Beyond The Numbers Baron Schwartz Who Am I? baron@percona.com - - PowerPoint PPT Presentation

Beyond The Numbers

Baron Schwartz

Who Am I?

• baron@percona.com
• @xaprb
• linkedin.com/in/xaprb
• xaprb.com/blog

Who Am I?

• Maatkit
• Innotop
• Aspersa
• JavaScript Libraries
• Percona T
• olkit
• Monitoring Plugins
• Online T
• ols

• Consulting
• Support
• Remote DBA
• Engineering
• Conferences &

Training

• Percona Server
• Percona XtraBackup
• Percona XtraDB

Cluster

• Percona T
• olkit
• Many More

Today's Agenda

• Benchmarks
• Aggregation and Distributions
• Performance, Capacity & Utilization
• Rules of Thumb
• Queueing Theory and Scalability

Benchmarks

What's Missing?

• Distribution
• Time Series
• Response Times
• Parameters
• Goals
• System Specs

What's Misleading?

• Logarithmic X-Axis
• Interpolation

What's Good?

• Y-Axis Reaches 0
• No Fake-Smoothing

Behind a Single Dot

Look At All That Data...

What's With The Grid Lines?!?!?

Better Benchmarks

What does an ideal benchmark report look like?

Clear Benchmark Goals

• Validating hardware configuration
• Comparing two systems
• Checking for regressions
• Capacity planning
• Reproducing bad behavior to solve it
• Stress-testing to find bottlenecks

Hardware and Software

• Specs for CPU, disk, memory, network
• Software versions (OS, SUT, benchmark)
• Filesystem, RAID controller
• Disk queue scheduler

Presenting Results

• Ideally, make raw results available
• Include metrics from OS (CPU, RAM, IO,

network)

• Generate some plots to summarize
• This is where the rubber meets the road!

Better Aggregate Measures

• Average
• Percentiles
• 95th
• 99th
• Maximum
• Observation Duration
• Question: how bad can 95th percentile be?

More Aggregate Measures

• Median (50th Percentile)
• Standard Deviation
• Index of Dispersion

Better...

Better Still...

Keep It Coming...

Throughput AND Response Time

Performance

• What is Performance?
• T

wo Metrics

• Response Time (time per task)
• Throughput (tasks per time)
• They're not reciprocals
• More on this later

What Performance Isn't

• CPU Usage
• Load Average
• Other metrics of resource consumption

Performance

• I often focus on response time
• It represents user experience
• Throughput indicates capacity rather than

performance

• For benchmarking, throughput is primary

Utilization

• The portion of time during which the

resource is busy

• i.e. there is at least one thing in progress

Utilization is Confusing

• Be very careful with tools that report

utilization

• From the Linux iostat man page:
• “%util: Percentage of CPU time during which

I/O requests were issued to the device (bandwidth utilization for the device). Device saturation occurs when this value is close to 100%.”

• Can you parse that? Is it true?

Capacity

• What is Capacity?

Capacity

Capacity – My Definition

Capacity is the maximum throughput ... at achievable concurrency ... with acceptable performance ... as defined by response time ... meeting specified constraints ... over specified observation intervals.

Capacity Example

• What is capacity of the system at a

concurrency of 32 with 10-second 95th- percentile response time not to exceed 2ms over a 60-minute duration?

• T
• determine this, we need goal-seeking

benchmark software

• Most benchmark software can't do this

Benchmarks, etc Recap

• Most benchmarks reveal very little
• Benchmark reports reveal even less
• It's good to go beyond the surface

Amdahl's Law

• “The speedup of a program using multiple

processors in parallel computing is limited by the time needed for the sequential fraction of the program.” - Wikipedia

• It's basically a law of diminishing returns.

Should I Defragment My Disk?

• Method 1: Google “defragment”
• Method 2: Try it and see
• Method 3: Measure if the disk is a

bottleneck

Spolsky -vs- Millsap

Spolsky -vs- Millsap

Amdahl's Law

• Don't try to optimize little things.

Little's Law

• N = XR
• That is,
• Concurrency = Throughput * Response Time
• This holds regardless of queueing, arrival

rate distribution, response time distribution, etc.

Little's Law Example

• If disk IOs average 4ms...
• And there are 280 IOs per second...
• Then the disk's average concurrency is:
• N = 280 * .004
• N = 1.12
• Do you believe this?
• When might it not be true?

Little's Law Example #2

• If disk utilization is 98%
• And there are 280 IOs per second
• What do we know?

Utilization Law

• U = SX
• Also independent of distributions, etc...
• That is,
• Utilization = Service Time * Throughput
• Utilization = 98% and Throughput = 280
• S = U/X
• Service Time = .98 / 280 = .0035

Queueing Theory

• How can we predict the amount of

queueing in a system?

• How can we predict its response times?
• How can we predict capacity?

Erlang Queueing

• Erlang's formulas model the probability of

queueing for a given arrival rate, service time, and number of servers.

• A “server” is anything capable of serving

a request.

• CPUs
• Disks

CPU -vs- Disk Queueing

• Scenario: 4-CPU, 4-disk (RAID0) server
• Thought experiment:
• How do processes queue for CPU?
• How do I/O requests queue on disks?

Notation

• T

ypically see something like M/M/1

• Each letter is a placeholder in A/S/n
• A = Arrival distribution
• S = Service-time distribution
• n = Number of servers
• A and S can be one of:
• Markov
• Deterministic
• General

CPUs -vs- Disks

• CPUs: M/M/4
• Disks: 4 x {M/M/1}

M/M/1 Queueing

cmg.org

M/M/n Queueing

cmg.org

Erlang C Function

• M/M/n queueing is modeled by Erlang C
• See http://en.wikipedia.org/wiki/Erlang_(unit)

What's Wrong With Erlang C?

• You must validate your arrival times.
• You must validate your service times.
• The equation is hard to work with.
• In practice, it's hard to use Erlang C.

Scalability

• Queueing causes non-linear scaling.
• But first, let's talk about linearity.

System Scalability

Concurrency Throughput Why?

Universal Scalability Law

Concurrency Throughput Linear Amdahl USL

Amdahl Scalability

USL Scalability

USL Scalability Modeling

USL Performance Modeling

Scalability Limitations

• Locks
• Synchronization points
• Shared resources
• Duplicated data to be kept in sync
• Weakest-link problems

RAID10 On EBS

• Which is faster?
• RAID 10 over 10 EBS volumes
• RAID 10 over 20 EBS volumes
• Hint: http://goo.gl/Xm92Y
• Also, http://goo.gl/fAEIL

Debunking “Linear”

• Ask to see the actual numbers.
• They shouldn't be rounded off suspiciously.
• They must be truly linear.
• They must intersect the point (0, 0).

Debunking, Example #1

Is it Linear?

It's Not Linear

Resources

• Naomi Robbins' Blog
• http://blogs.forbes.com/naomirobbins/
• Percona White Papers
• http://www.percona.com/
• Neil J. Gunther
• Guerrilla Capacity Planning
• http://www.contextneeded.com/

Questions?

baron@percona.com @xaprb