A Review - Synchronization Approaches to Digital systems

Sukanya. K Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 6, Issue 3, (Part - 1) March 2016, pp.01-10
www.ijera.com 1|P a g e
A Review - Synchronization Approaches to Digital systems
Sukanya. K*
, Dr. G. Laxminarayana**
*(Department of E.C.E, TKR College of Engineering and Technology, Ranga Reddy, Telangana-500097)
**( Department of E.C.E, CVSR College of Engineering, Ranga Reddy, Telangana-501301)
ABSTRACT
Synchronization is a prime requirement in the process of Digital systems. Wherein new devices are upcoming
towards providing higher service level, advanced distributed systems are been integrated onto a single platform
for higher service provision. However with the integration of large processing units, the distributed processing
needs a high level synchronization with minimum processing overhead. The issue of synchronization was
processed by various approaches. This paper outlines a brief review on the developments made in the field of
synchronization approach to digital system, under distributed mode operation.
Keywords: Digital system, Distributed system, Processing overhead, Service level, Synchronization
I. INTRODUCTION
The demand of service over different
applications has lead to the development of new
devices with high speed processing. The systems are
being designed with distributed processing unit,
operating in parallel to provide faster computation.
However the upcoming services are demanding for
high processing efficiency and parallel processing of
multiple operations. To cope up with the current
demand and future demands, new processing
architectures with multiple cores, distributed
processing or pipelining process are emerging. With
such high level processing, issue of synchronization
among various sub-components exist. Hence, an
architectural improvement would not satisfy the
current and up-coming processing demands. To
provide synchronization in an efficient manner, new
synchronization approaches with pipelining, and
distributed processing is required. Towards
optimizing the processing efficiency, the processing
digital system efficiency is measured in term of
energy consumption, speed of operation and
accuracy of operation. Wherein various approaches
were developed to achieve these fundamental issues,
a hybrid model of data and clock signaling was
developed in past termed as ‗Mesochronous
processing‘. In such coding, the data and clock
signals were simultaneously communicated over the
processing units to achieve synchronous operation.
Wherein Mesochronous communication is benefited
with synchronous processing, the efficiency of
synchronous coding is required.
II. PREVIOUS METHODS
In the digital systems, synchronization has
become an essential part. Towards optimizing
synchronization problem with efficiency, various past
approaches were made. These approaches are briefly
reviewed in this paper. The approach of
synchronization in the approach of processing,
register optimization, circuit and architectural
optimization, multiple core parallelizations, and its
application to various fields such as network
synchronization and communication synchronization
is presented. The proper synchronization techniques
will improve the power saving along with data
processing speed.
a) Processing Approaches
A Mesochronous pipelining approach for
pipelining process is suggested is presented in [1].
The clock period in conventional pipeline scheme is
observed to be proportional to the maximum stage
delay, whereas in a mesochronous pipelining it is
proportional to the maximum delay difference, which
leads to higher clock speeds in Mesochronous
coding. The clock distribution is simpler hence
leading to lower load and improves power saving. In
[2], a new design methodology of digital system,
called data-oriented methodology, to deal with the
synchronization problems, by using Blue spec
System was defined. Instead of conventional timing
control mechanism, the data-oriented methodology
adopts simple handshaking protocol, blocking
transferring, and explicitly register/FIFO declaration
for communicating between adjacent modules. In
[3], a novel approach for parallelization of two-
dimensional (2D) finite discrete element method
(FDEM), aiming at clusters and desktop computers is
developed. Dynamic domain decomposition based
parallelization solvers covering all aspects of FDEM
have been developed. The overall performance and
scalability of the parallel code is been evaluated
numerically and the suitability of the proposed
approach is presented. In [4], a mergesoft and sample
sorting approach is proposed. The suggested
approach provides a mapping based on integer linear
RESEARCH ARTICLE OPEN ACCESS

