Lecture 21 - HPC

0.0(0)
studied byStudied by 0 people
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/31

encourage image

There's no tags or description

Looks like no tags are added yet.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

32 Terms

1
New cards

What is a computer?

An electronic device that stores and processes digital info and follows programmed logic and instruction set.

2
New cards

What are the 5 components of a computer system?

CPU, memory, control unit, and I/O units.

3
New cards

What is the CPU (Central Processing Unit)?

Circuitry that carries out instructions of a computer program by using arithmetic, logical, and controlling I/O operations.

4
New cards

What controls the speed of the CPU?

The system clock.

5
New cards

What does the system clock do?

Generates electronic pulses at regular intervals to coordinate CPU activities, ensuring even the slowest operation can finish.

6
New cards

How is the performance of a CPU measured?

In clock speed (in GHz) and FLOPS (floating point operations per second).

7
New cards

What does FLOPS tell us?

How fast a single computation can be done by the CPU.

8
New cards

What does clock speed indicate?

How many instructions are performed per second by the CPU.

9
New cards

What is the thermal brick wall?

The clock-rate reached an upper limit because more cooling power is required.

10
New cards

Why do we need more cooling power for CPUs?

Higher CPU speed results in a higher clock rate, leading to faster electric current that generates more heat and lowers signal-to-noise ratio.

11
New cards

What is the current thermal brick wall for CPU speeds?

Hard to get above 4.0 GHz.

12
New cards

What are memory modules?

Any physical device capable of storing information for immediate use.

13
New cards

What is parallel computing?

Computation where many calculations are carried out simultaneously by breaking a big problem into smaller ones and solving them concurrently.

14
New cards

What is computational gain?

Serial time divided by parallel time.

15
New cards

What is parallel efficiency?

Computational gain divided by the number of processors.

16
New cards

What is serial computing?

A single processor running the computer program.

17
New cards

What is shared memory parallelism (OpenMP)?

Multiple processors or threads working on different parts of the program, sharing memory but sometimes competing for resources.

18
New cards

What is distributed parallelism (Message Passing Interface)?

Multiple processors working separately without having to contend with resources.

19
New cards

What is a supercomputer?

A computer cluster made of nodes (connected computers) that work together as a single system.

20
New cards

What is an OS?

Operating system is software closest to the computer hardware that manages all hardware and software, abstracting hardware from user programs.

21
New cards

What is SSH?

Secure Shell is a cryptographic network protocol for operating network services over an unsecured network, using encryption to secure connections between client and server.

22
New cards

What are the statistics of the Midway3 compute nodes?

192GB of memory, 100Gbps network, 24 cores, 3 GHz base frequency.

23
New cards

What is the storage of Midway3?

2.2 PB.

24
New cards

What is the shell?

Text-based terminal that takes in keyboard input and outputs text.

25
New cards

What is SLURM?

Workload manager that schedules jobs and manages resources between multiple users.

26
New cards

What are the 3 V's of big data?

Volume of data, Velocity of data transfers, Variety of types of data.

27
New cards

What is structured data?

Data that is formatted to be easily used with other databases.

28
New cards

What are examples of structured data?

Databases, JSON, HTML, CSV, etc.

29
New cards

What is unstructured data?

Data that is not structured, such as web pages, documents, pdfs, emails, media, sensor data.

30
New cards

What are the challenges of unstructured data?

They must be cleaned, outliers removed, pre-processed, edited, scraped, integrated, prepared, and analyzed.

31
New cards

What are the 4 rules of data trust?

Not all data is trustworthy, not all trustworthy data is correct, untrustworthy data is not always incorrect, even if data is correct, the answer may be wrong.

32
New cards

Why is data visualization necessary?

It helps create meaningful interpretations of results.