Quanah Gibson-Mount <[EMAIL PROTECTED]> writes: > That's the guideline for optimal performance. But it's also rather > expensive memory wise. I set Stanford's servers to duplicate values, > and they've done just fine. ;) Just as the 50k cache size Stanford sets > is far below the optimal value for performance as well. ;)
Okay. Does that parameter (idlcachesize) exist for BDB as well, or do we have to add it conditionally only if HDB is selected? -- Russ Allbery ([EMAIL PROTECTED]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