programming to address load balancing and latency
considerations. The suggested approach describes a
prototype implementation of proposed approach for
its runtime feasibility requirement. In [5], an
idealized mesoscale simulation for multi processing
unit is proposed. Good scalability is observed up to
49K processors in such system. Timing breakdown of
computational tasks are determined with relocation of
Lagrangian particles and interpolation of those
particles to the grid identified as the most expensive
operation and ideal for optimization. In [6], towards
parallelization a heterogeneous compute kernels and
predicting performance of the compilation results
was suggested. In [7], a new intermediate format
called unrolled concurrent control-flow graph with
surface and depth to represent the structure of an
Esterel program in pipelined computing is proposed.
In [8] a mesochronous pipeline scheme is described
operating on only one data set at a time. In this
mesochronous scheme, pipeline stages operate on
multiple data sets simultaneously. The variations in
supply current drawn by clock network is
significantly less in mesochronous scheme, thus
power supply noise is observed to be less. In most of
the digital systems the clock skew decreases the
performance of the digital systems in terms of
providing good sensitivity or to maintain data
synchronization. In the conventional pipeline system
facing problems due to improper synchronization of
clock pulses. In [9], a universal problem observed in
all digital systems called jitter or skew was focused.
The propagation of information in the digital systems
mainly controlled on the basis of clock pulses. In
most of the digital systems the clock skew decreases
the performance of the digital systems. A new system
is implemented in the path of the clock to reduce the
clock skew. In [10], a low complexity link micro
architecture for mesochronous on-chip
communication that enables skew constraint
looseness in the clock tree synthesis, frequency
speedup, power consumption reduction, and faster
back-end turnarounds is proposed. In [11] a regular
synchronizer and six multi synchronous
synchronizers are implemented on a programmable
logic device and the synchronization is measured.
An experiment system and method for measuring
synchronizers and metastable flip-flops are described.
Two separate settling time constants are shown for a
metastable flop. Clocking cross-talk between
asynchronous clocks is demonstrated. A regular
synchronizer useful for communications between
asynchronous clock domains, while the other
synchronizers providing higher bandwidth
communications between multi-synchronous and
mesochronous domains is presented. In [12] a work
on Synchronous Digital Hierarchy is reviewed. A
brief review regarding the problems in
synchronization of different data rate signals in single
clock, and master slave technique overcome the issue
of synchronization problem. In [13] a low complexity
link micro architecture for mesochronous on-chip
communication that enables skew constraint
looseness in the clock tree synthesis, frequency
speedup, power consumption reduction, and faster
back-end turnarounds is proposed. With respect to
the state of the art, the proposed link architecture
stands for its low power and low complexity
overheads. Moreover, it can be easily integrated into
a conventional digital design flow since it is
implemented by means of standard cells only. In [14]
a short tutorial attempts to present the approach and
the criticality of the subject of metastability and
synchronizers is presented. The issue with system
metastability for false operation is proposed. In [15]
for generating sequential code, the concurrency
expressed in the synchronous programs is
sequentialized. The developed approach was
designed to run on single-core processors. An attempt
generating multi-threaded code from polychronous
data-flow model is proposed. In [16] an approach of
synchrony conditions for purely asynchronous model
in a way that avoids any reference to message delays
and computing step times, as well as system-wide
constraints on execution patterns and network
topology is proposed. A Asynchronous Bounded-
Cycle (ABC) model just bounds the ratio of the
number of forward and backward-oriented messages
in certain relevant cycles in the space–time diagram
of an asynchronous execution is proposed. The clock
synchronization and lock-step rounds is implemented
and proved for the suggested ABC model.
b) Register Optimization approach
In [17] for synchronization in a breadth search
approach, a fine grain processing for synchronization
issue is developed. The data locality in this system is
evaluated. A codelet model focusing on the fine
grain parallelism of processing in synchronous mode
is proposed. The Codelet Model more efficiently
exploits data locality than the Open MP-like
execution models which traditionally focus on
coarse-grain parallelism inside loops. In [18], a high
level implementation of L1-Cache system based on
asynchronous communication oriented design style is
proposed. Each of the units of cache architecture is
pipelined asynchronously using asynchronous
interfaces. The enhanced pipelining increases the area
and energy resulting to the increased control circuitry
requirement for handling large number of
handshakes, resulting in saving of considerable area
omitting clock generation, distribution and gating
circuitry. The developments on the architectural
level developments were reviewed and the future
development towards the synchronization
requirement is presented in [19]. Due to the fact that
each cache memory implementation is time

consuming and error prone process, a synthesizable
and a configurable model proves out to be of
immense help as it aids in generating a range of
caches in a reproducible and quick fashion. The
micro pipelined cache, implemented using C-
Elements acts as a distributed message-passing
system. The RTL cache model implemented,
comprising of data and instruction caches which has
a wide array of configurable parameters. In addition
to timing robustness this implementation has high
average cache throughput and low latency. The
implemented architecture comprises of two direct-
mapped, write-through caches for data and
instruction.
c) Multiple core parallelization
In [20], focus on two different kinds of parallel
desktop architectures: multicore processors and GPU
accelerators was made. The GPU implementation
was carried out based on compute unified device
architecture, while the multicore implementation was
realized using open multiprocessing. The parallel
performance and energy consumption were compared
in the context of a quantitative remote sensing
retrieval application. In [21], a multi core architecture
following parallel processing is developed. The
limitation of the multi core aware programs are
analyzed, and at the same time, parallel programming
issues are presented. It is focused to develop a
method with little latency and synchronization
overheads, for which a loose synchronization
mechanism called ―weak synchronization‖ is
outlined. [22] explores the concepts of multi-core,
trending research areas in the field of multi-core
processors and then concentrates on power
management issues in multi-core architectures.
In [23], a novel scheduling algorithm for pipelines
streaming applications onto multi/many core
architectures is presented. The algorithm generates
pipeline schedules by formulating and solving MILP
(Mixed Integer Linear Programming) problems. The
suggested approach generates schedules that use up
to 71% smaller amount of buffers needed for
communication between kernels compared to
conventional approaches. In [24], an architectural
design of a six-way VLIW digital signal processor
(DSP) with clustered register files was proposed. The
architecture was designed for a variable length
instruction set and supporting dynamic instruction
dispatching. The one-level memory system
architecture of the processor instruction and data
caches and with instruction and data on-chip RAM
was proposed.
d) Applications to Network synchronization
The synchronization in network on chip, a
mesochronous scheme for communication over serial
buses in network on chips (NoC) is proposed in [25].
The technique, which removes metastability errors in
mesochronous communications, makes use of only
one strobe line with the bus is defined. The strobe
line toggles once with every frame of the data. In the
suggested method, the frequencies of the transmitter
and the receiver are kept independent. In [26], a
switch design which greatly reduces the overhead for
mesochronous synchronization and different layout
constraints is proposed. A framework of
mesochronous links that can direct the selection of
synchronization options on a port by- port basis for
all the switches in the NoC, based on timing and
layout constraints is proposed. In [27], a scheme to
handle mesochronous communication in NoC is
presented. The proposed approach analyzes (i) the
circuit design, (ii) the timing properties, (iii) the
requirements to support flow control across
mesochronous links, (iv) the implementation cost of
such a scheme after placement and routing is
proposed. In [28], two high-throughput, low-latency
converters which could be used to convert
synchronous communication protocol to
asynchronous one and vice versa is proposed. The
two hardware components have been designed to use
in Multi-Processor System on Chip communicating
by a full asynchronous Network on Chip (NoC). The
proposed architecture was made generic, which
allows the system designer to make various trade-off
between latency and robustness, depending on the
selected synchronizer. In [29], a run-time
environment that supports classes of programming
models and their composition for synchronization in
network processor was presented. This approach
combines the single-process multi-threaded execution
inside the compute clusters and I/O subsystems, with
set of specific Inter-Process Communication (IPC)
primitives that exploit the NoC architecture. This
approach provides a run-time support for the different
target programming models in multi-process
environment. In [30], a low-power, high-speed
source-synchronous link transceiver was proposed.
The techniques for NoC applications for power
optimization in relate to synchronization is proposed.
The issue of random mismatch, crosstalk, and
different count of transceiver unit in the network is
suggested. In [31], focus on NP-based high
performance inline stateful deep inspection is made.
This approach provides a dominant function block in
intrusion prevention systems (IPSs). This approach
investigates architecture-aware session design on
NPs, and provides an instance of implementation to
exploit the parallelism of the multi-core,
multithreaded NP. In [32], stateful pipeline buffers
added to long links allowing higher clock rate. The
wastage of resources on links needing only low
bandwidth was focused. In the asynchronous NoCs,
link pipelining is applied only to those that will
benefit from both increased through-put and

