liveBook · Manning

Simultaneous multithreading

Also called Hyper-Threading, hyperthreading, SMT

Simultaneous multithreading is a processor technology that enables multiple hardware threads to share and concurrently use one physical core’s execution resources.

How 3 Manning books explain it

The Coder Cafe

SMT, or simultaneous multithreading, is a processor technique that allows multiple hardware threads to share a physical core's execution resources.

How it is used
Explains how processors improve physical-core utilization and why visible hardware-thread counts do not equal physical-core counts.
How deep it goes
Explains architectural state, instruction queues, scheduling onto shared execution units, and performance tradeoffs without presenting a hardware implementation.
Examples
Compares four-core CPUs with four or eight visible threads and uses the i7-7700K to illustrate non-linear scaling.
What you can do after
Evaluate/understand SMT benefits by distinguishing hardware threads from cores and benchmarking workloads rather than assuming linear speedups.

Read how The Coder Cafe explains simultaneous multithreading →

Latency

Hyper-Threading is a processor technology that allows one physical core to execute multiple hardware threads concurrently.

How it is used
Frames SMT as both a way to increase concurrent execution and a potential source of latency variance.
How deep it goes
Explains duplicated architectural state, shared resources, and contention, but does not provide a configuration or measurement procedure.
Examples
Illustrates contention with a resource-heavy thread stalling another thread on the same physical core.
What you can do after
Evaluate SMT's throughput–latency trade-off and consider disabling it for ultra-low-latency applications.

Read how Latency explains simultaneous multithreading →

Quantization and Fast Inference

SMT is a processor capability that allows multiple hardware threads to share the resources of one physical core.

How it is used
Guides CPU inference thread-count decisions by testing whether SMT siblings improve decode throughput on the deployment host.
How deep it goes
Explains sibling-thread cache contention and measured throughput changes, without detailing processor implementation.
Examples
Measures llama.cpp decode from 11 to 22 threads, including a 12-thread dip on an 11-core VM.
What you can do after
Operate/configure inference thread counts by benchmarking SMT siblings instead of assuming physical-core count is optimal.

Read how Quantization and Fast Inference explains simultaneous multithreading →

Also covered in

What to know first

Related concepts

Questions about simultaneous multithreading

What is simultaneous multithreading?

Simultaneous multithreading is a processor technology that enables multiple hardware threads to share and concurrently use one physical core’s execution resources.

Which Manning books explain simultaneous multithreading?

The Coder Cafe, Latency and Quantization and Fast Inference.

What should I know before learning simultaneous multithreading?

Manning books that teach simultaneous multithreading usually introduce compute unit, context switching, CPU and logical CPU core first.

What is the difference between simultaneous multithreading and efficiency core?

Simultaneous multithreading is a processor technology that enables multiple hardware threads to share and concurrently use one physical core’s execution resources. An efficiency core is a processor core designed to maximize computational work per unit of power, typically at lower power and performance levels than a performance core.

What is simultaneous multithreading part of?

Simultaneous multithreading is part of hardware. Hardware is the physical equipment and components of a computing system, including processors, memory, storage, and peripherals.

What is simultaneous multithreading also called?

Simultaneous multithreading is also called Hyper-Threading, hyperthreading or SMT.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.