Model

Download as pdf or txt
Download as pdf or txt
You are on page 1of 14

1

MODULE I
Parallel computer models – Evolution of Computer Architecture, System attributes to
performance, Amdahl's law for a fixed workload. Multiprocessors and Multicomputers,
Multivector and SIMD computers, Architectural development tracks, Conditions of parallelism.

1.PARALLEL COMPUTER MODELS

 Parallel processing has emerged as a key enabling technology in modern computers, driven by
the ever-increasing demand for higher performance, lower costs, and sustained productivity in
real-life applications.
 Concurrent events are taking place in today's high-performance computers due to the common
practice of multiprogramming, multiprocessing, or multicomputing.

 Parallelism appears in various forms, such as pipelining, vectorization, concurrency,


simultaneity, data parallelism, partitioning, interleaving, overlapping, multiplicity, replication,
time sharing, space sharing, multitasking, multiprogramming, multithreading, and distributed
computing at different processing levels.

1.1THE STATE OF COMPUTING


1.1.1 Five Generation of Computers

Qn:Explain the five generations of computers?


2

1.1.2 Elements of Modern Computer

1.1.3 Evolution of Computer Architecture


Qn:Describe the evolution of parallel computer architecture?

Qn:Explain the term look ahead parallelism?


The study of computer architecture involves both programming/software requirements and hardware
organization. Therefore the study of architecture covers both instruction set architectures and machine
implementation organizations.

As shown in figure below, Evolution Started with the von Neumann architecture built as a sequential
machine executing scalar data . The sequential computer was improved from bit-serial to word—
parallel operations, and from fixed—point to floating point operations. The von Neumann architecture
is slow due to sequential execution of instructions in programs.

Lookahead , parallelism, and pipelining: Lookahead techniques were introduced to prefetch


instructions in order to overlap I/E (instruction fetch/ decode and execution) operations and to enable
functional parallelism .

Functional parallelism was supported by two approaches: One is to use multiple functional
units simultaneously, and the other is to practice pipelining at various processing levels.
The latter includes pipelined instruction execution, pipelined arithmetic computations, and
memory-access operations. Pipelining has proven especially attractive in performing identical
operations repeatedly over vector data strings.

A vector is one dimensional array of numbers. A vector processor is CPU that implements an
instruction set containing instructions that operate on one dimensional arrays of data called vectors.

Vector operations were originally carried out implicitly by software-controlled looping using
scalar pipeline processors.
3

Explicit vector instructions were introduced with the appearance of vector processors. A vector
processors equipped with multiple vector pipelines that can be concurrently used under hardware or
firmware control.

There are two Families of pipelined vector processors:


 Memory –to-memory- architecture supports the pipelined flow of vector operands directly
from the memory to pipelines and then back to the memory.
 Register-to register architecture uses vector registers to interface between the memory and
functional pipelines.

Another important branch of the architecture tree consists of the SIMD computers for synchronized
vector processing. An SIMD computer exploits spatial parallelism rather than temporal parallelism as
in a pipelined computer .SIMD computing is achieved through the use of an array of processing
elements [PEs] synchronised by the same controller. Associative memory can be used to build SIMD
associative processors.

Intrinsic parallel computers are those that execute programs in MIMD mode.
There are two major classes of parallel computers, namely, Shared memory multiprocessors and
message passing multicomputer. The major distinction between multiprocessors and multicomputer
lies in memory sharing and the mechanisms used for interprocessor communication.

The processors in a multiprocessor system communicate with each other through shared
variables in a common memory.
Each computer node in a multicomputer system has a local memory, unshared with other
nodes. lnterprocessor communication is done through message passing among the nodes.
4

Flynn’s Classification (classified into 4)

Qn:Describe briefly about the operational model of SIMD computer with an example?

Qn: Characterize the architectural operations of SIMD and MIMD computers?

Describe briefly about Flynn’s classification?


• Michael Flynn (1972) introduced a classification of various computer architectures based on
notions of instruction and data streams. Stream denote a sequence of items (instructions or
data) as executed or operated upon by a single processor. Two types of information flow into a
processor: instructions and data. The instruction stream is defined as the sequence of
instructions performed by the processing unit. The data stream is defined as the data traffic
exchanged between the memory and the processing unit.

• Both instructions and data are fetched from the memory modules. Instructions are decoded by
the control unit, which sends the decoded instruction to the processor units for execution. Data
streams flow between the processors and the memory bidirectionally. Each instruction stream is
generated by an independent control unit.

