Palestra ministrada por Luciano Palma no Intel Software Conference 2013, nos dias 6 de Agosto (NCC/UNESP/SP) e 12 de Agosto (COPPE/UFRJ/RJ).

1. Computação Paralela:
Benefícios e Desafios
Luciano Palma
Community Manager – Servers & HPC
Intel Software do Brasil
Luciano.Palma@intel.com

2. A Capacidade Computacional evoluiu.
O Software não acompanhou.
A densidade de transistores continua aumentando, mas…
… o aumento da velocidade (clock) não
A grande maioria dos PCs vendidos são multi-core, mas…
… muitos programas / cargas ainda não tiram proveito
do paralelismo possível
O Processamento Paralelo é chave para obter
o máximo
desempenho
possível

3. Lei de Moore
Se transistores fossem pessoas
Agora imagine o que 1,3 bilhões de pessoas poderiam fazer num palco.
Essa é a escala da Lei de Moore
0.1
1
10
100
1000
1970
1975
1980
1985
1990
1995
2000
2005
2010
2015
2020
Watts
Per Processor
Power
Wall
0.1
1
10
100
1000
0.1
1
10
100
1000
1970
1975
1980
1985
1990
1995
2000
2005
2010
2015
2020
1970
1975
1980
1985
1990
1995
2000
2005
2010
2015
2020
Watts
Per Processor
Power
Wall
1
10
100
1000
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
TPCC 4 Processor
Published
Single Core
Dual
Core
Quad
Core
1
10
100
1000
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
TPCC 4 Processor
Published
1
10
100
1000
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
TPCC 4 Processor
Published
Single Core
Dual
Core
Quad
Core

4. Por que Programação Paralela é Importante?
Competitividade
na Indústria
Pesquisa
Científica
Segurança
Nacional
Modelagem de
Clima/Tempo
Segurança Nacional
Pesquisa Farmacêutica
Maiores Desafios  Maior Complexidade Computacional…
… mantendo um “orçamento energético” realista
Imagens Médicas
Exploração de Energia
Simulações
Desempenho Computacional
Total por país
Análises Financeiras
Projeto de novos Produtos
CAD/manufatura
Criação de Conteúdo Digital
Corrida Computacional

5. Em Computação de Alto Desempenho
(HPC), a Simulação é crucial
• The Scientific Method is Dead-Long Live the (New) Scientific Method, June 2005
• Richard M. Satava, MD Journal of Surgical Innovation
A Nova Computação estimulou um Novo Método Científico*
Método Científico Clássico
Hipótese Análise
Conclusão
Refinamento
Experimentação
Modelagem e
Simulação/Refinamento do
Experimento
Previsão
Análise
Conclusão
Refinamento
Hipótese Experimentação
Para simular efetivamente… precisamos nos basear em computação paralela!

6. Exemplo de como a
Computação Paralela tem ajudado...
Problema: Processamento de
Imagens de Ressonância
Magnética Pediátrica é difícil. A
reconstrução da imagem precisa
ser rápida
• Crianças não ficam paradas
não seguram a respiração
• Baixa tolerância a exames demorados
• Custos e Riscos da Anestesia
Solução: MRI Avançada:
Processamento Paralelo de
Imagens e Compressed Sensing
reduziram dramaticamente o
tempo de aquisição da MRI
• Reconsturção 100x mais rápida
• MRI de qualidade mais alta e mais rápida
• Esta imagem: paciente de 8 meses com
massa cancerígena no fígado
– Reconstrução Serial: 1 hora
– Reconstrução Paralela: 1 minuto

7. Computação Serial vs. Paralela
Serialização
 Faz Sentido!
 Fácil para fazer “debug”
 Determinístico
 No entanto…
… aplicações serializadas não maximizam o desempenho de saída
(output) e não evoluirão no tempo com o avanço das tecnologias
Paralelização – dá para fazer?!
 Depende da quantidade de trabalho
a realizar e da habilidade
de dividir a tarefa
 É necessário entender a entrada (input), saída (output)
e suas dependências
for (i=0; i< num_sites; ++i) {
search (searchphrase, website[i]);
}
parallel_for (i=0; i< num_sites; ++i) {
search (searchphrase, website[i]);
}