buffering capacity, and is especially useful in
heterogeneous embedded SoCs. Two strategies that
determine where link pipeline buffers should be
placed in the topology is evaluated. The first
compares available link bandwidth, based on physical
wirelength, to the throughput needed by each source-
to-destination path, for each link. In [33], a
composable and predictable NoC architecture, that
offers only GS, based on flit-synchronous Time
Division Multiplexing (TDM) is proposed. In
contrast to other TDM based NoCs, scalability on the
physical level is achieved by using mesochronous or
asynchronous links. Functional scalability is
accomplished by completely isolating applications,
and by having a router architecture that does not limit
the number of service levels or connections.
e) Circuit and architectural optimization
In [34], a stencil computation for implementation
of state-of-the-art general purpose graphics
processing units is suggested. Stencil codes are used
for the core of numerical solvers and physical
simulation codes used for computing. The operation
of stencil code used for their superior floating point
performance and memory bandwidth is presented.
Especially memory bound stencil codes were
proposed exploiting the enlarged on-chip shared
memory to perform two time step updates per sweep.
This represents the application of temporal blocking
for stencils on processing units. It has been presented
a concise methodology to enable Invasive
Programming properties on an embedded Multi-
Processor System-on-Chip (MPSoC) in [35]. This is
achieved by combining a designer-guided code
parallelization approach with a virtualization,
generic, and scalable embedded MPSoC architecture.
To resolve data dependencies during task invasion, a
processor-independent task-based communication
scheme for the MPSoC is proposed. The approach is
demonstrated by the generation of an MPSoC,
featuring eight processors executing an application
which dynamically switches at runtime between
sequential and parallel execution. For the application
applicability, and with the number of events used in
[36], a multi-threaded CPU Open MP coding is
proposed giving best performance under parallel
processing. The study also shows that there is a
―break-even‖ point of the number of events where the
use of GPUs helps performance and improves
computation time. In [37], a distributed clock
generation scheme for Systems-on-Chip under
variant fault condition was proposed. A self-
stabilizing hardware blocks and a hybrid
synchronous/asynchronous state machine enabling
metastability-free transitions of the algorithm‘s states
was proposed. In [38], a clock network design
methodology that optimizes register placement was
proposed. In the suggested work, for a given cell
placement, incremental modifications are performed
based on the clock skew specifications by moving
registers toward preferred locations that may reduce
the clock network size. At the same time, the side-
effects to logic cell placement, such as signal net wire
length and critical path delay, were controlled. In
[39], a simple method using logic gates are taken to
check the previous flip flop binary information. The
logic gates create simple delay in producing the clock
to the next stage. Until the logic gates identify the
next binary bit from previous stage it will not allow
the clock generator to pass the next clock pulse to the
next stage of the circuit. In [40], an architecture
called asynchronous array of simple processors
(AsAP) is defined. AsAP uses a simple processor
architecture with small memories to increase energy
efficiency. The globally asynchronous locally
synchronous (GALS) clocking style and nearest-
neighbor communications improves the scalability,
and provide opportunities to mitigate effects of
device variations, global wire limitations, and
processor failures. In [41], a new structure of a DLL
circuit with clock alignment capability of both
leading and trailing output pulse edges is defined.
This circuit is used to obtain correct the duty-cycle
factor in a multistage clock buffer operation. In [42],
a transistor level modeling for delay tolerance logic
was developed. The suggested approach avoids a
significant increase in delay by allowing a full rail-to-
rail swing on a lightly loaded second dynamic node
with a small parasitic capacitance and which is
inversely coupled to the output. In [43], parallel
application signature for performance prediction
(PAS2P) is proposed. Based on the application‘s
message-passing activity, to identify and extract
representative phases, with which a parallel
application signature is created which enable to
predict the application‘s performance under
synchronous mode of communication is proposed.
The approach experimented with different scientific
applications on different clusters. In [44], several new
design strategies, which represent the current design
trends to deal with the emerging issues were
explored. For recognizing the stringent requirements
on power consumption, memory bandwidth/latency,
and transistor variability, novel power/thermal
management, multi-processor SoC, reconfigurable
logic, and design for verification and testing have
been incorporated into system design. In [45],
analysis on cluster global time continuity, using
global time change models for the node dynamics is
proposed. The obtained results shown that defining
the global time using the cluster average time (AGT)
is more stable than defining it using a single node‘s
local time (SGT). With normally distributed clock-
parameter assumptions, the AGT change bounds are
at most 70.7% of those for the SGT‘s. In [46], the
impact of the Total Store Order (TSO) memory

