More Tools
Done basic, most useful programming tools: MPI OpenMP What else is out there? Languages and compilers? Parallel computing is significantly more complicated than serial computing. Places a significant burden on the application developer, as we
Nic Brummell
Page 1
10/16/2014
AMS machines
http:/ams.soe.ucsc.edu/resources/computing
AMS has some decent sized machines!
NEW! 3 interactive analysis servers, 32 cores each (Dell PE R820: 4 x Intel Xeon
Sandy Bridge E5-4640 processor, each of which has 8
Chapter 4: Putting Components Together
So far, concentrated on deriving efficient parallel algorithms for individual components of a program.
Complete programs (BIG CODES!) may incorporate multiple parallel algorithms each requiring different
data structu
Case Study 1: Atmospheric Model
Conservation of momentum:
u #
tan &
1
p
% f + u
+ F
(v =
t $
a '
a cos
v #
tan &
1 p
+% f + u
+ F
(v =
t $
a '
a
Solution in two
horizontal directions,
latitude and longitude
Hydrostatic approximation:
g=
1 p
z
Conse
Performance Analysis and Debugging Tools
Performance analysis and debugging intimately connected since they both involve monitoring
of the software execution. Just different goals:
Debugging - achieve correct code behaviour
Performance analysis - achiev
Chapter 3: A Quantative Basis for Design
Real design tries to reach an optimal compromise between a number of thing
Execution time
Memory requirements
Implementation costs
Simplicity
Portability
Etc
Here try and form an understanding and make some e
Chapter 2: Designing Parallel Algorithms
Got ourselves some
parallel machine abstraction
parallel programming abstractions
Now need to translate a specification of a problem into an algorithm that displays concurrency,
scalability and locality (will do
Cloud Computing and
Big Data Processing
Shivaram Venkataraman
UC Berkeley, AMP Lab
Slides from Matei Zaharia
UC BERKELEY
Cloud Computing, Big Data
Hardware
Software
Open MPI
Google 1997
Data, Data, Data
Storage space must be used efficiently to store
i
AMS 250
Introduction to
High Performance Computing
Nic Brummell
brummell@soe.ucsc.edu
(831) 459 2122
Applied Math & Statistics (AMS)
University of California, Santa Cruz
Maybe we want to model something complicated
Global simulations: Differential rotati
Fortran 95 crash course
overview
File I/O
Introduction
Modern
features
Arrays # 1
Arrays # 2
Procedures
Fortran95
1
Part I: Getting started with
Fortran 95
Outline
First encounter with Fortran
Variables and their assignment
Control structures
Why learn