According to Flynn’s classification, either of the instruction or data streams can be single or multiple.
Computer architecture can be classified
into the
 single-instruction single-data streams (SISD);
 single-instruction multiple-data streams (SIMD);
 multiple-instruction single-data streams (MISD); and
 multiple-instruction multiple-data streams (MIMD).

1. SISD(Single Instruction Single Data Stream)

• Conventional sequential machines are called SISD -[single instruction stream over single data
stream] computers. Instructions are executed sequentially but may be overlapped in their
execution stages (pipelining).

2. SIMD(Single Instruction Multiple Data Stream) – Represents vector computers/array processors


equipped with scalar and vector hardware.There are multiple processing elements supervised by the
same control unit. All PEs receive the same instruction broadcast from the control unit but operate on
different data sets from distinct data streams.
5

3. MIMD(multiple instructions over multiple data stream) – most popular model. parallel computers
ters are reserved for MIMD

4. MISD(multiple instruction over single data stream)

The same data stream flows through a linear array of processors executing different instruction streams.
This architecture is also known as systolic arrays For pipelined execution of specific algorithms.

Of the four machine models, most parallel computers built in the past assumed the MIMD model
for general purpose computations. The SIMD and MISD models are more suitable for special-purpose
6

computations. For this reason, MIMD is the most popular model, SIMD next, and MISD the least
popular model being applied in commercial machines.

Six Layers for Computer System Development


Qn:describe Six layers of computer system development?

A layered development of parallel computers is illustrated in Fig. above, based on a


classification by Lionel Ni [I990].
• Hardware configurations differ from machine to machine, even those of the same model.

• The address space of a processor in a computer system varies among different architectures. It
depends on the memory organization, which is machine—dependent. These features are up to the
designer and should match the target application domains.

• On the other hand, we want to develop application programs and programming environments
which are machine-independent. Independent of machine architecture, the user programs can be
ported to many computers with minimum conversion costs.

• High- level languages and communication models depend on the Architectural choices made in a
computer system. From a programmer's viewpoint, these two layers should be architecture-
transparent.

• Programming languages such as Fortran, C, C++, Pascal, Ada, Lisp and others can be supported
by most computers. However, the communication models, shared variables versus message
passing, are mostly machine-dependent.
Challenges in Parallel Processing: Major challenge is on the software and application side.It is still difficult
to program parallel and vector computers.High performance computers should provide fast and accurate
solutions to scientific ,engineering , business,social and defense problems.

2.SYSTEM ATTRIBUTES TO PERFORMANCE


Qn:Define the terms a)clock rate, b)CPI, MIPS rate, Throughput rate?
Qn:List out the matrics affecting scalability of a computer system for a given application?(
Ic,p,m,k t)
Qn:List and explain four system attributes affecting the performance of CPU?(
instruction-set architecture, compiler technology, CPU implementation and control, and
cache and memory hierarchy)
7

System Attributes versus Performance Factors

The ideal performance of a computer system requires a perfect match between machine
capability and program behaviour.

Machine capability can be enhanced with better hardware technology, however program behaviour
is difficult to predict due to its dependence on application and run-time conditions.

Below are the five fundamental factors for projecting the performance of a computer.

CPU is driven by a clock of constant clock with a cycle time ().he inverse of cycle time is the clock
rate (f=1/

Size of the program is determined by the Instruction Count(Ic). Different instructions in a particular
program may require different number of clock cycles to execute. So,

Cycles per instruction (CPI):-is an important parameter for measuring the time needed to execute an
instruction .

Execution Time/CPU Time (T): Let Ic be Instruction Count or total number of instructions in the
program. The Execution Time or CPU time (T) will be:

T = Ic CPI  Eq.1.1

• The execution of an instruction involves the instruction fetch, decode, operand fetch, execution
and store results.
Only instruction decode and execution phases are carried out in CPU. The remaining three operations
may require access to memory
The CPI of an instruction type can be divided into two component terms corresponding to the total
processor cycles and memory cycles needed to complete the execution of the instruction. That is :-
CPI = Instruction Cycle = Processor Cycles +Memory Cycles. ie
CPI = Instruction cycle = p + m + k
where
m = number of memory references
P = number of processor cycles
k = latency factor (how much the memory is slow w.r.t to CPU)

Therefore, Equation 1.1 can be rewrite as follows:-

Eq.1.2
T = Ic (p+m+k) 

From the above equation the five factors affecting performance are:-Ic ,p,m,k,

System Attributes: The above five performance factors (Ic ,p,m,k,) are influenced by four system
attributes: instruction-set architecture, compiler technology, CPU implementation and control,
and cache and memory hierarchy, as specified in Table 1.2 below.

