Hi Jonathan, Thanks for the prompt response. It is helpful for sure and give me more insight than I had before - Thank you for that. It is strange though that I do see different sums once running your query with summary select type , pipe_name, sum(chunk_size) /1024/1024 chunk_size from ( select /*+ ordered use_nl(h) */ c.inst_id instance, kglnaobj pipe_name, decode(kglobt00,1,'PRIVATE','PUBLIC') type, ksmchcom chunk_com, ksmchptr chunk_ptr, ksmchsiz chunk_size, ksmchcls alloc_class from x$kglob c, x$ksmhp h where c.kglhdnsp = 7 and c.kglobsta != 0 and h.ksmchds = c.kglobhd0 and ksmchcom = 'kkxpr' ) group by type , pipe_name order by 3 ,2 ; Result TYPE ------- PIPE_NAME -------------------------------------------------------------------------------- CHUNK_SIZE ---------- PUBLIC ROPIPE .000228882 PUBLIC PUBSUBPIPENAME .00062561 PUBLIC ADVISORPIPE .000915527 PUBLIC MAKEPIPE .001235962 PUBLIC OUTBOUNDPIPE .006538391 PUBLIC PUBSUBPIPE 1.86437225 vs. the below query select * from ( select row_number () over ( partition by namespace order by sharable_mem desc ) row_within, namespace, sharable_mem/1024/1024 sharable_mem , substr(name, 1,40 ) short_name from v$db_object_cache order by sharable_mem desc ) where namespace like '%PIPE' and row_within <= 5 order by sharable_mem desc, namespace, row_within / Result ROW_WITHIN NAMESPACE ---------- ---------------------------------------------------------------- SHARABLE_MEM ------------ SHORT_NAME -------------------------------------------------------------------------------- 1 PIPE 2.44610596 PUBSUBPIPE 2 PIPE .39125061 OUTBOUNDPIPE 3 PIPE .023963928 ROPIPE 4 PIPE .008384705 ACXPROD45 5 PIPE .008369446 CUSTVEHPIPE In any this is very helpful and will check it during the different times of the day. Best, Hanan On Mar 17, 2014, at 7:36 PM, Jonathan Lewis <jonathan@xxxxxxxxxxxxxxxxxx> wrote: > > > Is this the sort of thing you'd be interested in: > > > http://jonathanlewis.wordpress.com/2009/01/30/pipes/ > > > Regards > Jonathan Lewis > http://jonathanlewis.wordpress.com > @jloracle > > ________________________________________ > From: oracle-l-bounce@xxxxxxxxxxxxx [oracle-l-bounce@xxxxxxxxxxxxx] on behalf > of Hanan Hit [hithanan@xxxxxxxxx] > Sent: 17 March 2014 21:13 > To: Oracle-L@xxxxxxxxxxxxx > Cc: Hanan Hit > Subject: Question about Oracle pipes sizes and the messages at the queue > > Hi All, > > I have a quick question about Oracle pipes sizes and the messages at the > queue. > > Basically what I am looking for is a method to identify the size of the pipe > in the the shared pool as well as to identify the number of messages (queue > size) and possibly their content. > > I am using the below query (extracted and modified from an Hotsos) > presentation about Shared pool , which lets me see the memory usage of the > different Pipes but just wanted to check whether there is other option and > still looking at the messages at the queue. > > select row_within , > namespace , > sharable_mem , > short_name > from ( select row_number () over ( partition by namespace order by > sharable_mem desc ) row_within, namespace, > sharable_mem/1024/1024 sharable_mem, > substr(name, 1,60 ) as short_name > from v$db_object_cache order by sharable_mem desc ) > where namespace like '%PIPE' > order by sharable_mem desc, namespace, row_within > / > > > > Best, > Hanan > > > -- > //www.freelists.org/webpage/oracle-l > >