model is investigated, which is used by Intel x86 and
Sun SPARC processors, on secure information flow.
Two approaches of data flow for information flow in
such processor architecture is proposed. In [47], a
methodology for the design and implementation of a
self-timed reconfigurable control device suitable for a
parallel cascaded flip-flop synchronizer based on a
principle known as wagging, operating through the
application of distributed feedback graphs is
proposed. By modifying the endpoint adjacency of a
common behavior graph via one-hot codes, several
configurable modes can be implemented in a single
design specification, thereby facilitating direct
control over the synchronization time and the mean-
time between failures of the parallel master-slave
latches in the synchronizer. In [48], a Clock
distribution networks is proposed which synchronize
the flow of data signals among synchronous data
paths. The designs of these networks dramatically
affect system-wide performance and reliability. A
theoretical background of clock skew is provided in
order to better understand how clock distribution
networks interact with data paths. Minimum and
maximum timing constraints are developed from the
relative timing between the localized clock skew and
the data paths. These constraint relationships are
reviewed, and compensating design techniques were
discussed. In [49], an on-chip measurement circuit to
measure multi-giga bit cycle-to-cycle jitter based on
the vernier oscillator (VO) is proposed. This is
inherited from the famous vernier delay line. The
calibration of this method is also given. The circuit
adopts a differential digital controlled delay element,
which makes the circuit flexible in adjusting the
measurement resolution, and a highly sensitive phase
capturer, which makes the circuit able to measure
jitters in pico-second range. In [50], a special purpose
processing element is described which is used to
optimize the operational outline of the processing
element. It has been used to configure the FPGAs of
the massively parallel hardware platform RIVYERA.
In [51], a novel low power adaptive pulse triggered
flip-flop (PTFF) featuring exclusive-or gate based
clock gating with replica path delay scheme is
proposed. Clock gating is a very accepted technique
to reduce dynamic power of idle clocking
subsystems. Incorporating clock gating with PTFF
leads to reduction in dynamic power consumption
and replica path delay pulse generator simplifies
design effort and achieves robust timing
characteristics as compared to the conventional
PTFF. The proposed PTFF features best power delay
product performance. In [52], to achieve the issue of
synchronization, a technique for reducing the S/N
ratio of emanating information by signal reduction
and noise generation have been proposed. A
technique that fluctuate frequencies of synchronizing
signal to reduce correlation between timings of
display and transmission is proposed. In [53], the
phase-stabilized transfer of a chirped frequency as a
tool for synchronization and time transfer is
proposed. A remote measure for the synchronization
of remote counters‘ gate intervals without using an
external time reference is proposed. The results were
directly applicable to the remote synchronization of
frequency measurements. In [54], A robust, scalable,
and power efficient dual-clock first-input first-out
(FIFO) architecture which is useful for transferring
data between modules operating in different clock
domains is presented. The architecture supports
correct operation in applications where multiple
clock cycles of latency exist between the data
producer, FIFO, and the data consumer; and with
arbitrary clock frequency changes, halting, and
restarting in one or both clock domains. In [55], a
parallel computing demands synchronous clocking of
multiple core processors to reliably carry out joint
computations is proposed. The mutually coupled
phase-locked loops (PLLs) enable synchronous
clocking in large-scale systems with transmission
delays. A phase description of coupled PLLs that
includes filter kernels and delayed signal
transmission is proposed. The transmission delays in
the coupling, enable the existence of stable
synchronized states, while instantaneously coupled
PLLs do not tend to synchronize. The filtering and
transmission delays combinly govern the collective
frequency and the time scale of synchronization.
f) Communication application
In [56], an approach to timing and data handling
that support the operation of single and correlated
communication systems at their maximum capacity
was proposed. The suggested approach was focused
for FPGA device operating for higher operating
frequency of operation. In [57], a novel lumped time-
delay compensation scheme for an communication
system was proposed. The transfer of data in the flow
was synchronized with a time delay of different
quantized pulse, accurately compensated with a
simple structure compared to the multiple time-delay
lines. In [58], a formal specification and verification
of a new fault-tolerant real-time communication
protocol, called DoRiS, is suggested. This approach is
designed for supporting distributed real-time systems
that use a shared high-bandwidth medium. In [59],
certain value on the maximum number of messages
communicated regardless of the sparsity pattern of
the matrix is developed. The downside, however, is
the increased message volume and the negligible
redundant computation. Reducing the message
latency costs at the expense of increasing message
volume was suggested. Two iterative-improvement
based on heuristics to alleviate the increase in the
volume through one-to-one task-to-processor
mapping is proposed. In [60], focuses on the

