Baron Schwartz: Measuring Scalability and Performance With TCP

diff --git a/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/hCalendar.html b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/hCalendar.html
new file mode 100644 (file)
index 0000000..f70a595
--- /dev/null
+++ b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/hCalendar.html
@@ -0,0 +1,16 @@
+<!-- http://microformats.org/wiki/hcalendar-cheatsheet -->
+<div class="vevent">
+ <span class="organiser">
+Baron Schwartz
+ </span>
+ <span class="summary">
+Measuring Scalability and Performance With TCP
+ </span>
+ <a class="url" href="http://www.percona.tv/percona-live/measuring-scalability-and-performance-with-tcp">Percona Live NYC 2011</a>,
+ <span class="location">New York</span>,
+ <abbr class="dtstart" title="20110526">May 26th, 2011</abbr>
+ <div class="description">
+What if you had all the data you needed to measure system performance and scalability at any tier, discover performance and stability problems before they happen, and plan for capacity by modeling the system’s behavior at greater load than you currently have? Now it is as easy as running tcpdump and processing the result with a tool. In this talk you will learn a quick overview of mathematical scaling models, how the inputs can be derived from observing query start and completion times, and how the mk-tcp-model tool from Maatkit can transform tcpdump output into the inputs you will need for tools such as Aspersa’s usl tool for applying the Universal Scalability Model.
+<p>
+ </div>
+</div>

diff --git a/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/presentation.pdf b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/presentation.pdf
new file mode 120000 (symlink)
index 0000000..2e27c9d
--- /dev/null
+++ b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/presentation.pdf
@@ -0,0 +1 @@
+PerconaLiveNYC2011-Extracting-Performance-and-Scalability-Metrics-From-TCP.pdf
\ No newline at end of file

diff --git a/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/video.ogv b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/video.ogv
new file mode 120000 (symlink)
index 0000000..057e70e
--- /dev/null
+++ b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/video.ogv
@@ -0,0 +1 @@
+6_Baron-percona-live-caraveggio-percona.mov
\ No newline at end of file