The instruction-set architecture affects the program length (1,) and processor cycles needed (p). The
compiler technology affects the values of Ic, p, and the memory reference count (m). The CPU
implementation and control determine the total processor time (p *) needed. Finally, the memory
technology and hierarchy design affect the memory access latency (k * ). The above CPU time can be
used as a basis in estimating the execution rate of a processor.
8

MIPS Rate. The processor speed is often measured in terms of million instructions per second
(MIPS). We simply call it the MIPS rate of a given processor.

Let C be the total number of cycles needed to execute a given program (ie C=Ic * CPI).
Then the CPU time in Eq. 1.2 can be estimated as
T = C* 
= C/f .

Furthermore, CPI = C/Ic and


T = Ic* CPI * 
= Ic * CPI/f.

MIPS rate = Ic/(T*106 )


Eq.1.3
=f/(CPI*106)
=(f*Ic)/(C*106)

Or

Now Based on Equation 1.2 and 1.3


CPU Time ,T = (IC*10-6)/MIPS

MFLOPS:
Most compute intensive applications in science and engineering make heavy use of floating point
operations. For such applications a more relevant measure of performance is floating point operations
per second abbreviated as mflops. With prefix mega(10^6),giga(10^9) tera(10^12) or peta(10^15).
Floating-point performance is expressed as millions of floating-point operations per second (MFLOPS),
defined as follows ,only with floating-point instructions.
Number of executed floating-point operations in a program
9

MFLOPS = _______________________________________________
6
execution time * 10

Throughput Rate:- Number of programs executed per unit time is called system throughput ws(in
programs per second).In a multiprogrammed system, the system throughput is often lower than CPU
throughput Wp defined by

Wp = f Eq.1.4

Ic *CPI

W=1/T
OR
W =( MIPS*106 )/Ic

Problems:-
1 A benchmark program is executed on a 40MHz processor. The benchmark program has the
following statistics.

Calculate average CPI, MIPS rate & execution time for the above benchmark program
Given Clock speed of the processor= 40 MHz=40*106 Hz
Average CPI= C/Ic
=Total cycles to execute the program/Instruction count
=(45000 *1 + 32000*2 + 1500*2 + 8000*2) /(45000 + 3200 + 15000 + 8000)
=155000/100000
=1.55

Execution Time ,T= C / f


T = 150000 / 40 *106
T = 0.155 / 40
T = .003875 s
T = 3.875 ms (since 1s=1000ms)

MIPS rate = Ic / T * 106


MIPS rate = 100000/(0.003875*106)
=
25.8

2. Consider the execution of an object code with 2 *106 instructions on a 400 MHz processor. The
program consists of four major types of instructions. The instruction mix and the number of
cycles [CPI] needed for each instruction type are given below based on the result of a program
trace experiment:
10

(a) Calculate the average CPl when the program is executed on a uniprocessor with the above
trace results.
(b) Calculate the corresponding MIPS rate based on the CPI obtained in part (a).

Answer:
average CPI = 0.6 +(2 *0.18) +(4* 0.12) +(8 *0.1) =2.24.
MIPS rate = f/(CPI*106)
=(400* 106) /(2.24 *106) =178

Programming Environments:
Qn: Difference between Implicit Parallelism and Explicit forms of Parallelism?
The programmability of a computer depends on the programming environment provided to the users.
conventional uniprocessor computers are programmed in a sequential environment in which
instructions are executed one after another in a sequential manner. Parallel computers employs parallel
environment where parallelism is automatically exploited.. Based on the programming environments
required parallelism can be of two types:
11

IMPLICIT PARALLELISM EXPLICIT PARALLELISM


1. In computer science, implicit 1. In computer programming, explicit
parallelism is a characteristic of a parallelism is the representation of
programming language that allows a concurrent computations by means of
compiler or interpreter to primitives in the form of special-purpose
automatically exploit the parallelism directives or function calls. Most parallel
inherent to the computations expressed primitives are related to process
by some of the language's constructs. synchronization, communication or task
partitioning.
2. Uses conventional languages such as
C, C++, Fortran or Pascal to write 2. Requires more effort by programmers to
source program develop a source program using parallel
dialects like C, C++, Fortran and Pascal.

