SlideShare a Scribd company logo
1 of 46
Basic Concepts Of DSM:
๏‚ž A DSM system provides a logical abstraction
of shared memory which is built using a set
of interconnected nodes having physically
distributed memories.
Types of DSMs
๏‚ž Hardware level DSM
๏‚ž Software level DSM
๏‚ž Hybrid level DSM
Advantages of DSM
๏‚ž Simple abstraction
๏‚ž Improved portability of distributed
application programs
๏‚ž Provides better performance in some
applications
๏‚ž Large memory space at no extra cost
๏‚ž Better than message passing systems
Comparison of IPC
paradigms
๏ƒ˜ DFS
๏‚— Single shared address space
๏‚— Communicate, synchronize using load / store
๏‚— Can support message passing
๏ƒ˜ Message Passing
๏‚— Send / Receive
๏‚— Communication + synchronization
๏‚— Can support shared memory
Hardware architectures
๏‚ž On chip memory
๏‚ž Bus based multiprocessor
๏‚ž Ring based multiprocessor
๏‚ž Switched multiprocessor
On chip memory
In this CPU portion of the chip has a address and data lines that directly
connect to the memory portion.
Such chips are used in cars appliances and even toys
Bus based multiprocessor
All CPUs connected to one bus (backplane)
Memory and peripherals are accessed via shared bus. System looks the
same from any processor.
Consistency protocols
Ring DSM architecture
๏‚ž Shred memory :
๏‚— Private areas
๏‚— Shared areas
(Ring) Memnet: Node memory
Switched multiprocessor:
๏‚ž Multiple clusters interconnected by a bus offer better
scalability
๏‚ž Example : Dash system
NUMA Multiprocessor
๏‚ž Non-uniform memory access (NUMA) is a
computer memory design used multiprocessing,
where the memory access time depends on the
memory location relative to the processor.
๏‚ž Under NUMA, a processor can access its own
local memory faster than non-local memory.
๏‚ž The benefits of NUMA are limited to particular
workloads, notably on servers where the data is
often associated strongly with certain tasks or
users.
UMA Multiprocessor
๏‚ž Uniform memory access (UMA) is a shared memory
architecture used in parallel computers. All the processors
in the UMA model share the physical memory uniformly.
๏‚ž In a UMA architecture, access time to a memory location
is independent of which processor makes the request or
which memory chip contains the transferred data.
๏‚ž Uniform memory access computer architectures are often
contrasted with non-uniform memory access (NUMA)
architectures.
๏‚ž In the UMA architecture, each processor may use a
private cache. Peripherals are also shared in some
fashion.
๏‚ž The UMA model is suitable for general purpose and time
sharing applications by multiple users. It can be used to
speed up the execution of a single large program in time-
critical applications.
D/f bw two Multiprocessor:
BUS BASED
MULTIPROCCESORS
RING BASED
MULTIPROCCESORS
๏‚ž They are tightly coupled
with the CPUโ€™S
normally in a single
rack.
๏‚ž It has separate global
memory
๏‚ž Machines here can be
much more loosely
coupled and this loose
coupling can affect their
performance.
๏‚ž It has no separate
global memory.
DSM design issues
๏‚ž Granularity of sharing
๏‚ž Structure of data
๏‚ž Consistency models
๏‚ž Coherence protocols
Granularity
๏‚ž False sharing
๏‚ž Thrashing
Thrashing:
๏‚ž False sharing
๏‚ž Techniques to reduce
thrashing:
๏‚— Application controlled lock
๏‚— Pin the block to a node for
specific time
๏‚— Customize algorithm to
shared data usage pattern
DSM structure
๏‚ž Organization of data items in the shared
memory
Consistency models
//Important.
๏‚ž Refers to how recent the shared memory updates
are visible to all the other processes running on
different machines
Strict consistency
๏‚ž Strongest form of consistency.
๏‚ž All write operations are visible to all processes.
Sequential consistency
๏‚ž All processors in the system observe the same
ordering of reads and writes which are issued in
sequence by the individual processors
Causal consistency
๏‚ž Weakening of sequential consistency for
better concurrency
๏‚ž Causally related operation is the one which
has influenced the other operation
PRAM consistency
๏‚ž Pipelined Random Access Memory consistency
๏‚ž Write operations performed by different processes may
be seen by different processes in different orders
๏‚ž All write operations of a single process are pipelined
๏‚ž Simple, easy to implement and has good performance.
Processor consistency
๏‚ž Adheres to the PRAM consistency
๏‚ž Constraint on memory coherence
๏‚ž Order in which the memory operations
are seen by two processors need not be
identical, but the order of writes issued
by each processor must be preserved
Weak consistency
๏‚ž Use a special variable called the
synchronization variable.
๏‚ž Very difficult to show and keep track of the changes at
time to time.
Properties of the weak
consistency model:
๏‚ž Access to synchronization variables is
sequentially consistent.
๏‚ž Only when all previous writes are
completed everywhere, access to
synchronizations variable is allowed.
๏‚ž Until all previous accesses to
synchronization variables are performed,
no read write data access operations will
be allowed.
Release consistency
๏ถ Uses two synchronization variables (Acquire and
Release)
๏ถ Release consistency model uses synchronization
mechanism based on barriers.
Eager Release
Consistency(ERC)
Lazy Release
Consistency(LRC)
Entry consistency
โ€ข Use acquire and release at the start and
end of each critical section, respectively.
โ€ข Each ordinary shared variable is
associated with some synchronization
variable such as a lock or barrier.
โ€ข Entry consistency (EC) is similar to LRC
but more relaxed; shared data is explicitly
associated with synchronization primitives
and is made consistent when such an
operation is performed
Scope consistency
๏‚ž A scope is a limited view of memory with
respect to which memory references are
performed
Comparison of consistency
models-1
๏‚ž Most common: sequential consistency
model .
Comparison of Consistency
models-2
๏‚ž Based on efficiency and programmability
Coherence protocols
๏‚ž Specifies how the rules set by the
memory consistency model are to be
implemented
Coherence algorithms
๏‚ž Maintain consistency among replicas
Multiple Reader/ Multiple Writer
algorithm
๏‚ž Uses twin and diff creation technique
Write Protocols for consistency
๏‚ž Write Update (WU)
๏‚ž Write Invalidate (WI) protocols
Replication versus migration
strategies
๏‚ž Replication strategy:
๏‚— No replication
๏‚— Replication
๏‚ž Migration strategy
๏‚— No migration
๏‚— Migration
๏‚ž Non Replicated and Non
Migrating Block- NRNMB.
๏‚ž Non Replicated, Migrating
Block- NRMB.
๏‚ž Replicated, Migrating Block-
RMB.
๏‚ž Replicated Non Migrating
Block-RNMB.
Case studies
๏‚ž Munin
๏‚ž Linda
๏‚ž Midway
Munin
Linda-1
Linda-2
Linda-3
D/F Munin and Midway:

More Related Content

What's hot

File models and file accessing models
File models and file accessing modelsFile models and file accessing models
File models and file accessing modelsishmecse13
ย 
Load balancing in Distributed Systems
Load balancing in Distributed SystemsLoad balancing in Distributed Systems
Load balancing in Distributed SystemsRicha Singh
ย 
CS9222 ADVANCED OPERATING SYSTEMS
CS9222 ADVANCED OPERATING SYSTEMSCS9222 ADVANCED OPERATING SYSTEMS
CS9222 ADVANCED OPERATING SYSTEMSKathirvel Ayyaswamy
ย 
resource management
  resource management  resource management
resource managementAshish Kumar
ย 
Clock synchronization in distributed system
Clock synchronization in distributed systemClock synchronization in distributed system
Clock synchronization in distributed systemSunita Sahu
ย 
Distributed shared memory shyam soni
Distributed shared memory shyam soniDistributed shared memory shyam soni
Distributed shared memory shyam soniShyam Soni
ย 
Distributed File Systems
Distributed File Systems Distributed File Systems
Distributed File Systems Maurvi04
ย 
Chap 4
Chap 4Chap 4
Chap 4suks_87
ย 
Communication primitives
Communication primitivesCommunication primitives
Communication primitivesStudent
ย 
Fault tolerance in distributed systems
Fault tolerance in distributed systemsFault tolerance in distributed systems
Fault tolerance in distributed systemssumitjain2013
ย 
Distributed Shared Memory
Distributed Shared MemoryDistributed Shared Memory
Distributed Shared MemoryPrakhar Rastogi
ย 
8. mutual exclusion in Distributed Operating Systems
8. mutual exclusion in Distributed Operating Systems8. mutual exclusion in Distributed Operating Systems
8. mutual exclusion in Distributed Operating SystemsDr Sandeep Kumar Poonia
ย 
Distributed system
Distributed systemDistributed system
Distributed systemSyed Zaid Irshad
ย 
distributed Computing system model
distributed Computing system modeldistributed Computing system model
distributed Computing system modelHarshad Umredkar
ย 
Synchronization in distributed systems
Synchronization in distributed systems Synchronization in distributed systems
Synchronization in distributed systems SHATHAN
ย 
Deadlock in Distributed Systems
Deadlock in Distributed SystemsDeadlock in Distributed Systems
Deadlock in Distributed SystemsPritom Saha Akash
ย 
Operating system paging and segmentation
Operating system paging and segmentationOperating system paging and segmentation
Operating system paging and segmentationhamza haseeb
ย 
Distributed data processing
Distributed data processingDistributed data processing
Distributed data processingAyisha Kowsar
ย 
Distributed deadlock
Distributed deadlockDistributed deadlock
Distributed deadlockMd. Mahedi Mahfuj
ย 

