Clock skew

Last updated August 18, 2023

Clock skew (sometimes called timing skew) is a phenomenon in synchronous digital circuit systems (such as computer systems) in which the same sourced clock signal arrives at different components at different times due to gate or, in more advanced semiconductor technology, wire signal propagation delay. The instantaneous difference between the readings of any two clocks is called their skew.

The operation of most digital circuits is synchronized by a periodic signal known as a "clock" that dictates the sequence and pacing of the devices on the circuit. This clock is distributed from a single source to all the memory elements of the circuit, which for example could be registers or flip-flops. In a circuit using edge-triggered registers, when the clock edge or tick arrives at a register, the register transfers the register input to the register output, and these new output values flow through combinational logic to provide the values at register inputs for the next clock tick.

Ideally, the input to each memory element reaches its final value in time for the next clock tick so that the behavior of the whole circuit can be predicted exactly. The maximum speed at which a system can run must account for the variance that occurs between the various elements of a circuit due to differences in physical composition, temperature, and path length.

In a synchronous circuit, two registers, or flip-flops, are said to be "sequentially adjacent" if a logic path connects them. Given two sequentially adjacent registers R_i and R_j with clock arrival times at the source and destination register clock pins equal to T_Ci and T_Cj respectively, clock skew can be defined as: $T skew i, j = T C i - T C j$ .

In circuit design

Clock skew can be caused by many different things, such as wire-interconnect length, temperature variations, variation in intermediate devices, capacitive coupling, material imperfections, and differences in input capacitance on the clock inputs of devices using the clock. As the clock rate of a circuit increases, timing becomes more critical and less variation can be tolerated if the circuit is to function properly.

There are two types of clock skew: negative skew and positive skew. Positive skew occurs when the receiving register receives the clock tick later than the transmitting register. Negative skew is the opposite: the transmitting register gets the clock tick later than the receiving register. Zero clock skew refers to the arrival of the clock tick simultaneously at transmitting and receiving register.^[1]

Harmful skew

There are two types of violation that can be caused by clock skew. One problem is caused when the clock reaches the first register and the clock signal towards the second register travels slower than output of the first register into the second register - the output of the first register reaches the second register input faster and therefore is clocked replacing the initial data on the second register, or maybe destroying the integrity of the latched data. This is called a hold violation because the previous data is not held long enough at the destination flip-flop to be properly clocked through. Another problem is caused if the destination flip-flop receives the clock tick earlier than the source flip-flop - the data signal has that much less time to reach the destination flip-flop before the next clock tick. If it fails to do so, a setup violation occurs, so-called because the new data was not set up and stable before the next clock tick arrived. A hold violation is more serious than a setup violation because it cannot be fixed by increasing the clock period. Positive skew and negative skew cannot negatively impact setup and hold timing constraints respectively (see inequalities below).

Beneficial skew

Where a signal broadly clocks a circuit, the signals/state-transitions it initiates must be stabilized before it signals another set of state transitions -- and that limits the clock's upper frequency. Skew thus decreases the clock frequency at which the circuit will operate correctly. For each source register and destination register connected by a path, the following setup and hold inequalities must be obeyed:

$T\geq {\text{reg}}+{\text{path}}_{\text{max}}+J+S-(s_{d}-s_{s})$
$(s_{d}-s_{s})\leq {\text{reg}}+{\text{path}}_{\text{min}}-J-H$

where

T is the clock period,
reg is the source register's clock to Q delay,
${\text{path}}_{\text{max}}$ is the path with the longest delay from source to destination,
J is an upper bound on jitter,
S is the setup time of the destination register
$(s_{d}-s_{s})$ represents the clock skew from the source to the destination registers,
${\text{path}}_{\text{min}}$ is the path with the shortest delay from source to destination,
H is the hold time of the destination register,
$s_{d}$ is the clock skew to the destination register, and
$s_{s}$ is the clock skew to the source register.

Positive clock skews are good for fixing setup violations, but can cause hold violations. Negative clock skew can guard against a hold violation, but can cause a setup violation.

In the above inequalities, a single parameter, J, is used to account for jitter. This parameter must be an upper bound for the difference in jitter over all source register/destination register pairs. However, if the structure of the clock distribution network is known, different source register/destination register pairs may have different jitter parameters, and a different jitter value may be used for the hold constraint in contrast to the value for the setup constraint. For example, if the source register and destination register receive their clock signals from a common nearby clock buffer, the jitter bound for that hold constraint can be very small, since any variation in that clock signal will affect the two registers equally. For the same example, the jitter bound for the setup constraint must be larger than for the hold constraint, because jitter can vary from clock tick to clock tick. If the source register receives its clock signal from a leaf buffer of the clock distribution network that is far removed from the leaf buffer feeding the destination register, then the jitter bound will have to be larger to account for the different clock paths to the two registers, which may have different noise sources coupling into them.