partitioning of transceiver systems and on the
implementation of application-specific components
to introduce an advanced multiprocessor system-on-
chip interface and control architecture which is able
to fulfill the requirements of future transceiver
integrations is proposed. The proposed framework
demonstrates a high degree of scalability, flexibility,
and reusability. With the suggested approach the time
to market for products can be reduced and fast
adaptations to the requirements of the market are
made feasible.
g) Globally asynchronous - Locally synchronous
(GALS)
A novel burst-mode globally asynchronous -
locally synchronous (GALS) technique was proposed
in [61]. The goal of this technique was to improving
the performance of the GALS approach for systems
with predominantly bursty data transfer. The
simulation results observed illustrated a significant
performance improvement in comparison with the
classical implementation of GALS using pausible
clocking. In [62], the problem of synthesizing the
asynchronous wrappers and glue logic needed for
correct GALS implementation of a modular
synchronous system is proposed. Suggested approach
was developed based on the weakly endochronous
synchronous model, which defines high-level,
implementation-independent conditions guaranteeing
correct de-synchronization at the level of the abstract
synchronous model. In [63], a novel GALS
architecture for a general-purpose multiprocessor
platform that is intended specifically for the use in
hard real-time systems is proposed. The platform
contains one NoC offering access to a shared
memory and one NoC supporting message passing,
each of them optimized for its purpose of use is
developed. A message passing NoC architecture and
implementation reflecting two main requirements: 1)
support for hard real-time applications and 2)
implementation of GALS is developed. In [64], a
Synchronizers and arbiters for any Globally
Asynchronous, Locally Synchronous network is
proposed. They contribute to latency, because of the
synchronization time required for reliability, and to
metastability delay in the arbiters. In [65], a clock
distribution and synchronization approach to address
the issue of jitter in large scale circuit is proposed. A
reconfigurable data path processor (RDP) is
proposed, which is carried out in several stages for
synchronization. The operation of synchronization
was carried over a operand routing network (ORN)
and clock control, synchronization scheme.
III. CONCLUSION
Synchronization is observed to be a very
important issue in digital systems. The approach of
providing synchronization is variant. In this survey, it
is illustrated that various modes of parallelization
architectures and processing methods were
developed. Various modes of register optimization
methods were developed to minimize the delay
latency issue in digital system operation. In the
modeling of synchronization approach to SoC
modeling the application to network synchronization
was observed. In the current developments and future
prospective new NoC units are developing, though
the integration has various advantages, the processing
synchronization and power consumptions are a major
issue. In various real time applications this approach
of global synchronization issue is to be solved for a
better services. This observation leads to the
requirement of synchronization in digital system
designing, and its limitation in resource utilization in
term of power and area coverage constraints.
REFERENCES
[1] Suryanarayana B. Tatapudi and Jose G.
Delgado- Frias, ―A Mesochronous Pipeline
Scheme for High Performance Low Power
Digital Systems‖, ISCAS, IEEE, 2006.
[2] S. L. Chu, M. J. Lo, ―A New Design
Methodology for Composing Complex
Digital Systems‖, Journal of Applied
Research and Technology, vol.11, 2013.
[3] T. Lukas, G.G. Schiava D Albano, A.
Munjiza, ―Space decomposition based
parallelization solutions for the combined
finite discrete element method in 2D‖,
Journal of Rock Mechanics and
Geotechnical Engineering 6, 007-615, 2014.
[4] Nicolas Melot, Christoph Kessler, Kenan
Avdic, Patrick Cichowski, J¨org Keller,
―Engineering parallel sorting for the Intel
SCC‖, International Conference on
Computational Science, ICCS 2012, WEPA,
Procedia Computer Science 9, 1890 – 1899,
Elsevier, 2012.
[5] Joseph R. Peterson, Charles A. Wight,
Martin Berzins, ―Applying high-
performance computing to petascale
explosive Simulations‖, International
Conference on Computational Science,
Procedia Computer Science 18, 2259 –
2268, Elsevier, 2013.
[6] Nicolas Benoit and Stephane Louise, ―A
First Step to Performance Prediction for
Heterogeneous Processing on Manycores‖,
International Conference On Computational
Science, Procedia Computer Science, 51,
2952–2956, Elsevier, 2015.
[7] Simon Yuan, Sidharta Andalam, Li Hsien
Yoong, Partha S. Roop, Zoran Salcic,
―STARPro - A new multithreaded direct
execution platform for Esterel‖, Electronic

