Posted by Riyaj Shamsudeen on March 24, 2016
Over the last many years, some of you have invited me to attend conferences in India, and talk about Oracle RAC and performance. I have not had an opportunity to make it to conferences in India, until now
I am excited to announce that I will be participating in OTN sponsored Oracle ACE Director’s tour in India (April 23rd to May 2nd 2016), and presenting ( with deep dive demos ) about RAC, performance, and in-memory. This is a golden opportunity for you to learn some of the internal stuff that I talk about in my class too.
Refer http://otnyathra.com for further details.
I am also excited to be in the tour with Connor McDonald, Biju Thomas, Sai, and many others. Don’t forget to send an email to Sai, the conference rooms will fill out quickly.
Did I mention deep dive demos?
Update 1: Updating the presentation slides and scripts. Do ask if I miss any script that you are looking for
Posted in 12c, Performance tuning, Presentations, RAC | Tagged: oracle performance, RAC performance | 4 Comments »
Posted by Riyaj Shamsudeen on November 18, 2015
Thanks for coming to my presentations in RAC day at Dublin, Ohio. Please find the presentation files below. Hopefully, I will get video files and upload that here too.
OOUG presentation files and scripts
md5 checksum of the zip file is:
Posted in Oracle database internals, Performance tuning, Presentations, RAC | Tagged: oracle performance, RAC, RAC performance | 3 Comments »
Posted by Riyaj Shamsudeen on October 22, 2015
I will be talking about RAC and performance in-depth, with lots of demos, in a RAC day training with Ohio Oracle User group on Nov 16,2015 Monday. Venue for the presentation is Dublin, Ohio.
Agenda for the day:
08:00a – 09:00: Registration / Breakfast
09:00a – 09:15: Announcements -Introduction of the speaker
09:15a – 10:30: Underpinning for Oracle RAC and Clusterware
10:30a – 10:45: Break
10:45a – 11:45: RAC cache fusion internals
11:45a – 01:00: Lunch
01:00p – 02:00: RAC Performance tuning Part 1 – Wait events and object tuning
02:00p – 02:15: Break
02:15p – 03:30: RAC performance tuning Part 2 – locks, library cache locks etc.
03:30p – 03:45: Member Announcements, Gift Drawings
Please RSVP to the co-ordinators so that you will have a seat
OOUG RAC day
Posted in Oracle database internals, Performance tuning, Presentations, RAC | Tagged: oracle performance, RAC, RAC performance, RAC performance myths | 1 Comment »
Posted by Riyaj Shamsudeen on April 15, 2015
I am an ardent believer of “show me how it works” principle and usually, I have demos in my presentation. So, I was presenting “Tools for advanced debugging in Solaris and Linux” with demos in IOUG Collaborate 2015 in Las Vegas on April 13 and my souped-up laptop (with 32G of memory, SSD drives, and an high end video processor etc ) was not responding when I tried to access folder to open my presentation files.
Sometimes, demos do fail. At least, I managed to complete the demos with zero slides Apologies to the audience for my R-rated rants about laptop issues.
You can download presentations files from the links below.
Posted in in-memory, inmemory, Performance tuning, Presentations | Tagged: dtrace, oracle performance, perf record, perf tool, pstack, truss | Leave a Comment »
Posted by Riyaj Shamsudeen on March 25, 2015
I will be presenting two topics in IOUG Collaborate 2015 in Vegas. Use the show planner and add my presentations to your schedule
Session #189: April 13 Monday 9:15 to 10:15AM Topic: Oracle Database 12c In-Memory Internals. Room Palm B
Session #145: April 13 Monday 12:45PM-1:45PM Topic: Tools and Techniques for Advanced Debugging in Solaris & Linux (mostly live demo). Room Palm B.
Posted in inmemory, Oracle database internals, Performance tuning, Presentations, RAC | Tagged: collaborate 2015, in-memory internals, ioug, presentations, strace, truss | Leave a Comment »
Posted by Riyaj Shamsudeen on January 19, 2015
I will be talking in Rocky Mountain Oracle User Group Training Days 2015( http://www.rmoug.org), with live demos (hopefully there will be no failures in the demo). My topics are:
Feb 17: Deep dive: 3:15PM to 5:15PM – RAC 12c optimization: I will discuss RAC global cache layer in detail with a few demos. You probably can’t find these deep Global Cache layer details anywhere else
Feb 19: Wednesday: 2:45PM to 3:45PM – Advanced UNIX tools: I will discuss both Solaris and Linux advanced tools to debug deep performance issues.
Feb 19: Wednesday: 12:15PM – 1:15PM – Exadata SIG panel with Alex Fatkulin.
Come to Denver. Come on, it won’t be cold ( I think )
Uploading presentation and scripts of the presentation files: RMOUG_2015_Riyaj_RAC_12c_optim files
Presentation files for Unix tools demos: RMOUG_2015_Riyaj_Unix_tools
Posted in 11g, 12c, Presentations, RAC | Tagged: gc buffer busy, RAC performance, RAC training, rmoug | 1 Comment »
Posted by Riyaj Shamsudeen on October 6, 2014
While presenting at Oaktable World 2014 in San Fransisco, I discussed the in-memory pre-population speed and indicated that it takes about 30 minutes to 1 hour to load ~300GB of tables. Someone asked me “Why?” and that was a fair question. So, I profiled the in-memory pre-population at startup.
I profiled all in-memory worker sessions using Tanel’s snapper script and also profiled the processes in OS using Linux perf tool with 99Hz sample rate. As there is no other activity in the database server, it is okay to sample everything in the server. Snapper output will indicate where the time is spent; if the time is spent executing in CPU, then the perf report output will tell us the function call stack executing at that CPU cycle. Data from these two profiling methods will help us to understand the root cause of slowness.
