = (((dbl)BATcount(b)/SAMPLE_SIZE) * uc);
- if (ratio > 0)
- return (1<<bits(*N * ratio)) - 1;
+ assert (d.histo);
+ histo = BATdescriptor(d.histo);
+ BATloop(histo, p, q) {
+ /* + 4 for a average chain list of 4 */
+ r += ((dbl)cnt/SAMPLE_SIZE)/(*(int*)BUNtloc(histo,p)+4);
}
+ BBPunfix(d.histo);
+ grp_unfix(&d);
+ if (bits(r) > *N)
+ return (1<<bits(r))-1;
}
/* default to */
return (1<<*N) - 1;
@@ -735,7 +748,7 @@
mapentry_t entry, *e;
BAT *map;
[EMAIL PROTECTED]
- hash_t mask = derive_mask(ct_histo, b);
+ hash_t mask = derive_mask(ct_map, ct_histo, b);
int custom_rng = BATcount(ct_histo); /* expected number of groups */
hash_t custom_MASK = mask;
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Monetdb-checkins mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-checkins