8. Dividindo o trabalho…
Decomposição de Tarefas:
Executa diferentes funções do
programa em paralelo.
Limitada escalabilidade, portanto
precisamos de…
Decomposição de Dados:
Dados são separados em blocos e
cada bloco é processado em uma
task diferente.
A divisão (splitting) pode ser
contínua, recursiva.
Maiores conjuntos de dados 
mais tasks
O Paralelismo cresce à medida que
o tamanho do problema cresce
#pragma omp parallel shared(data, ans1, ans2)
{
#pragma omp sections
{
#pragma omp section
ans1=do_this(data);
#pragma omp section
ans2=do_that(data);
}
}
#pragma omp parallel_for shared(data, ans1)
private(i)
for(i=0; i<N; i++) {
ans1(i) = do_this(data(i));
}
#pragma omp parallel_for shared(data, ans2)
private(i)
for(i=0; i<N; i++) {
ans2(i) = do_that(data(i));
}

9. Técnicas de Programação Paralela:
Como dividir o trabalho entre sistemas?
É possível utilizar Threads (OpenMP, TBB, Cilk, pthreads…) ou
Processos (MPI, process fork..) para programar em paralelo
Modelo de Memória Compartilhada
 Único espaço de memória utilizado por múltiplos processadores
 Espaço de endereçamento unificado
 Simples para trabalhar, mas requer cuidado para evitar condições de corrida
(“race conditions”) quando existe dependência entre eventos
Passagem de Mensagens
 Comunicação entre processos
 Pode demandar mais trabalho para implementar
 Menos “race conditions” porque as mensagens podem forçar o sincronismo
“Race conditions”
acontecem quando processos ou
threads separados modificam ou
dependem das mesmas coisas…
Isso é notavelmente difícil de
“debugar”!

10. Exemplo de uso
OpenMP – Cálculo de π

11. Cálculo de π – Teoria
π pode ser calculado através da integral abaixo
π = ∫ f(x) * dx = 4 / (1 + x2) * dx = 3.141592654...0
1

12. π = 4 / (1 + 0,52) * 1 = 4 / 1,25  π = 3,2
Cálculo de π – Primeira Aproximação
Valor grosseiramente aproximado de π

13. f(0,25) = 3,735; f(0,75) = 2,572  π = 3,153
3,735 * 0,5
= 1,867
2,572 * 0,5
= 1,286
Cálculo de π – Segunda Aproximação
Com mais aproximações, o valor de π tende ao valor teórico

14. Aproximação com 10 intervalos  π = 3,14243,990*0,1=0,399
3,912*0,1=0,391
3,765*0,1=0,376
3,563*0,1=0,356
3,326*0,1=0,333
3,071*0,1=0,397
2,812*0,1=0,281
2,560*0,1=0,256
2,322*0,1=0,232
2,102*0,1=0,210
Cálculo de π – Aumentando os intervalos…
Quanto mais intervalos, maior a precisão do cálculo

15. Solução do Problema com Paralelismo
Cada thread realiza o cálculo de alguns intervalos
3,912*0,1=0,391
3,765*0,1=0,376
3,563*0,1=0,356
3,326*0,1=0,333
3,071*0,1=0,397
2,812*0,1=0,281
2,560*0,1=0,256
2,322*0,1=0,232
2,102*0,1=0,210
3,990*0,1=0,399
REDUÇÃO
PI = 3,1424
3,912*0,1=0,3913,912*0,1=0,391
3,765*0,1=0,3763,765*0,1=0,376
3,563*0,1=0,3563,563*0,1=0,356
3,326*0,1=0,3333,326*0,1=0,333
3,071*0,1=0,3973,071*0,1=0,397
2,812*0,1=0,2812,812*0,1=0,281
2,560*0,1=0,2562,560*0,1=0,256
2,322*0,1=0,2322,322*0,1=0,232
2,102*0,1=0,2102,102*0,1=0,210
3,990*0,1=0,3993,990*0,1=0,399
3,912*0,1=0,3913,912*0,1=0,3913,912*0,1=0,3913,912*0,1=0,391
3,765*0,1=0,3763,765*0,1=0,3763,765*0,1=0,3763,765*0,1=0,376
3,563*0,1=0,3563,563*0,1=0,3563,563*0,1=0,3563,563*0,1=0,356
3,326*0,1=0,3333,326*0,1=0,3333,326*0,1=0,3333,326*0,1=0,333
3,071*0,1=0,3973,071*0,1=0,3973,071*0,1=0,3973,071*0,1=0,397
2,812*0,1=0,2812,812*0,1=0,2812,812*0,1=0,2812,812*0,1=0,281
2,560*0,1=0,2562,560*0,1=0,2562,560*0,1=0,2562,560*0,1=0,256
2,322*0,1=0,2322,322*0,1=0,2322,322*0,1=0,2322,322*0,1=0,232
2,102*0,1=0,2102,102*0,1=0,2102,102*0,1=0,2102,102*0,1=0,210
3,990*0,1=0,3993,990*0,1=0,3993,990*0,1=0,3993,990*0,1=0,399
Thread Thread Thread Thread

