Re: About net-snmp table iterator performance
bai haoquan <[email protected]>
| Newsgroups | gmane.network.net-snmp.user |
|---|---|
| Message-ID | <[email protected]> |
Hi, Thank you for your reply. Now I had known that the iterator helper is inefficient for large tables and I should try other helpers for my apps. And if the data requested is internal my agent, which kind of helper should I choose instead of the iterator helper. Looking forward to your reply. Best regards On Mon, Mar 15, 2010 at 7:10 PM, Dave Shield <[email protected]>wrote: > On 15 March 2010 10:29, bai haoquan <[email protected]> wrote: > > To my understanding, it will cause performance problem when there are > mass > > of data. > > Yes - the iterator helper is notoriously inefficient for large tables. > It's probably not a sensible choice if performance is an issue. > Use one of the other helpers instead. > > > I have no idea that why we have to traverse > > the whole 200 nodes every time to get a value > > Remember that the iterator helper was specifically designed > for tables where the data is stored in "random" order. > So the expectation is that you'd always have to search the > whole internal table, in order to find the "next row" in SNMP ordering. > > There is a flag that can improve performance if the internal > indexing matches the SNMP ordering, but it's still probably > not the most efficient of the helpers. > > > Remember too that the "get_first/get_next" loop is run > for each individual GETNEXT request. > The "get_next" calls do *NOT* correspond one-to-one > with the incoming GETNEXT requests of the snmpwalk > or snmptable command. > > > > Do I have any misunderstanding? Look forward to your early reply and > > some good advices for this problem. > > My best advice is to use a different helper! > > Dave > ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Net-snmp-users mailing list [email protected] Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users