Hyper-Threading technology

Hyper-Threading technology, or HT technology, allows each microprocessor core to execute two software threads at the same time. Simultaneous execution of two threads by the same core is accomplished by:

• Adding the second set of a few internal core components, such as interrupt controller, general, control and special registers. The second set of registers allows the CPU to keep the state of both cores, and effortlessly switch between them by switching the register set.
• Modifying other core units to switch between two threads. Some internal units automatically switch from one thread to another, and back, while others switch to another thread only when the current thread is stalled, i.e. is waiting for data.
• Limiting maximum size of internal buffers used by each thread. This is necessary to ensure that both threads get sufficient resources required for thread execution.

HT technology results in better utilization of core resources and improved performance. Using this technology, many core resources can be shared between threads without negative impact, although sharing of some resources, like level 1 cache, may lower process performance under certain conditions. Overall, for large number of multi-threaded applications the Hyper-Threading can provide noticeable performance boost, up to 20%, or up to 50% in rare cases. On operating system level, a single-core CPU with HyperThreading technology will be reported as two logical processors, dual-core CPU with HT is reported as four logical processors, and so on. HyperThreading Technology does not require any special drivers to be installed, but does require BIOS to support this feature.

HyperThreading was implemented in the following processor families:

Desktop/Server families	Mobile families
	Atom
Core i3	Core i3 Mobile
Core i5	Core i5 Mobile
Core i7	Core i7 Mobile
Core i7 Extreme Edition	Core i7 Mobile Extreme Edition
Pentium 4	Mobile Pentium 4
Pentium 4 Extreme Edition
Pentium Extreme Edition
Xeon

Hyper-Threading technology is one of a few methods of improving CPU performance in multi-threaded applications and/or multiple single-threaded tasks. Other methods are multi-core and multi-processing. The table below summarizes difference between Hyper-Threading, multi-core and multi-processing methods.

Method	Core resources	Other package resources	Motherboard resources	Performance in multi-threaded tasks
HyperThreading Technology	Most resources shared	Shared	Shared	Minor improvement
Multi-core (N cores)	Not shared	Shared	Shared	Up to N times
Multi-processing (N CPUs)	Not shared	Not shared	Shared	Up to N times