OpenGenus Tech Review Team - OpenGenus IQ: Learn Algorithms, DL, System Design (Page 143)

Central Processing Unit (CPU) vs Graphics Processing Unit (GPU) vs Tensor Processing Unit (TPU)

Central Processing Unit (CPU), Graphics Processing Unit (GPU) and Tensor Processing Unit (TPU) are processors with a specialized purpose and architecture. We have compared these in respect to Memory Subsystem Architecture, Compute Primitive, Performance, Purpose, Usage and Manufacturers.

Software Engineering

CAP theorem (Brewer's theorem)

The CAP theorem (or Brewer Theorem) is the idea that a distributed computing system is not able to provide partition tolerance, consistency and availability at the same time. It was developed in 2000 by Eric Allen Brewer. CAP theorem has applications in database and server infrastructure

instruction set

Explicitly parallel instruction computing (EPIC)

EPIC (Explicitly Parallel Instruction Computing) is a 64-bit microprocessor instruction set which is an improvement to the VLIW (Very Large Instruction Word) architecture. It has been developed by Intel and Hewlett Packard. It uses speculative loading, predication, and explicit parallelism

instruction set

Very long instruction word (VLIW)

Very long instruction word (VLIW) is an instruction set architecture designed to take full advantage of instruction level parallelism in form of pipelining, multiple processors, superscalar implementation and multiple independent operations. It has its advantages and disadvantages and is used widely

TensorFlow

How TensorFlow uses Graph data structure concepts?

In this article, we explain various concepts in TensorFlow such as tensors, dataflow graphs and several optimizations such as decision tree pruning and demonstrate the use of graph data structure and algorithm concepts in TensorFlow

instruction set

Minimal Instruction Set Computer (MISC)

Minimal instruction set computer (MISC) is a processor architecture with a very small number of basic operations and corresponding opcodes in its instruction set. MISC has its own advantages and disadvantages. It is commercially used as STEREO and INMOS transputer

instruction set

One Instruction Set Computer (OISC)

One Instruction Set Computer (OISC) is a computer architecture that has only one instruction in its instruction set. It is based on bit manipulating machine, transport triggered architecture and arithmetic based turing complete machines. It has its own advantages and used commercially as high subleq

instruction set

Zero instruction set computer (ZISC)

Zero instruction set computer (ZISC) is a computer architecture based on two fundamental ideas like pattern matching and absence of micro instructions. ZISC has its own advantages and is commercially used by IBM in ZISC35 and by Intel s NI1000. CM1K another ZISC has been developed recently.

instruction set

Examples of Instruction Sets

We have demonstrated examples of instruction set architectures from various categories such as RISC, CISC, MISC, VLIW, EPIC, OISC and ZISC. Examples include ARM, MIPS, OpenRISC, SPARC, x86, z architecture, Intel 8080, Transputer, Transmeta Crusoe, Elbrus 2000, Itanium, Cryptoleq, NI1000 and CM1K.

instruction set

7 Types of Instruction Set

Instruction sets are Reduced Instruction Set Computer (RISC), Complex Instruction Set Computer (CISC), Minimal instruction set computers (MISC), Very long instruction word (VLIW), Explicitly parallel instruction computing (EPIC), One instruction set computer (OISC) and Zero instruction set computer

Computer Architecture

Complex Instruction Set Computer (CISC) architecture explained

CISC Complex Instruction Set Computer architecture focuses on reducing the number of instructions per program It has emphasis on hardware design, has multi clock complex instructions, memory to memory instructions, high cycles per second, small code size and uses transistors for storing instructions

Computer Architecture

Reduced Instruction Set Computer (RISC) architecture explained

RISC (Reduced Instruction Set Computer) architecture focuses on reducing the number of cycles per instruction. It has emphasis on software design, has single clock, reduced instructions only, register to register independent instruction, low cycles per second and large code size. See a RISC example

Machine Learning (ML)

Key ideas that makes Graphics Processing Unit (GPU) work so fast

We have explored the key ideas that are used in Graphics Processing Unit to make it so fast. Ideas include many cores in parallel, pack cores full of ALUs by sharing instruction stream by explicit SIMD vector instruction and avoid latency stalls by interleaving execution of many groups.

Machine Learning (ML)

Basic Graphics Processing Unit (GPU) design concepts

we have explored some of the basic architecture concepts in Graphics Processing Unit (GPU) such as graphics pipeline, vector processing, primitive processing, rasterization, fragment processing, pixel operations, graphics architecture and shader programming model There are five basic graphics entity

Machine Learning (ML)

What is a Graphics Processing Unit (GPU)?

A graphics processing unit (GPU) is a processor like CPU and TPU for faster graphics processing. Specifically, it designed to rapidly manipulate and alter memory to accelerate the creation of images in a frame buffer to be displayed on a screen. GPUs are developed by Intel, Nvidia and AMD (ATI).

Machine Learning (ML)

Tensor Processing Unit (TPU) explained

A tensor processing unit (TPU) is a proprietary processor designed by Google in 2016 for use in neural networks inference. Norm Jouppi was the Technical leader of the TPU project. Key ideas in TPU include Matrix Multiplier Unit, Unified Buffer, Activation Unit and systolic array

Software Engineering

Difference between forward proxy server and reverse proxy server

We explored the difference between forward proxy server and reverse forward proxy. Forward proxies are used for Content Filtering, eMail security, NAT’ing and Compliance Reporting. Reverse proxy are used for Content Distribution Network

Algorithms

KMP (Knuth-Morris-Pratt) Algorithm

Given a string S of length n and a pattern P of length m , you have to find all occurences of pattern P in string S provided n > m. Knuth Morris Pratt algorithm is an efficient pattern searching algorithm. Time and space complexity of KMP algorithm is O(m + n) linear.

Software Engineering

Branch prediction explained with a code example

Branch prediction is an optimization technique which predicts the path a code will take before it is known for sure. We have explained the concept with a C++ example of branch prediction where a condition statement runs slower in case of unsorted data compared to sorted data.

Machine Learning (ML)

When to use Multilayer Perceptrons (MLP)?

Multilayer Perceptrons (MLPs) are the buiding blocks of neural network. They are comprised of one or more layers of neurons. MLPs are used for classification prediction problems, regression prediction problems and tabular datasets.

Software Engineering

Multirepo vs Monorepo: Which one to choose?

Choosing between Multirepo and Monorepo depends on various factors such as Tooling infrastructure and investment, build challenges, source code version control, large scale refactoring and engineering culture. Monorepo is used by Google and Facebook while Multirepo is used by Amazon and Netfilx

Software Engineering

Multirepo explained

Multirepo is a software architecture which means a separate repository for each project. Each project is an independent working unit. It has minimum tooling investment and is in use at companies such as Amazon and Netflix.

monorepo

Monorepo explained

Monorepo is a repository that contains more than one logical project. It comes with several advantages such as collaboration among various teams and at the same time, disadvantages such as massive tooling and infrastructure investment Monorepo architecture is followed by Google, Facebook and Twitter

Machine Learning (ML)

When to use Recurrent Neural Networks (RNN)?

Recurrent Neural Networks (RNNs) are designed to work with sequence prediction problems. RNNs can be used on Text data, Speech data, Classification prediction problems, Regression prediction problems and Generative models. Sequence prediction problems come in many forms.

Machine Learning (ML)

When to use Convolutional Neural Networks (CNN)?

Use CNN for data with a spatial relationship. Convolutional Neural Networks (CNNs) are designed to map image data to an output variable. They have proven so effective that they are the ready to use method for any type of prediction problem involving image data as an input