16. Agora um pouco de hardware…

17. 17
Engenharia e Arquitetura…

18. Intel inside, inside Intel…
Um processador com múltiplos cores possui
componentes “comuns” aos cores.
Estes compontentes recebem o nome de Uncore.

19. Intel inside, inside Intel…
E agora, dentro do core…

20. Por que descer neste nível?
Utilizar todas as “threads de hardware”
 Não esquecer o HyperThread
Utilizar todas as unidades de execução
 Retirar o máximo de instruções por ciclo de clock
Otimizar o uso dos unidades de vetores (AVX/AVX2)
 Loops otimizados
Manter as caches com dados/instruções válidos
 Evitar “cache misses”
Aproveitar o “branch prediction”
 Evitar o “stall” da pipeline
Para tirar o máximo proveito dos recursos do hardware!

21. Por que descer neste nível?
Utilizar todas as “threads de hardware”
 Não esquecer o HyperThread
Utilizar todas as unidades de execução
 Retirar o máximo de instruções por ciclo de clock
Otimizar o uso dos unidades de vetores (AVX/AVX2)
 Loops otimizados
Manter as caches com dados/instruções válidos
 Evitar “cache misses”
Aproveitar o “branch prediction”
 Evitar o “stall” da pipeline
Para tirar o máximo proveito dos recursos do hardware!

22. Por que descer neste nível?
Utilizar todas as “threads de hardware”
 Não esquecer o HyperThread
Utilizar todas as unidades de execução
 Retirar o máximo de instruções por ciclo de clock
Otimizar o uso dos unidades de vetores (AVX/AVX2)
 Loops otimizados
Manter as caches com dados/instruções válidos
 Evitar “cache misses”
Aproveitar o “branch prediction”
 Evitar o “stall” da pipeline
Para tirar o máximo proveito dos recursos do hardware!

23. Existem recursos para lhe ajudar!
Ferramentas de Software da Intel
IDZ – Intel Developer Zone
http://software.intel.com

24. Intel® Cilk™ Plus
• Extensões para
as linguagens
C/C++ para
simplificar o
paralelismo
• Código aberto
Também um
produto Intel
Intel® Threading
Building Blocks
• Template
libraries
amplamente
usadas em C++
para paralelismo
• Código aberto
Também um
produto Intel
Domain Specific
Libraries
• Intel® Integrated
Performance
Primitives
• Intel® Math
Kernel Library
Padrões
estabelecidos
• Message Passing
Interface (MPI)
• OpenMP*
• Coarray Fortran
• OpenCL*
Modelos de Programação Paralela
Níveis de abstração conforme a necessidade
Mesmos modelos para multi-core (Xeon) e
many-core (Xeon Phi)

25. Nota sobre Otimização

26. • INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR
OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS
OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING
TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE,
MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.
• A "Mission Critical Application" is any application in which failure of the Intel Product could result, directly or indirectly, in personal injury or
death. SHOULD YOU PURCHASE OR USE INTEL'S PRODUCTS FOR ANY SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL
AND ITS SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS, OFFICERS, AND EMPLOYEES OF EACH, HARMLESS AGAINST ALL
CLAIMS COSTS, DAMAGES, AND EXPENSES AND REASONABLE ATTORNEYS' FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY CLAIM OF PRODUCT
LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS
SUBCONTRACTOR WAS NEGLIGENT IN THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS.
• Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics
of any features or instructions marked "reserved" or "undefined". Intel reserves these for future definition and shall have no responsibility whatsoever
for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design
with this information.
• The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published
specifications. Current characterized errata are available on request.
• Intel processor numbers are not a measure of performance. Processor numbers differentiate features within each processor family, not across different
processor families. Go to: http://www.intel.com/products/processor_number.
• Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
• Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1-800-548-
4725, or go to: http://www.intel.com/design/literature.htm
• Intel, Core, Atom, Pentium, Intel inside, Sponsors of Tomorrow, Pentium, 386, 486, DX2 and the Intel logo are trademarks of Intel Corporation in the
United States and other countries.
• *Other names and brands may be claimed as the property of others.
• Copyright ©2012 Intel Corporation.
Legal Disclaimer

