Data locality

Q: How deeply does Fast Python: High performance techniques for large datasets cover data locality?

Explains memory-hierarchy costs, cache proximity, and virtualization caveats, but gives no optimization procedure.

As explained in Fast Python: High performance techniques for large datasets

Data locality describes the relationship between where data is located and where or how it is accessed or processed, including proximity, reuse, and location constraints.

Data locality concerns how data placement affects processing cost. Keeping needed information close to the CPU can reduce waiting for slower memory; storage and network access costs also vary widely. In virtual machines, cache conditions may be unpredictable, limiting the reliability of cache-aware optimizations. section 1.2.1 section 1.2.3 section 6.1.3

Compression and locality

Processing compressed data can be faster than processing raw data when the compressed block fits in CPU cache: decompression can use cycles otherwise spent waiting for RAM. A NumPy array representation choice can also affect cache locality and processing speed. section 1.2.1

Table 1.1 Memory hierarchy with sizes and access times for a hypothetical, but realistic modern desktop (view table figure)

Type	Size	Access time
CPU
L1 cache	256 KB	2 ns
L2 cache	1 MB	5 ns
L3 cache	6 MB	30 ns
RAM
DIMM	8 GB	100 ns
Secondary storage
SSD	256 GB	50 µs
HDD	2 TB	5 ms
Tertiary storage
NAS - Network Access Server	100 TB	Network dependent
Cloud proprietary	1 PB	Provider dependent

Compare the capacities and access times shown across CPU caches, RAM, disks, and network storage. section 1.2.1

How Fast Python: High performance techniques for large datasets covers it

Cover of Fast Python: High performance techniques for large datasets

liveBook manning.com

Read Fast Python: High performance techniques for large datasets on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Deep Learning with PyTorch, Second EditionThe tendency for nearby memory accesses to hit nearby stored data, which improves performance.
liveBook manning.com

Idiomatic RustData locality is the tendency for a program to access data near recently accessed or related data in memory.
liveBook manning.com

100 Go Mistakes and How to Avoid ThemData locality is the tendency for related or recently accessed data to reside near one another in memory or storage.

Compare how 4 books explain data locality →

Related concepts

linked listcontrasts with · An ordered sequence of elements stored in separate nodes connected by links or references rather than in one contiguous storage region.
spatial localityincludes · Spatial locality is the tendency for memory locations near a recently accessed location to be accessed soon afterward.
data-oriented programmingproduced by · Data-oriented programming is an approach to structuring software around data representations, access patterns, and operations suited to the data's characteristics and intended use.
temporal localityincludes · Temporal locality is the tendency for recently accessed data or memory locations to be accessed again soon.

This explainer was generated by Manning's knowledge graph from the text of Fast Python: High performance techniques for large datasets. The section links point to the parts of the book it draws on.