When collecting network traces to solve performance problems, we need to balance the need to capture enough data to re-construct the upper protocols with the need to limit the amount of trace data collected overall. When capturing NFS traffic, the packet size can grow to be quite large, since the NFS packets will...
Read more »
Performance
Collecting NFS trace data for performance tasks.
Presentation to RTP OpenSolaris group on Filebench.
On Janurary 5th 2010 I gave a short presentation on the FileBench workload modeling tool to the RTP OpenSolaris group. Click on the slide to advance. The “Slides” link will take you to an HTML version.
Slides
Read more »
Performance related videos from Velocity 09
A set of recorded presentations from the Velocity 09 conference, many of which are related to performance. Particularly “Internet Scale” problems.
Here’s the link: Velocity 2009
Read more »
Benchmarking Blunders and Things That Go Bump in the Night
A 2004 paper from Neil Gunther which essentially discusses the pitfalls of relying on observed date without applying the known performance laws (most notably little’s law) to the problem at hand.
PDF
The original link to arXiv.org (whatever that is)
Read more »
Sorting TPC-H lineitem table.
Sometimes it is useful to sort the lineitem table, in the order that Query-6 will read the data back.
Query-6 looks like so :-
set timing on;
select
sum(l_extendedprice * l_discount) as revenue
from
lineitem
where
l_shipdate >= to_date( '1995-01-01', 'YYYY-MM-DD')
and l_shipdate < add_months(to_date( '1995-01-01', 'YYYY-MM-DD'), 12)
and l_discount between 0.07 - 0.01 and 0.07 + 0.01
and l_quantity < 25;
exit;
and...
Read more »