Guerrilla Capacity Planning
Guerrilla Capacity Planning
Purpose
This 5-day course presents guerrilla-style tactics for
doing performance analysis, system sizing, and capacity planning in
todayÕs challenging business environment where time-to-market is
everything, and management is impatient for results. In such an
environment, performance analysis and planning are often dropped in
favor of just getting the project completed on schedule. Building a
detailed simulation model for the project might be a cool thing to do,
but it could also be career-limiting in an environment where management
just wants a sense of direction and not the compass-bearing!
Conventional performance analysis and capacity planning techniques are
often too cumbersome and time consuming to address such immediate needs.
Guerrilla Capacity Planning presents a more tactical approach to
meet today's demanding capacity challenges.
A brief overview of Guerrilla Capacity Planning concepts and
techniques was published in
IEEE IT Professional
magazine (PDF 1.3 MB).
Certification
This class corresponds to Guerrilla Capacity Planner: Level II certification.
The levels are defined as:
- Entry level. No training class for this grade exists yet.
- Exposure to a wide variety of computer systems capacity planning concepts, methods, and
tools that can be adapted opportunistically to support the needs of
enterprise-level platform-independent performance management.
- Detailed study of a particular capacity planning technique or performance analysis tool.
An example class is Guerrilla Data Analysis.
Another Level III class, based on using PDQ, is being planned.
A printed certificate reflecting the level of achievement is awarded to each attendee who completes the course.
Dates
Check the
schedule
for the latest information.
Online
registration
is available. Additional registration details are provided at the end of this page.
5-Day Course Outline
Class begin each day at 9am and end at 5pm.
A serviced morning break of half an hour around 10:30am.
Seated lunch service is provided from Noon until 1pm.
A serviced afternoon break of half an hour around 3:00pm.
On the last day, we usually finish around 3:00pm for those needing to head for the airport.