Notes in Theoretical Computer Science 238,
37–55, Elsevier, 2009.
[8] Suryanarayana B. Tatapudi and José G.
Delgado-Frias, ―A Reduced Clock Delay
Approach for High Performance
Mesochronous Pipeline‖, IEEE, 2006.
[9] N. Suresh Kumar, D.V.Rama KotiReddy, B.
Pramod Kumar, A. Chaitanya Prabhu, ―Two
Way Clock Scheme In Pipeline To
Minimize The Clock Skew‖, Global Journal
of Computer Science and Technology, Vol.
10, 2010.
[10] Francesco Vitullo, Nicola E. L‘Insalata, Esa
Petri, Sergio Saponara, Luca Fanucci, ―Low-
Complexity Link Micro architecture for
Mesochronous Communication in Networks
on Chip‖, IEEE Transactions on Computers,
Vol. 57, No. 9, September 2008.
[11] Yaron Semiat and Ran Ginosar, ―Timing
Measurements of Synchronization Circuits‖,
Proceedings of the 9th
International
Symposium on Asynchronous Circuits and
Systems, IEEE, 2003.
[12] Abhishek Agwekar, Mohammed Ahmed,
Ramanand Singh and A. Riyaz,
―Synchronization Problems in Synchronous
Digital Hierarchy (SDH) Communication
System and Master Slave Strategies,‖
International Journal of Scientific
Engineering and Technology, 2012.
[13] Francesco Vitullo, Nicola E. L‘Insalata, Esa
Petri, Sergio Saponara, Luca Fanucci,
Michele Casula, Riccardo Locatelli, and
Marcello Coppola, ―Low-Complexity Link
Micro architecture for Mesochronous
Communication in Networks on Chip,‖
IEEE Transactions On Computers,
September 2008.
[14] Ran Ginosar, ―Metastability and
Synchronizers: A Tutorial,‖ IEEE Design &
Test, 2011.
[15] Bijoy A. Jose, HirenD. Patelb,Sandeep K.
Shuklaa and Jean-Pierre Talpin, ―Generating
Multi-Threaded Code from Polychronous
Specifications,‖ Electronic Notes in
Theoretical Computer Science, 2009.
[16] Peter Robinson and Ulrich Schmid, ―The
Asynchronous Bounded-Cycle model,‖
[17] Chen Chen, Souad Koliai, and Guang Gao,
―Exploitation of Locality for Energy
Efficiency for Breadth First Search in Fine-
Grain Execution Models‖, tsinghua science
and technology, 18, 6, 2013.
[18] Mansi Jhamb, R.K. Sharma , A.K. Gupta,
―A high level implementation and
performance evaluation of level-I
asynchronous cache on FPGA‖, Journal of
King Saud University – Computer and
Information Sciences, Elsevier 2015 .
[19] Stefano Ruffini, ―Network Synchronization
— Stand-Alone Products That Support The
Design Of Synchronization Networks,‖
Ericsson Review No. 1, 2004.
[20] Jia Liu, Dustin Feld, Yong Xue, Jochen
Garcke, and Thomas Soddemann,
―Multicore Processors and Graphics
Processing Unit Accelerators for Parallel
Retrieval of Aerosol Optical Depth From
Satellite Data: Implementation,
Performance, and Energy Efficiency‖, IEEE
Journal of Selected Topics in applied earth
observations and remote sensing, Vol. 8, No.
5, 2015.
[21] Stephane Louise, Vincent David, Fabien
Calcado, ―A single processor approach for
loosely synchronized execution of parallel
flows on heterogeneous multicore‖,
International Conference on Computational
Science, ICCS 2011, Procedia Computer
Science 4, 1997–2006, Elsevier, 2011.
[22] Khaled M. Attia , Mostafa A. El-Hosseini,
Hesham A. Ali, ―Dynamic power
management techniques in multi-core
architectures: A survey study‖, Ain Shams
Engineering Journal, Elsevier, 2015.
[23] A. Hatanaka, N. Bagherzadeh, ―A software
pipelining algorithm of streaming
applications with low buffer requirements‖,
Computer Science & Engineering and
Electrical Engineering, Scientia Iranica D,
19 (3), 627–634, Elsevier, 2012.
[24] Shen Zheng, He Hu, Yang Xu, Jia Di and
Sun Yihe, ―Architecture Design of a
Variable Length Instruction Set VLIW
DSP,‖ Tsinghua Science and Technology,
October 2009.
[25] Mohsen Saneei, Ali Afzali-Kusha,
Zainalabedin Navabi, ―A Mesochronous
Technique for Communication in Network
on Chips‖, the 18th
International Conference
on Microelectronics (ICM), IEEE, 2006.
[26] Daniele Ludovici, Alessandro Strano,
Georgi N. Gaydadjiev, Luca Benini, Davide
Bertozzi, ―Design Space Exploration of a
Mesochronous Link for Cost-Effective and
Flexible GALS NoCs‖, EDAA, 2010.
[27] Igor Loi, Federico Angiolini, Luca Benini,
―Developing Mesochronous Synchronizers
to Enable 3D NoCs‖, EDAA, 2008.
[28] A. Sheibanyrad and A. Greiner, ―Two
Efficient Synchronous Asynchronous
Converters Well-Suited for Network on
Chip in GALS Architectures,‖ PATMOS
2006.