3. The sequentially coded source 3. Parallelism is explicitly specified in the


program is translated into parallel user programs.
object code by a parallelizing
compiler.

4. Compiler detects parallelism and 4. Burden on compiler is reduced as


assigns target machine resources. parallelism specified explicitly.

5. Success relies on intelligence of 5. Programmer’s effort is more- special s/w


parallelizing compiler. Requires less tools needed to make environment more
effort from programmers. friendly to user groups.

6. Applied in shared memory 6. Applied in Loosely coupled


multiprocessors. Multiprocessors to tightly coupled VLIW

3. MULTIPROCESSORS AND MULTICOMPUTERS


Qn. Distinguish between multiprocessor and multicomputers based on their structure,
resource sharing, and interprocessor communication)?

Qn:List differences between UMA, NUMA,COMA Models?

Qn:Describe in detail the types of shared memory multiprocessors models?


- 2 categories of parallel computer
- Distinguished by having shared memory(multiprocessors) or unshared distributed memory(multi
computers)

Multiprocessors Multicomputer
1. Single computer with multiple 1. Multiple autonomous computers
processors

2. Each PE’s(CPU/processing elements) 2. Each PE’s has its own memory and
do not have their own individual resources – no sharing – Thus called
memories – memory and I/O resources Distributed Memory Multicomputers
are shared – Thus called Shared
Memory Multiprocessors
3. Communication between PE’s not
12

3. Communication between PE’s a must mandatory

4. Tightly coupled – due to high degree of 4. Loosely coupled as there is no resource


resource sharing sharing

5. Use Dynamic Network – thus 5. Use Static Network – connection of


communication links can be switching units is fixed
reconfigured
6. Ex: Message Passing Multicomputer
6. Ex: Sequent Symmetry S-81
7. NORMA model/ Distributed-
Memory Multicomputer
7. 3 Types

- UMA model
- NUMA model
COMA model

THE UMA MODEL

- Physical memory is uniformly shared by all processors


- All processors (PE1….PEn) take equal access time to memory – Thus its termed as Uniform
Memory Access Computers
- Each PE can have its own private Cache
- High degree of resource sharing(memory and I/O ) – Tightly Coupled
- Interconnection Network can be – Common bus, cross bar switch or Multistage n/w ( discussed
later)
- When all PE’s have equal access to all peripheral devices – Symmetric Multiprocessor
- In Asymmetric multiprocessor only one subset of processors have peripheral access. Master
Processors control Slave (attached) processors.

Applications of UMA Model

- Suitable for general purpose and time sharing application by multiple users
- Can be used to speed up execution of a single program in time critical application
13

DISADVANTAGES

- Interacting process cause simultaneous access to same locations – cause problem when an
update is followed by read operation (old value will be read)
- Poor Scalability – as no: of processors increase –shared memory area increase-thus n/w becomes
bottleneck.
- No: of processors usually in range(10-100)

THE NUMA MODEL (Ex: BBN Butterfly)

- Access time varies with location of memory


- Shared memory is distributed to all processors – Local memories
- Collection of all local memories forms global memory space accessible by all processors.
- Its faster to access content within local memory of a processor than to access remote
memory attached to another processor (delay through interconnection) – Thus named as
NON-Uniform Memory access – because access time depends on whether data available in
local memory of the processor itself or not.

Advantage

- – reduces n/w bottleneck issue that occurs in UMA as prcessors have a direct access path to
attached local memory

3 types of Memory access pattern

I. Fastest to access – local memory of PE itself


II. Next fastest – global memory shared by PE/Cluster
III. Slowest to access – remote memory(local memory of PE from another Cluster)
14

THE COMA MODEL

- Multiprocessor + Cache Memory = COMA Model


- Multiprocessor using cache only memory
- Ex: Data diffusion Machine , KSR-1 machine
- Special case of NUMA machine in which distributed main memories are converted to caches –
all caches together form a global address space
- Remote cache access is assisted by Distributed Cache directories (D in fig abv)

Application of COMA

- General purpose multiuser applications

DISTRIBUTED MEMORY / NORMA / MULTICOMPUTERS (Intel Paragon,


nCUBE, SuperNode1000)

- Consists of multiple computers(called nodes)


- A node is an autonomous computer consisting of processor, local memory attached disks or
I/O peripherals.
- Nodes interconnected by a message passing network – can be Mesh, Ring, Torus, Hypercube
etc (discussed later)
- All interconnection provide point to point static connection among nodes

You might also like