levo
Log inStart

Mobile Technology, Data Mining and Cloud Computing

Processing and analysing large volumes of data

  • 3 things to know
  • A2 Unit 1

Overview

How distributed processing speeds up analysis, map and reduce approach to processing large data sets and in-memory processing.

What you need to know

Answer each one before you reveal it.

Your recallNot started
Key term

Why are the results of mining presented visually?

Key term

Why can a single powerful server not analyse the largest data sets?

Process

Explain how distributed processing speeds up analysis.

Questions from this subtopic

Each one with the model answer underneath.

  1. 1

    Why can a single powerful server not analyse the largest data sets?

    Model answer

    One server cannot hold petabytes of data or process it within a useful time, so very large analyses are divided across many machines that each work on their own portion simultaneously.

  2. 2

    Explain how distributed processing speeds up analysis.

    Model answer

    Distributed processing splits the data across a cluster of servers, has each server process the part stored locally and then combines the partial results, so the whole job finishes in a fraction of the time.

  3. 3

    Describe the map and reduce approach to processing large data sets.

    Model answer

    A map and reduce approach applies the same operation to every record in parallel to produce intermediate results, then combines those intermediate results by key to produce the final totals.

  4. 4

    Distinguish scaling up from scaling out.

    Model answer

    Scaling up replaces a server with a more powerful one, while scaling out adds more ordinary servers to a cluster, and very large analyses scale out because that approach has no practical ceiling.

  5. 5

    What is in-memory processing, and what does it cost?

    Model answer

    In-memory processing holds the working data in RAM rather than reading it repeatedly from disk, which speeds analysis enormously but requires far more memory and therefore more expensive machines.

  6. 6

    Describe how a decision tree classifies a record.

    Model answer

    A decision tree is built by repeatedly splitting the training records on the field that best separates the outcomes, and a new record is classified by following the branch that matches it at each test until a leaf is reached.

See the rest and revise

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.

1Closed book
2Open book
3Closed book

The 3-step method, run for you

Answer cold, learn it, then prove it stuck. Levo tracks the jump between your first attempt and your last.

Systems Analysis and Development Approaches
Design, Development and Testing
Implementation and System Security
Programming

Cram a whole paper in one sitting

Sort every topic into what you know and what you do not, then drill only the weak ones.

A check requires a value.
presencerangeformat

Blank Blitz makes facts automatic

Rapid timed gap-fills on the terms and numbers you need to recall without thinking.

ANormal data is 12, which should be accepted.
BBoundary data is 1 and 50, which should be accepted.
C0 and 51, just outside the limits, should be rejected.

Build It gets the exam shape right

Assemble a full-mark answer from its marking points, in the order an examiner wants them.

One byte is made up of 8 bits.
TrueFalse

Test It catches misconceptions

Spot the fault in a statement. The traps are the mistakes students actually make.

721941163128510

Brain Breaks between sessions

Short focus games to reset attention, so the next set lands better than the last.

Every question in Mobile Technology, Data Mining and Cloud Computing, on a schedule.

Attempt a question from memory, write the correct version, then attempt it again. Levo records the jump between your first attempt and your last, and brings each question back on the day you are due to forget it.

Start revising, £5 a monthSee how the method works

Or £46 a year, about £3.83 a month. Cancel anytime.