[29] Benoıt Dupont de Dinechina, Pierre
Guironnet de Massasa, Guillaume Lagera,
Clement Legera, Benjamin Orgogozoa,
Jerome Reyberta and Thierry Strudel, ―A
Distributed Run-Time Environment for the
Kalray MPPAR-256 Integrated Manycore
Processor,‖ International Conference on
Computational Science, ICCS 2013.
[30] Daniel Schinkel, Eisse Mensink, Eric A. M.
Klumperink, Ed van Tuijl, and Bram Nauta,
―Low-Power, High-Speed Transceivers for
Network-on-Chip Communication,‖ IEEE
Transactions On Very Large Scale
Integration (VLSI) Systems, 2008.
[31] Xu Bo, He Fei, Xue Yibo And Li Jun,
―Architecture-Aware Session Lookup
Design for Inline Deep Inspection on
Network Processors,‖ Tsinghua Science And
Technology, February 2009.
[32] Daniel Gebhardt, Junbok You and Kenneth
S. Stevens, ―Link Pipelining Strategies for
an Application-Specific Asynchronous
NOC,‖ ACM 2011.
[33] Andreas Hansson, Mahesh Subburaman and
Kees Goossens, ―AELITE: A Flit-
Synchronous Network on Chip with
Composable and Predictable Services,‖
EDAA 2009.
[34] Andreas Schäfer, Dietmar Fey, ―High
Performance Stencil Code Algorithms for
GPGPUs‖, International Conference on
Computational Science, ICCS 2011,
Procedia Computer Science 4, 2027–2036,
Elsevier, 2011.
[35] Alexander Biedermann, Sorin A. Huss, ―A
Methodology for Invasive Programming on
Virtualizable Embedded MPSoC
Architectures‖, International Conference on
Computational Science, ICCS 2013,
Procedia Computer Science 18, 359 – 368,
Elsevier, 2013.
[36] S.J. Lewis b, D.G.Ireland b,n,
W.Vanderbauwhede , ―Code optimisation in
a nested-sampling algorithm‖,
NuclearInstrumentsandMethodsinPhysicsRe
searchA785, 105–109, Elsevier, 2015.
[37] Danny Doleva, Matthias Függerb, Markus
Poschb, Ulrich Schmidb, Andreas
Steiningerb and Christoph Lenzen,
―Rigorously Modelling Self-Stabilizing
Fault-Tolerant Circuits: Anultra- Robust
Clocking Scheme For Systems-On-Chip,‖
Journal of Computer and System Sciences,
2014.
[38] Zhou Qiang, Cai Yici, Huang Liang And
Hong Xianlong, ―Incremental Placement-
Based Clock Network Minimization
Methodology,‖ Tsinghua Science And
Technology, February 2008.
[39] N. Suresh Kumar, G. Babu Rao, K.V.
Ramana Rao, and Anil P, ―Clock
Synchronization in Digital Circuits,‖
International Journal of Engineering Trends
and Technology, 2011.
[40] Zhiyi Yu, Michael J. Meeuwsen, RyanW.
Apperson, Omar Sattari, Michael Lai,
Jeremy W. Webb, Eric W. Work, Dean
Truong, Tinoosh Mohsenin, and Bevan M.
Baas, ―AsAP: An Asynchronous Array of
Simple Processors,‖ IEEE Journal Of Solid-
State Circuits, vol. 43, No. 3, March 2008.
[41] Mile Stojcev and Goran Jovanovic, ―Clock
Aligner Based On Delay Locked Loop With
Double Edge Synchronization,‖
Microelectronics Reliability, 2008.
[42] Rahul Singh, Gi-Moon Hong, and Suhwan
Kim, ―Bitline Techniques with Dual
Dynamic Nodes for Low-Power Register
Files,‖ IEEE Transactions on Circuits and
Systems, April 2013.
[43] Alvaro Wong, Dolores Rexachs, and Emilio
Luque, ―Parallel Application Signature for
Performance Analysis and Prediction,‖
IEEE Transactions on Parallel and
Distributed Systems, July 2015.
[44] Yen-Kuang Chen and S. Y. Kung, ―Trend
and Challenge on System-on-a-Chip
Designs‖, Journal of Signal Processing
Systems, Volume 53, Issue 1, pp 217-229,
November 2008.
[45] Li Li, Liu Yongpan, Yang Huazhong and
Wang Hui, ―Influence of Node Dynamics on
Cluster Global Time Continuity,‖ Tsinghua
Science And Technology, April 2011.
[46] Jeffrey A. Vaughan and Todd Millstein,
―Secure Information Flow for Concurrent
Programs under Total Store Order,‖
Computer Security Foundations Symposium,
2015.
[47] James S. Guido, and Alexandre Yakovlev,
―Design of Self-Timed Reconfigurable
Controllers for Parallel Synchronization via
Wagging,‖ IEEE Transactions On Very
Large Scale Integration (VLSI) Systems,
Feburary 2015.
[48] Eby G. Friedman, ―Clock Distribution
Networks in Synchronous Digital Integrated
Circuits,‖ IEEE 2001.
[49] Zhang Jingkai, Chung Len Lee, Tian Chao
and Yu Fei, ―On-Chip Multi-Giga Bit
Cycle-to-Cycle Jitter Measurement Circuit,‖
Tsinghua Science and Technology, July
2007.
[50] Christoph Starke, Vasco Grossmann, Lars
Wienbrandt and Manfred Schimmler, ―An