- @snapper.sql out,gather=stw 600 10 “select sid from v$session where program like ‘%W00%'”
- Perf tool : perf record -F 99 -u oracle -g sleep 3600
Posted in 12c, in-memory, inmemory, Oracle database internals, Performance tuning, Presentations | Tagged: 12c in-memory, in-memory, kdzu, perf record, perf report, pre-population cpu time, pre-population speed, snapper | Leave a Comment »
Posted by Riyaj Shamsudeen on September 26, 2014
Many Oaktable members are planning to talk about deep technical topics in Oaktable world 2014. Looking at the agenda, I am excited, so many deep topics are planned. I will be talking about in-memory internals on Monday morning at 9AM, 9/29/2014, right after Mogens’ Keynote speech. You can find all details here: Oaktable world 2014. I will post my presentation slides after the presentation.
Start your open world week presentation with mine. Sorry, no beers planned at that time, it is 9AM, after all!
Thanks for attending my presentation at Oaktable World 2014. You can download the slides : In-memory_internals.pdf.
Also, our book Expert Oracle RAC 12c has been translated to Chinese language. You can find details about that book in one of the translator’s blog: Alex lizx.
Posted in 12c, in-memory, inmemory | Tagged: in-memory, oaktable world 2014, presentations | Leave a Comment »
Posted by Riyaj Shamsudeen on September 11, 2014
I enabled an huge 70G table for inmemory population, I expected the inmemory population to take a while, but the population didn’t complete even after letting it run for a day. Why?
Initial review of the server shows no issues, no resource starvation. This must be a problem with Oracle processes itself. I started digging further, and ASH data shows that in numerous samples the process was seen reading block using single block I/I calls. Also object_id matches with the table I was trying to populate.
select * from (
select start_time, end_time, sql_id,event, current_obj#, cnt_on_cpu + cnt_waiting tot_cnt,
rank () over ( order by (cnt_on_cpu + cnt_waiting) desc ) rnk
sum(decode(session_state,'ON CPU',1,0)) cnt_on_cpu,
first_value(sample_time) over( order by sample_time ) start_time,
last_value(sample_time) over( order by sample_time
rows between unbounded preceding and unbounded following ) end_time,
sql_id,event, session_state, current_obj#
(select * from v$active_session_history ash where session_id= &&sid and session_serial#=&&serial_number)
group by sql_id, event, current_obj#
START_TIME END_TIME SQL_ID EVENT CURRENT_OBJ# TOT_CNT RNK
------------------------- ------------------------- ------------- ------------------------------ ------------ ---------- ----------
18-AUG-14 08.42.03.702 AM 18-AUG-14 09.02.06.463 AM db file sequential read 168967 990 1
168967 156 2
direct path read 168967 50 3
bdwtqttka2w2y -1 3 4
bdwtqttka2w2y direct path read 168967 1 5
24uqc4aqrhdrs 168967 1 5
-1 1 5
Read the rest of this entry »
Posted in 12c, inmemory | Tagged: 12c inmemory, inmemory internals | 3 Comments »
Posted by Riyaj Shamsudeen on September 11, 2014
After the restart of a 12c inmemory database with 300GB+ SGA, I noticed that an Oracle background process sa00 was consuming a bit of CPU. Documentation suggests that it is SGA Allocator process, however, ipcs -ma command shows that the shared memory segment is already allocated. I was curious, of course, what would that background process will be allocating?.
Process stack of the process shows that it is touching SGA pages to pre-page SGA memory pages.
$ pstack 21131
#0 0x0000000000d9996e in ksmprepage_memory ()
#1 0x0000000000d99369 in ksm_prepage_sga_seg ()
#2 0x0000000003a5c78b in skgmapply ()
#3 0x0000000000da686a in ksmapply_v2 ()
#4 0x0000000000d9a82c in ksmprepage ()
#5 0x0000000000d99f89 in ksm_sslv_exec_cbk ()
#6 0x0000000000f79810 in ksvrdp ()
#7 0x00000000031013b7 in opirip ()
#8 0x0000000001bb0a08 in opidrv ()
#9 0x00000000026c0f71 in sou2o ()
#10 0x0000000000bbd85e in opimai_real ()
#11 0x00000000026cb6bc in ssthrdmain ()
#12 0x0000000000bbd72c in main ()
$ ps -ef|grep 21131
oracle 21131 1 96 15:00 ? 00:01:50 ora_sa00_XXXXXX
Two notable changes in this area:
1. Incidentally, pre_page_sga initialization parameter was defaulted to a value of FALSE until 11.2. In version 12.1, the parameter value defaults to TRUE.
2. As huge SGA is expected for inmemory databases, a new background process SA00 is also created to touch all SGA memory pages at startup.
As inmemory worker processes will be populating the inmemory column store soon after the startup, touching memory pages at instance startup makes sense, and the feature should improve the performance of inmemory population. At least, Worker processes doesn’t need to suffer from huge amount page faults. (note that this SGA is not using hugepages).
Change to the parameter pre_page_sga also should improve the performance of inmemory scan, as the memory map entries will be setup at the process startup. However, I am not quite clear, how this change will affect the performance of a connection storm, i.e. if there are numerous database connections in a short period of time and disconnects. Isn’t that the reason why the pre_page_sga was defaulted to FALSE? But, I need to test this thoroughly to understand the implications further.
Posted in 12c | Tagged: 12c inmemory, inmemory, inmemory internals, pre_page_sga | 4 Comments »