Contents
1 Motivations and Terminology
1.1 What is Guerrilla Capacity Planning?
1.2 The Guerrilla Manual
1.3 ITIL (IT Infrastructure Library) for Guerrillas
1.4 Getting the Jump on Queueing
2 Capacity Planning Techniques
2.1 Queueing Systems for Computers Systems
2.2 Multi-Tier Planning with PDQ
2.3 A Slow Introduction to Pretty Damn Quick (PDQ)
3 Quantitative Scalability on a Stick
3.1 Universal Law of Computational Scaling
3.2 Hardware Scalability
3.3 Software Scalability
3.4 Virtual Load Testing
4 Virtualization and Standards
4.1 Spectrum of Virtualization: From HTT to GRIDs
4.2 The Fair-Share Scheduler
4.3 Industry Standards and Benchmarks
5 Distributed Systems Planning
5.1 Website Planning
5.2 Internet Planning
5.3 Gargantuan Computing for Guerrillas: GRIDs and P2P
1 Motivations and Terminology
1.1 What is Guerrilla Capacity Planning?
- Why management resists capacity planning
- Risk Management vs. Risk Perception
- Tactical Planning as a Weapon
1.2 The Guerrilla Manual
1.3 ITIL (IT Infrastructure Library) for Guerrillas
- ITIL history and source material
- ITIL business process infrastructure
- ITIL capacity management process
- How ITIL connects with Guerrilla capacity planning
- The Wheel of Capacity Planning for ITIL
1.4 Getting the Jump on Queueing
- Capacity Modeling is Not Like a Model Railway
- Grocery store as a queue: Checking it out
- Fundamental Metric Relationships
- LittleÕs Law Means a Lot
- Comparative performance of various queues
2 Capacity Planning Techniques
2.1 Queueing Systems for Computers Systems
- Open-Circuit Queues
- Closed-Circuit Queues
- Multiple Workloads in Open and Closed Circuits
- Performance Bounds and Log Jams
2.2 Multi-Tier Planning with PDQ
- Client/Server Architectures
- Benchmarking Environment
- Capacity Analysis with PDQ
2.3 A Slow Introduction to Pretty Damn Quick (PDQ)
- How to Build PDQ Models
- The Perl PDQ Module
- Classic Queues in PDQ
3 Quantitative Scalability on a Stick
3.1 Universal Law of Computational Scaling
- Fundamental Concepts of Scaling
- Geometric Scaling
- Allometric Scaling
- Computer system scaling
3.2 Hardware Scalability
- Ideal Parallelism
- AmdahlÕs Law
- The Role of Coherency
3.3 Software Scalability
- AmdahlÕs Law for Software
- SPEC SDM benchmark example
- Windows-based applications
- Multi-tier architectures
3.4 Virtual Load Testing
- Steady-state testing procedures
- Minimal Dataset
- Capacity Ratios
- Regression Analysis using EXCEL
4 Virtualization and Standards
4.1 Spectrum of Virtualization: From HTT to GRIDs
- Performance of hyperthreaded multicore processors
- Performance of Virtual Machine Monitors (VMMs) e.g., VMWare, Xen
4.2 The Fair-Share Scheduler
- Principles of operation
- PDQ model of the fair-share scheduler
- Performance impact of share allocations
- Capacity planning guidelines
4.3 Industry Standards and Benchmarks
- The Central Problem
- ARM (Application Resource Mgmt)
- SNMP (Simple Network Management Protocol)
- OpenGroup UMA (Universal Measurement Architecture)
- SPEC Benchmarks: CINT/CFP, SDM, SFS, WEB, JVM
- TPC Benchmarks: A, B, C, D(H and R), W
- Custom benchmarking prats and pitfalls
5 Distributed Systems Planning
5.1 Website Planning
- Analysis of Daily Traffic
- Effective Demand metric
- Short-term Capacity Planning (multivariate regression)
- Long-term Capacity Planning (nonlinear regression)
- Estimating the Doubling Period
5.2 Internet Planning
- The Bellcore Traces
- Fractals and Self-Similar Internet traffic
- Impact on Buffer Sizing
- New Developments
5.3 Gargantuan Computing for Guerrillas: GRIDs and P2P
- Scalability of Gnutella, Napster, Kazaa, Skype
- Hypernet Topologies
- Capacity Metrics
- Relative Bandwidth
Who Should Attend
UNIX sysadms, mainframe sysprogs, performance engineers, consultants,
IT technical managers and software development engineers.
This course does not assume prior experience with performance
analysis methods, but a working knowledge of computer systems (especially UNIX)
and college level algebra is helpful.
Terms and Conditions
Tuition Fee
See
class schedule
page for current pricing.
Early Bird: if registered 30 days in advance of the course.
Enroll
online
now!
Discounts
A corporate discount of 10% will be applied to the TOTAL enrollment fee
for THREE (3) or more people from the same company. Once a seat is
booked, a penalty of $500 will be imposed for a one-time transfer of
that seat to another session date. Inability to attend after such a
one-time transfer will automatically forfeit the entire registration
fee.
Refunds
Requests for refunds must be received in writing at least 30 days before the
start date of the course. There is a $50 processing
fee for cancellations. Substitutions may be made at any time.
You may do a one-time transfer of your current class enrollment to hold for an alternative class,
but there will be an additional $500 fee for such transfers.
Transportation
Information will be sent upon receipt of enrollment. A packet will include
airport and transportation options.
Reservations
Enrollment is limited to 30 students. All confirmed
reservations must be must be accompanied by a purchase order number, a
check for the tuition, or credit card information for billing. Courtesy
Reservations will be held for up to 30 days in order for paperwork to be
processed so long as there is suffcient time and adequate space in
thecourse.
Textbook
A copy of Dr. Gunther's performance analysis textbook:
Guerrilla Capacity Planning
(Springer-Verlag 2007) is included in the price of admission.
Sorry, no refunds or exchanges can be given if you already have a copy of the book.
File translated from
TEX
by
TTH,
version 3.38.
On 13 Feb 2008, 16:53.