FPGA Implementation of an Investment
Strategy Processor,‖ International
Conference on Computational Science,
ICCS 2012.
[51] Krishna Kavali, S. Rajendar and R. Naresh,
―Design of Low Power Adaptive Pulse
Triggered Flip-Flop Using Modified Clock
Gating Schemeat 90nm Technology,‖ 2nd
International Conference on Nanomaterial’s
and Technologies, CNT, 2014.
[52] Takashi Watanabe, Hiroto Nagayoshi,
Hiroshi Sako and Toshirou Uemura,
―Synchronization Clock Frequency
Modulation Technique for Compromising
Emanations Security,‖ IEICE 2009.
[53] Sebastian M. F. Raupach and Gesine
Grosche, ―Chirped Frequency Transfer: A
Tool for Synchronization and Time
Transfer,‖ IEEE Transactions on
Ultrasonics, Ferroelectrics, and Fr 920
equency Control, June 2014.
[54] Ryan W. Apperson, Zhiyi Yu, Michael J.
Meeuwsen, Tinoosh Mohsenin, and Bevan
M. Baas, ―A Scalable Dual-Clock FIFO for
Data Transfers Between Arbitrary and
Haltable Clock Domains,‖ IEEE
Transactions On Very Large Scale
Integration (VLSI) Systems, October 2007.
[55] Alexandros Pollakis, Lucas Wetzel, David J
Jorg, Wolfgang Rave, Gerhard Fettweis and
Frank Julicher, ―Synchronization In
Networks Of Mutually Delay coupled
Phase-Locked Loops,‖ New Journal of
Physics, 2014.
[56] Alan Mink, Joshua C Bienfang, Robert
Carpenter, Lijun Ma, Barry Hershman,
Alessandro Restelli and Xiao Tang,
―Programmable Instrumentation And
Gigahertz Signalling For Single-Photon
Quantum Communication Systems,‖ New
Journal of Physics, 2009.
[57] Zhe Kang, Jinhui Yuan, Qiang Wu, Tao
Wang, Sha Li, Xinzhu Sang, Chongxiu Yu
and Gerald Farrell, ―Lumped Time-Delay
Compensation Scheme for Coding
Synchronization in the Nonlinear Spectral
Quantization-Based All-Optical Analog-to-
Digital Conversion,‖ IEEE Photonics
Journal, December 2013.
[58] Paul Regnier, George Lima and Aline
Andrade, ―A TLA+ Formal Specification
and Verification of a New Real-Time
Communication Protocol,‖ Electronic Notes
in Theoretical Computer Science, 2009.
[59] R. Oguz Selvitopi, Muhammet Mustafa
Ozdal, and Cevdet Aykanat, ―A Novel
Method for Scaling Iterative Solvers:
Avoiding Latency Overhead of Parallel
Sparse-Matrix Vector Multiplies,‖ IEEE
Transactions On Parallel And Distributed
Systems, March 2015.
[60] Siegfried Brandstätter and Mario Huemer,
―A Novel MPSoC Interface and Control
Architecture for Multistandard RF
Transceivers,‖ IEEE 2014.
[61] Milos Krstic, Xin Fan, Eckhard Grass and
Frank K. Gurkaynak, ―GALS for Bursty
Data Transfer based on Clock Coupling,‖
Electronic Notes in Theoretical Computer
Science, 2009.
[62] S. Dasgupta, D. Potop-Butucaru, B.Caillaud
and A.Yakovlev, ―Moving from Weakly
Endochronous Systems to Delay-Insensitive
Circuits,‖ Electronic Notes in Theoretical
Computer Science, 2006.
[63] Evangelia Kasapaki, Martin Schoeberl,
Rasmus Bo Sorensen, Christoph Muller,
Kees Goossens, and Jens Sparso, ―Argo: A
Real-Time Network-on-Chip Architecture
With an Efficient GALS Implementation‖
IEEE Transactions On Very Large Scale
Integration (VLSI) Systems, 2015.
[64] David Kinniment, ―Synchronization and
Arbitration in GALS,‖ Electronic Notes in
[65] Irina Kataeva, Hiroyuki Akaike, Akira
Fujimaki, Nobuyuki Yoshikawa and
Naofumi Takagi, ―Experimental
Demonstration of an Operand Routing
Network Prototype Employing Clock
Control and Data Synchronization Scheme,‖
Superconductivity Centennial Conference
2012.
K. SUKANYA presently working
as Associate professor in the
department of Electronics and
Communication Engineering at
TKR College of Engineering &
Technology, Medbowli, Meerpet,
SaroorNagar, Hyderabad,
Telangana State, INDIA. She has 7 years of teaching
experience. She is associated with ISTE as life
member. She has obtained B. Tech. degree in
Electronics and Communication Engineering from
Jayamukhi Institute of Technological Sciences,
Warangal, Jawaharlal Nehru Technological
University Hyderabad, in 2006, M.Tech. degree in
Embedded Systems from Ramappa Engineering
College, Warangal, Jawaharlal Nehru Technological
University Hyderabad, in 2011 and my area of
Research interest is Embedded Systems, Ph.D (ECE)
from Jawaharlal Nehru Technological University,
Hyderabad and it is my part of Research work.

Dr. G. LAXMINARAYANA
presently working as Principal of
Anurag College of Engineering.
He has obtained BE from
Osmania university, M.Tech
from Indian Institute of Science,
Bangalore and Ph.D from
JNTUH under the guidance of
Dr. K. Lalkishore (VC of JNTUA). He has 5 years of
industrial experience and 30 years of teaching
experience. He worked in Osmania University from
1979 to 1998. He worked as Head of the department,
ECE at Sreenidhi Engineering College, VBIT and
Aurora. He also worked as Director of Aurora
Scientific Technological and Research Academy and
Principal of Holy Mary Institute of Technology. He is
an industrial consultant in instrumentation and
worked in South Central Railways. He is associated
with IETE for last 20 years and also member of
IEEE, ISOI and ISTE. Presently he is an Executive
Committee member in the present body of
Hyderabad IETE chapter and R&D subcommittee
chair at IETE Hyderabad. He is supervising 12 Ph.D
students and published various papers in International
journals and reputated National journals.

A Review - Synchronization Approaches to Digital systems

More Related Content

What's hot (18)

Similar to A Review - Synchronization Approaches to Digital systems (20)

Recently uploaded (20)

A Review - Synchronization Approaches to Digital systems