27. Risk Factors
The above statements and any others in this document that refer to plans and expectations for the second quarter, the year and the future are forward-looking
statements that involve a number of risks and uncertainties. Words such as “anticipates,” “expects,” “intends,” “plans,” “believes,” “seeks,” “estimates,” “may,”
“will,” “should” and their variations identify forward-looking statements. Statements that refer to or are based on projections, uncertain events or assumptions
also identify forward-looking statements. Many factors could affect Intel’s actual results, and variances from Intel’s current expectations regarding such factors
could cause actual results to differ materially from those expressed in these forward-looking statements. Intel presently considers the following to be the
important factors that could cause actual results to differ materially from the company’s expectations. Demand could be different from Intel's expectations due
to factors including changes in business and economic conditions, including supply constraints and other disruptions affecting customers; customer acceptance of
Intel’s and competitors’ products; changes in customer order patterns including order cancellations; and changes in the level of inventory at customers.
Uncertainty in global economic and financial conditions poses a risk that consumers and businesses may defer purchases in response to negative financial events,
which could negatively affect product demand and other related matters. Intel operates in intensely competitive industries that are characterized by a high
percentage of costs that are fixed or difficult to reduce in the short term and product demand that is highly variable and difficult to forecast. Revenue and the
gross margin percentage are affected by the timing of Intel product introductions and the demand for and market acceptance of Intel's products; actions taken
by Intel's competitors, including product offerings and introductions, marketing programs and pricing pressures and Intel’s response to such actions; and Intel’s
ability to respond quickly to technological developments and to incorporate new features into its products. Intel is in the process of transitioning to its next
generation of products on 22nm process technology, and there could be execution and timing issues associated with these changes, including products defects
and errata and lower than anticipated manufacturing yields. The gross margin percentage could vary significantly from expectations based on capacity utilization;
variations in inventory valuation, including variations related to the timing of qualifying products for sale; changes in revenue levels; segment product mix; the
timing and execution of the manufacturing ramp and associated costs; start-up costs; excess or obsolete inventory; changes in unit costs; defects or disruptions
in the supply of materials or resources; product manufacturing quality/yields; and impairments of long-lived assets, including manufacturing, assembly/test and
intangible assets. The majority of Intel’s non-marketable equity investment portfolio balance is concentrated in companies in the flash memory market segment,
and declines in this market segment or changes in management’s plans with respect to Intel’s investments in this market segment could result in significant
impairment charges, impacting restructuring charges as well as gains/losses on equity investments and interest and other. Intel's results could be affected by
adverse economic, social, political and physical/infrastructure conditions in countries where Intel, its customers or its suppliers operate, including military conflict
and other security risks, natural disasters, infrastructure disruptions, health concerns and fluctuations in currency exchange rates. Expenses, particularly certain
marketing and compensation expenses, as well as restructuring and asset impairment charges, vary depending on the level of demand for Intel's products and
the level of revenue and profits. Intel’s results could be affected by the timing of closing of acquisitions and divestitures. Intel's results could be affected by
adverse effects associated with product defects and errata (deviations from published specifications), and by litigation or regulatory matters involving
intellectual property, stockholder, consumer, antitrust, disclosure and other issues, such as the litigation and regulatory matters described in Intel's SEC reports.
An unfavorable ruling could include monetary damages or an injunction prohibiting Intel from manufacturing or selling one or more products, precluding particular
business practices, impacting Intel’s ability to design its products, or requiring other remedies such as compulsory licensing of intellectual property. A detailed
discussion of these and other factors that could affect Intel’s results is included in Intel’s SEC filings, including the report on Form 10-K for the year ended Dec.
31, 2011.
Rev. 4/17/12