diff --git a/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/video.srt.yaml b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/video.srt.yaml
new file mode 100644 (file)
index 0000000..b86b6cc
--- /dev/null
+++ b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/video.srt.yaml
@@ -0,0 +1,222 @@
+--- 
+- '0.0'
+- '1.0'
+- "[1] Extracting Performance and Scalability Metrics From TCP"
+--- []
+
+--- 
+- '6.2'
+- '7.2'
+- "[2] Consulting Support Training Development For MySQL"
+--- 
+- '8.3'
+- '9.3'
+- "[3] Diamond Sponsors"
+--- 
+- '10.6'
+- '11.6'
+- "[4] Agenda"
+--- 
+- '71.6'
+- '72.6'
+- "[5] Capturing TCP Traffic"
+--- 
+- '230.2'
+- '231.2'
+- "[6] Process with mk-tcp-model"
+--- 
+- '329.9'
+- '330.9'
+- "[7] Plot as Time-Series and Inspect"
+--- 
+- '385.7'
+- '386.7'
+- "[8] Plot as Time-Series and Inspect"
+--- 
+- '513.2'
+- '514.2'
+- "[9] Stall Detection"
+--- 
+- '554.5'
+- '555.5'
+- "[10] If completions are what's affected, maybe we can see the evidence."
+--- 
+- '569.6'
+- '570.6'
+- "[11] Arrivals per 5ms"
+--- 
+- '650.7'
+- '651.7'
+- "[12] Completions Per 5ms"
+--- 
+- '703.8'
+- '704.8'
+- "[13] Hmmmmm"
+--- 
+- '723.5'
+- '724.5'
+- "[14] Is there really a problem?"
+--- 
+- '733.3'
+- '734.3'
+- "[15] If arrivals are fairly uniform, but completions cluster together, can we see it?"
+--- 
+- '757.1'
+- '758.1'
+- "[16] Arrivals"
+--- 
+- '769.8'
+- '770.8'
+- "[17] Completions"
+--- 
+- '780.6'
+- '781.6'
+- "[18] Hmmmmmmmmmmmmmmm"
+--- 
+- '804.0'
+- 805
+- "[19] Completions Minus Arrivals"
+--- 
+- '841.0'
+- 842
+- "[20] Seeing Hidden Patterns"
+--- 
+- '879.8'
+- '880.8'
+- "[21] Pile-Up Detection Algorithm"
+--- 
+- '953.5'
+- '954.5'
+- "[20] Seeing Hidden Patterns"
+--- 
+- '965.7'
+- '966.7'
+- "[22] Detecting Performance Problems"
+--- 
+- '1061.3'
+- '1062.3'
+- "[23] Index of Dispersion"
+--- 
+- '1076.5'
+- '1077.5'
+- "[24] Index of Dispersion"
+--- 
+- '1120.0'
+- 1121
+- "[25] Interpreting Index of Dispersion"
+--- 
+- '1126.5'
+- '1127.5'
+- "[26] Highly variable == highly optimizable."
+--- 
+- '1232.4'
+- '1233.4'
+- "[27] Time-Series, All Together Now"
+--- 
+- '1276.7'
+- '1277.7'
+- questions
+--- 
+- '1412.8'
+- '1413.8'
+- "[28] Modeling Scalability"
+--- 
+- '1448.2'
+- '1449.2'
+- "[29] The Scalability Function"
+--- 
+- '1475.5'
+- '1476.5'
+- "[30] This is Linear Scalability"
+--- 
+- '1541.0'
+- 1542
+- "[31] This is Not Linear Scalability"
+--- 
+- '1543.5'
+- '1544.5'
+- "[32] What Causes Non-Linearity?"
+--- 
+- '1570.3'
+- '1571.3'
+- "[33] Factor #1: Serialization"
+--- 
+- '1681.2'
+- '1682.2'
+- "[34] Factor #2: Coherency"
+--- 
+- '1753.6'
+- '1754.6'
+- "[35] Loss of Scalability"
+--- 
+- '1794.3'
+- '1795.3'
+- "[36] Scalability Modeling Method"
+--- 
+- '1825.3'
+- '1826.3'
+- "[37] Concurrency from TCP"
+--- 
+- '1841.1'
+- '1842.1'
+- "[38] Concurrency from TCP"
+--- 
+- '1862.2'
+- '1863.2'
+- "[39] Visually..."
+--- 
+- '1922.7'
+- '1923.7'
+- "[40] Using mk-tcp-model"
+--- 
+- '1942.0'
+- 1943
+- "[41] Using Aspersa's USL Tool"
+--- 
+- '1958.0'
+- 1959
+- "[42] Results"
+--- 
+- '2096.8'
+- '2097.8'
+- "[43] Forecasting Performance"
+--- 
+- '2221.1'
+- '2222.1'
+- "[44] Forecasting Performance"
+--- 
+- '2266.4'
+- '2267.4'
+- "[45] Validating Input"
+--- 
+- '2465.3'
+- '2466.3'
+- "[46] When Is This Useful?"
+--- 
+- '2474.8'
+- '2475.8'
+- "[47] Be Vewwy Vewwy Quiet"
+--- 
+- '2483.6'
+- '2484.6'
+- "[48] Worst-Case Scaling"
+--- 
+- '2519.7'
+- '2520.7'
+- "[49] Double Rainbow!"
+--- 
+- '2614.4'
+- '2615.4'
+- "[50] \xE2\x80\x9CAssuming Linear Scalability...\xE2\x80\x9D"
+--- 
+- '2623.2'
+- '2624.2'
+- "[51] Best-Case Bounds"
+--- 
+- '2668.0'
+- 2669
+- "[52] Summary"
+--- 
+- '2724.0'
+- '2725'
+- "[53] Further Research"

diff --git a/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/www.sh b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/www.sh
new file mode 100755 (executable)
index 0000000..806d2c6
--- /dev/null
+++ b/media/pl-nyc2011-Measuring_Scalability_and_Performance_With_TCP/www.sh
@@ -0,0 +1,4 @@
+#!/bin/sh
+
+#get_flash_videos http://blip.tv/play/hKlBgsOWcQA.html
+wget -nc http://www.percona.com/files/presentations/percona-live/nyc-2011/PerconaLiveNYC2011-Extracting-Performance-and-Scalability-Metrics-From-TCP.pdf