Figures 1 and 2 illustrate a situation where intentional clock skew can benefit a synchronous circuit.^[2] In the zero-skew circuit of Figure 1, a long path goes from flip-flop FF1 to flip-flop FF2, and a short path, such as a shift-register path, from FF2 to FF3. The FF2 -> FF3 path is dangerously close to having a hold violation: If even a small amount of extra clock delay occurs at FF3, this could destroy the data at the D input of FF3 before the clock arrives to clock it through to FF3's Q output. This could happen even if FF2 and FF3 were physically close to each other, if their clock inputs happened to come from different leaf buffers of a clock distribution network.

Figure 2 shows how the problem can be fixed with intentional clock skew. A small amount of extra delay is interposed before FF2's clock input, which then safely positions the FF2 -> FF3 path away from its hold violation. As an added benefit, this same extra clock delay relaxes the setup constraint for the FF1 -> FF2 path. The FF1 -> FF2 path can operate correctly at a clock period that is less than what is required for the zero clock skew case, by an amount equal to the delay of the added clock delay buffer.

A common misconception about intentional clock skew is that it is necessarily more dangerous than zero clock skew, or that it requires more precise control of delays in the clock distribution network. However it is the zero skew circuit of Figure 1 that is closer to malfunctioning - a small amount of positive clock skew for the FF2 -> FF3 pair will cause a hold violation, whereas the intentional skew circuit of Figure 2 is more tolerant of unintended delay variations in clock distribution.

Optimal skew

If the clock arrival times at individual registers are viewed as variables to be adjusted in order to minimize the clock period while satisfying the setup and hold inequalities for all of the paths through the circuit, then the result is a Linear Programming problem.^[3] In this linear program, zero clock skew is merely a feasible point - the solution to the linear program generally gives a clock period that is less than what is achieved by zero skew. In addition, safety margins greater than or equal to the zero skew case can be guaranteed by setting setup and hold times and jitter bound appropriately in the linear program.

Due to the simple form of this linear program, an easily programmed algorithm is available for arriving at a solution.^[2] Most CAD systems for VLSI and FPGA design contain facilities for optimizing clock skews.

Confusion between clock skew and clock jitter

In addition to clock skew due to static differences in the clock latency from the clock source to each clocked register, no clock signal is perfectly periodic, so that the clock period or clock cycle time varies even at a single component, and this variation is known as clock jitter. At a particular point in a clock distribution network, jitter is the only contributor to the clock timing uncertainty.

As an approximation, it is often useful to discuss the total clock timing uncertainty between two registers as the sum of spatial clock skew (the spatial differences in clock latency from the clock source), and clock jitter (meaning the non-periodicity of the clock at a particular point in the network). Unfortunately, spatial clock skew varies in time from one cycle to the next due to local time-dependent variations in the power supply, local temperature, and noise coupling to other signals.

Thus, in the usual case of sending and receiving registers at different locations, there is no clear way to separate the total clock timing uncertainty into spatial skew and jitter. Thus some authors use the term clock skew to describe the sum of spatial clock skew and clock jitter. This of course means that the clock skew between two points varies from cycle to cycle, which is a complexity that is rarely mentioned. Many other authors use the term clock skew only for the spatial variation of clock times, and use the term clock jitter to represent the rest of the total clock timing uncertainty. This of course means that the clock jitter must be different at each component, which again is rarely discussed.

Fortunately, in many cases, spatial clock skew remains fairly constant from cycle to cycle, so that the rest of the total clock timing uncertainty can be well approximated by a single common clock jitter value.

On a network

On a network such as the internet, clock skew describes the difference in frequency (first derivative of offset with time) of different clocks within the network.^[4] Network operations that require timestamps which are comparable across hosts can be affected by clock skew. A number of protocols (e.g. Network Time Protocol) have been designed to reduce clock skew, and produce more stable functions. Some applications (such as game servers) may also use their own synchronization mechanism to avoid reliability problems due to clock skew.

Interfaces

Clock skew is the reason why at fast speeds or long distances, serial interfaces (e.g. Serial Attached SCSI or USB) are preferred over parallel interfaces (e.g. parallel SCSI).^{[ citation needed ]}

Related Research Articles

In digital logic and computing, a counter is a device which stores the number of times a particular event or process has occurred, often in relationship to a clock. The most common type is a sequential digital logic circuit with an input line called the clock and multiple output lines. The values on the output lines represent a number in the binary or BCD number system. Each pulse applied to the clock input increments or decrements the number in the counter.

