Hyper-Threading Technology

From Conservapedia
Jump to: navigation, search

Hyper-Threading Technology is a technical method developed by Intel which increases the flow of data to a single processor core. It was first released in 2002 on Xeon and Pentium 4 processors. Hyper-threading enables a more efficient use of a processor, since it is less likely to stall (stop and wait) when data is needed after a state change or for other reasons. Some software (such as the Windows task manager) sees a hyper-threading device as two cores. However, this is only a slightly better utilization of a single core. While a statistically significant improvement, Hyper-Threading does not run with nearly as much efficiency as a pair of cores. Nonetheless, hyper-threading technology can be applied to multi-core systems, enabling slightly better performance for all cores.[1][2][3]

Security risk

In 2018, it was revealed to the public that hyper-threading technology creates an opportunity for a side-channel attack. Because the processor changes state to cycle between threads, there is a possibility that a malicious process running on the processor could access information which it should not be able to. Although such an attack has not yet been seen in the wild, this potentially enables the loss of vital data such as the device's RSA private key. For this reason, it will probably be phased out, unless it can be altered to maintain process barriers.[4][5][6][7]

References