Intel Corporation

Choupin Huang

Stefan Rusu

Nicholas B Peterson

Vijaya K Boddu

Drin-Guang W Chen

Ramesh K R Velugoti

Charles T Ballou

Inho Kim

Soren Sharifi

Integrated clock differential buffering. A first phase locked loop (PLL) circuit having a first clocking ratio is coupled to receive an input differential clock signal. The first PLL circuit generates a first reference clock signal. A second PLL circuit having a second clocking ratio is coupled to receive the input differential clock signal. The second PLL circuit to generate a second reference clock signal. A first set of clock signal output buffers are coupled to receive the first reference clock signal and to provide a first differential reference clock signal corresponding to the first reference clock signal. A second set of clock signal output buffers is coupled to receive the second reference clock signal and to provide a second differential reference clock signal corresponding to the second reference clock signal. The first PLL circuit, the second PLL circuit, the first set of output buffers and the second set of output buffers reside within an integrated circuit package also having a die to receive at least the first differential reference clock signal.

Integrated clock differential buffering

Scalable, common reference-clocking architecture and method for blade and rack servers. A common reference clock source is configured to provide synchronized clock input signals to a plurality of blades in a blade server or servers in a rack server. The reference clock signals are then used for clock operations related to serial interconnect links between blades and/or servers, such as QuickPath Interconnect (QPI) links or PCIe links. The serial interconnect links may be routed via electrical or optical cables between blades or servers. The common reference clock input and inter-blade or inter-server interconnect scheme is scalable, such that the plurality of blades or servers can be linked together in communication. Moreover, when QPI links are used, coherent memory transactions across blades or servers are provided, enabling fine grained parallelism to be used for parallel processing applications.

Scalable, common reference-clocking architecture using a separate, single clock source for blade and rack servers

A method, apparatus, article of manufacture, and system, the method including, in some embodiments, providing a differential clock ganging structure to receive complementary differential clock signals, the differential clock ganging structure outputting clock ganging output signals, providing a source termination structure for each of the clock ganging output signals, and providing an inductance and capacitance compensation structure to receive an output of the source termination structure and to connect to a terminal interconnect for at least one of the clock ganging output signals.

Method and system for clock routing and compensation

Methods and arrangements to gang differential clock signals to attenuate pin-to-pin output skew for a clock driver are disclosed. Embodiments may comprise a pattern of conductors to interconnect output pins for differential clock signals with termination resistors. The pattern of conductors comprises a group of conductors for a positive clock (p-clock) signal and a group of conductors for a negative clock (n-clock) signal. The conductors for the p-clock signal intersect at a gang point between the output pins and pads for the termination resistors. Similarly, the conductors for the n-clock signals intersect at a gang point between the pins and the pads. In many embodiments, the distance between the pins and pads may be approximately 120 mils. In further embodiments, the distance may be longer or shorter than 120 mils. Other embodiments are disclosed and claimed.

Growing community of inventors

San Jose, CA, United States of America

Choupin Huang