In electronics and telecommunications, jitter is the deviation from true periodicity of a presumably periodic signal, often in relation to a reference clock signal. In clock recovery applications it is called timing jitter. Jitter is a significant, and usually undesired, factor in the design of almost all communications links.

A phase-locked loop or phase lock loop (PLL) is a control system that generates an output signal whose phase is related to the phase of an input signal. There are several different types; the simplest is an electronic circuit consisting of a variable frequency oscillator and a phase detector in a feedback loop. The oscillator's frequency and phase are controlled proportionally by an applied voltage, hence the term voltage-controlled oscillator (VCO). The oscillator generates a periodic signal of a specific frequency, and the phase detector compares the phase of that signal with the phase of the input periodic signal, to adjust the oscillator to keep the phases matched.

A shift register is a type of digital circuit using a cascade of flip-flops where the output of one flip-flop is connected to the input of the next. They share a single clock signal, which causes the data stored in the system to shift from one location to the next. By connecting the last flip-flop back to the first, the data can cycle within the shifters for extended periods, and in this configuration they were used as computer memory, displacing delay-line memory systems in the late 1960s and early 1970s.

In automata theory, sequential logic is a type of logic circuit whose output depends on the present value of its input signals and on the sequence of past inputs, the input history. This is in contrast to combinational logic, whose output is a function of only the present input. That is, sequential logic has state (memory) while combinational logic does not.

In electronics and especially synchronous digital circuits, a clock signal is an electronic logic signal which oscillates between a high and a low state at a constant frequency and is used like a metronome to synchronize actions of digital circuits. In a synchronous logic circuit, the most common type of digital circuit, the clock signal is applied to all storage devices, flip-flops and latches, and causes them all to change state simultaneously, preventing race conditions.

In electronic instrumentation and signal processing, a time-to-digital converter (TDC) is a device for recognizing events and providing a digital representation of the time they occurred. For example, a TDC might output the time of arrival for each incoming pulse. Some applications wish to measure the time interval between two events rather than some notion of an absolute time.

Source-Synchronous clocking refers to a technique used for timing symbols on a digital interface. Specifically, it refers to the technique of having the transmitting device send a clock signal along with the data signals. The timing of the unidirectional data signals is referenced to the clock sourced by the same device that generates those signals, and not to a global clock. Compared to other digital clocking topologies like system-synchronous clocks, where a global clock source is fed to all devices in the system, a source-synchronous clock topology can attain far higher speeds.

10 Gigabit Attachment Unit Interface is a standard for extending the XGMII between the MAC and PHY layer of 10 Gigabit Ethernet (10GbE) defined in Clause 47 of the IEEE 802.3 standard. The name is a concatenation of the Roman numeral X, meaning ten, and the initials of "Attachment Unit Interface".

In digital circuits, the contamination delay is the minimum amount of time from when an input changes until any output starts to change its value. This change in value does not imply that the value has reached a stable condition. The contamination delay only specifies that the output rises to 50% of the voltage level for a logic high. The circuit is guaranteed not to show any output change in response to an input change before t_cd time units have passed. The determination of the contamination delay of a combined circuit requires identifying the shortest path of contamination delays from input to output and by adding each t_cd time along this path.

In digital electronics, a synchronous circuit is a digital circuit in which the changes in the state of memory elements are synchronized by a clock signal. In a sequential digital logic circuit, data are stored in memory devices called flip-flops or latches. The output of a flip-flop is constant until a pulse is applied to its "clock" input, upon which the input of the flip-flop is latched into its output. In a synchronous logic circuit, an electronic oscillator called the clock generates a string (sequence) of pulses, the "clock signal". This clock signal is applied to every storage element, so in an ideal synchronous circuit, every change in the logical levels of its storage components is simultaneous. Ideally, the input to each storage element has reached its final value before the next clock occurs, so the behaviour of the whole circuit can be predicted exactly. Practically, some delay is required for each logical operation, resulting in a maximum speed limitations at which each synchronous system can run.

In electronics, metastability is the ability of a digital electronic system to persist for an unbounded time in an unstable equilibrium or metastable state. In digital logic circuits, a digital signal is required to be within certain voltage or current limits to represent a '0' or '1' logic level for correct circuit operation; if the signal is within a forbidden intermediate range it may cause faulty behavior in logic gates the signal is applied to. In metastable states, the circuit may be unable to settle into a stable '0' or '1' logic level within the time required for proper circuit operation. As a result, the circuit can act in unpredictable ways, and may lead to a system failure, sometimes referred to as a "glitch". Metastability is an instance of the Buridan's ass paradox.

