Write consistency and DML restart

Few weeks ago Tanel Poder published a great video talking about Oracle’s write consistency , DML restarts and demonstrating also how we can detect them.He also published a script that allow finding UPDATE/DELETE statements that hit the write consistency issue & have to restart under the hood. But as he stated it uses V$SQL_PLAN_MONITOR, so requires Diag+Tuning pack licenses.

The purpose of this blog post is to show another way to detect statements hitting the write consistency issue.

Continue reading

From memory request to PL/SQL source line [ errorstack dump ]

In my last blog post i described a geeky way to trace back the responsible PL/SQL code for a particular memory request into the PGA. It was based essentially on a dynamic tracing tool “systemtap” to probe on  specific functions entry such as KGHAL memory allocator functions (Based on Stefan Koehler dtrace script )  and relied on calling an internal oracle function “pfrln0lookup” using “oradebug call” to  get the actual PL/SQL line number.It would have been safer if we reverse engineered the function “pfrln0lookup” to extract only the thing that matter and avoid calling it but this need time and is forbidden :p  !

So here i will describe a safer and simpler approach relying only on collecting multiple errorstack dump samples and a little shell script to parse the trace file !

Continue reading

Memory bandwidth vs latency response curve

Memory bound applications are sensitive to memory latency and bandwidth that’s why it’s important to measure and monitor them.Even if this two concepts are often described  independently they are inherently interrelated.

According to Bruce Jacob in ” The memory system: you can’t avoid it, you can’t ignore it, you can’t fake it” the bandwidth vs latency response curve for a system has three regions.

  • Constant region: The latency response is fairly constant for the first 40% of the sustained bandwidth.
  • Linear region:  In between 40% to 80% of the sustained bandwidth, the latency response increases almost linearly with the bandwidth demand of the system due to contention overhead by numerous memory requests.
  • Exponential region:  Between 80% to 100% of the sustained bandwidth,  the memory latency is dominated by the contention latency which can be as much as twice the idle latency or more.
  • Maximum sustained bandwidth :  Is 65% to 75% of the theoretical maximum bandwidth.
 Armed with Intel Memory Latency Checker (MLC) let’s check our current system !

Continue reading

Geeky PL/SQL tracer/profiler : Another step

This is my second post under the theme of how to extend our capabilities to trace and profile PL/SQL code.This time motivated by a comment from Luca Canali on my previous post  :

So based on my previous work on geeky PL/SQL tracer let’s see how we can obtain a geeky PL/SQL on-CPU Flame Graph !

Continue reading

Geeky PL/SQL tracer/profiler : First step

This blog post is about how to extend our capabilities to trace and profile PL/SQL code.It’s primarily motivated by few tweets from Franck Pachot and of course because it’s FUN !

Capture 02

Capture 01

So in the first part of this series we are going to answer to this questions : Can we map those underling function to the source PL/SQL object and line number ? Can we obtain a full trace ? Of course yes otherwise there will be no blog post :p

Continue reading