Static timing analysis (STA) is a simulation method of computing the expected timing of a synchronous digital circuit without requiring a simulation of the full circuit.

In computer architecture, clock gating is a popular power management technique used in many synchronous circuits for reducing dynamic power dissipation, by removing the clock signal when the circuit is not in use or ignores clock signal. Clock gating saves power by pruning the clock tree, at the cost of adding more logic to a circuit. Pruning the clock disables portions of the circuitry so that the flip-flops in them do not have to switch states. Switching states consumes power. When not being switched, the switching power consumption goes to zero, and only leakage currents are incurred.

In digital electronic design a clock domain crossing (CDC), or simply clock crossing, is the traversal of a signal in a synchronous digital circuit from one clock domain into another. If a signal does not assert long enough and is not registered, it may appear asynchronous on the incoming clock boundary.

The Timing closure in VLSI design and electronics engineering is the process by which a logic design of a clocked synchronous circuit consisting of primitive elements such as combinatorial logic gates and sequential logic gates is modified to meet its timing requirements. Unlike in a computer program where there is no explicit delay to perform a calculation, logic circuits have intrinsic and well defined delays to propagate inputs to outputs.

Retiming is the technique of moving the structural location of latches or registers in a digital circuit to improve its performance, area, and/or power characteristics in such a way that preserves its functional behavior at its outputs. Retiming was first described by Charles E. Leiserson and James B. Saxe in 1983.

<span class="mw-page-title-main">Physical design (electronics)</span>

In integrated circuit design, physical design is a step in the standard design cycle which follows after the circuit design. At this step, circuit representations of the components of the design are converted into geometric representations of shapes which, when manufactured in the corresponding layers of materials, will ensure the required functioning of the components. This geometric representation is called integrated circuit layout. This step is usually split into several sub-steps, which include both design and verification and validation of the layout.

In electronics, flip-flops and latches are circuits that have two stable states that can store state information – a bistable multivibrator. The circuit can be made to change state by signals applied to one or more control inputs and will output its state. It is the basic storage element in sequential logic. Flip-flops and latches are fundamental building blocks of digital electronics systems used in computers, communications, and many other types of systems.

Low Power flip-flops are flip-flops that are designed for low-power electronics, such as smartphones and notebooks. A flip-flop, or latch, is a circuit that has two stable states and can be used to store state information.

References

Friedman, Eby G. (1995). Clock Distribution Networks in VLSI Circuits and Systems . IEEE Press. ISBN 978-0-7803-1058-2.
Friedman, Eby G. (May 2001). "Clock Distribution Networks in Synchronous Digital Integrated Circuits" (PDF). Proceedings of the IEEE . 89 (5): 665–692. CiteSeerX 10.1.1.7.7824 . doi:10.1109/5.929649. Archived from the original (PDF) on 2015-06-01. Retrieved 2017-01-09.
Tam, S.; Limaye, D.L.; Desai, U.N. (April 2004). "Clock Generation and Distribution for the 130-nm Itanium 2 Processor with 6-MB On-Die L3 Cache". IEEE Journal of Solid-State Circuits . 39 (4). doi:10.1109/JSSC.2004.825121. S2CID 31388328.

↑ Friedman, E. G. Clock distribution design in VLSI circuits-an overview, 1993 IEEE International Symposium on Circuits and Systems (pp. 1475-1478). IEEE, 1993.
1 2 Maheshwari, N., and Sapatnekar, S.S., Timing Analysis and Optimization of Sequential Circuits, Kluwer, 1999.
↑ Fishburn, J.P. (July 1990). "Clock skew optimization" (PDF). IEEE Transactions on Computers. 39 (7): 945–951. doi:10.1109/12.55696.
↑ Mills, D. (1992). "Network Time Protocol (Version 3) Specification, Implementation and Analysis". tools.ietf.org. doi: 10.17487/RFC1305 . Retrieved 2017-10-30.

This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.

[friedman_1993-1] Friedman, E. G. Clock distribution design in VLSI circuits-an overview, 1993 IEEE International Symposium on Circuits and Systems (pp. 1475-1478). IEEE, 1993.

[maheshwari_sapatnekar_1999-2] 1 2 Maheshwari, N., and Sapatnekar, S.S., Timing Analysis and Optimization of Sequential Circuits, Kluwer, 1999.

[3] Fishburn, J.P. (July 1990). "Clock skew optimization" (PDF). IEEE Transactions on Computers. 39 (7): 945–951. doi:10.1109/12.55696.

[4] Mills, D. (1992). "Network Time Protocol (Version 3) Specification, Implementation and Analysis". tools.ietf.org. doi: 10.17487/RFC1305 . Retrieved 2017-10-30.

[1]

[2]

[3]

[4]