What's hot (20)

File models and file accessing models
File models and file accessing modelsFile models and file accessing models
File models and file accessing models
ย 
Load balancing in Distributed Systems
Load balancing in Distributed SystemsLoad balancing in Distributed Systems
Load balancing in Distributed Systems
ย 
CS9222 ADVANCED OPERATING SYSTEMS
CS9222 ADVANCED OPERATING SYSTEMSCS9222 ADVANCED OPERATING SYSTEMS
CS9222 ADVANCED OPERATING SYSTEMS
ย 
resource management
  resource management  resource management
resource management
ย 
Clock synchronization in distributed system
Clock synchronization in distributed systemClock synchronization in distributed system
Clock synchronization in distributed system
ย 
Distributed shared memory shyam soni
Distributed shared memory shyam soniDistributed shared memory shyam soni
Distributed shared memory shyam soni
ย 
Distributed File Systems
Distributed File Systems Distributed File Systems
Distributed File Systems
ย 
Chap 4
Chap 4Chap 4
Chap 4
ย 
Communication primitives
Communication primitivesCommunication primitives
Communication primitives
ย 
Message passing in Distributed Computing Systems
Message passing in Distributed Computing SystemsMessage passing in Distributed Computing Systems
Message passing in Distributed Computing Systems
ย 
Fault tolerance in distributed systems
Fault tolerance in distributed systemsFault tolerance in distributed systems
Fault tolerance in distributed systems
ย 
Distributed Shared Memory
Distributed Shared MemoryDistributed Shared Memory
Distributed Shared Memory
ย 
8. mutual exclusion in Distributed Operating Systems
8. mutual exclusion in Distributed Operating Systems8. mutual exclusion in Distributed Operating Systems
8. mutual exclusion in Distributed Operating Systems
ย 
Distributed system
Distributed systemDistributed system
Distributed system
ย 
distributed Computing system model
distributed Computing system modeldistributed Computing system model
distributed Computing system model
ย 
Synchronization in distributed systems
Synchronization in distributed systems Synchronization in distributed systems
Synchronization in distributed systems
ย 
Deadlock in Distributed Systems
Deadlock in Distributed SystemsDeadlock in Distributed Systems
Deadlock in Distributed Systems
ย 
Operating system paging and segmentation
Operating system paging and segmentationOperating system paging and segmentation
Operating system paging and segmentation
ย 
Distributed data processing
Distributed data processingDistributed data processing
Distributed data processing
ย 
Distributed deadlock
Distributed deadlockDistributed deadlock
Distributed deadlock
ย 

Similar to 6.distributed shared memory

Symmetric multiprocessing and Microkernel
Symmetric multiprocessing and MicrokernelSymmetric multiprocessing and Microkernel
Symmetric multiprocessing and MicrokernelManoraj Pannerselum
ย 
Parallel Processing (Part 2)
Parallel Processing (Part 2)Parallel Processing (Part 2)
Parallel Processing (Part 2)Ajeng Savitri
ย 
Lecture 6
Lecture  6Lecture  6
Lecture 6Mr SMAK
ย 
Lecture 6
Lecture  6Lecture  6
Lecture 6Mr SMAK
ย 
Lecture 6
Lecture  6Lecture  6
Lecture 6Mr SMAK
ย 
Multiprocessor Scheduling
Multiprocessor SchedulingMultiprocessor Scheduling
Multiprocessor SchedulingKhadija Saleem
ย 
Shared memory Parallelism (NOTES)
Shared memory Parallelism (NOTES)Shared memory Parallelism (NOTES)
Shared memory Parallelism (NOTES)Subhajit Sahu
ย 
message passing vs shared memory
message passing vs shared memorymessage passing vs shared memory
message passing vs shared memoryHamza Zahid
ย 
Wiki 2
Wiki 2Wiki 2
Wiki 2Sid Hegde
ย 
Overview of Distributed Systems
Overview of Distributed SystemsOverview of Distributed Systems
Overview of Distributed Systemsvampugani
ย 
Parallel Processing Presentation2
Parallel Processing Presentation2Parallel Processing Presentation2
Parallel Processing Presentation2daniyalqureshi712
ย 
Multiprocessor
MultiprocessorMultiprocessor
MultiprocessorKamal Acharya
ย 
Distributed Shared Memory Systems
Distributed Shared Memory SystemsDistributed Shared Memory Systems
Distributed Shared Memory SystemsAnkit Gupta
ย 
Non-Uniform Memory Access ( NUMA)
Non-Uniform Memory Access ( NUMA)Non-Uniform Memory Access ( NUMA)
Non-Uniform Memory Access ( NUMA)Nakul Manchanda
ย 
Computer architecture multi processor
Computer architecture multi processorComputer architecture multi processor
Computer architecture multi processorMazin Alwaaly
ย 

Similar to 6.distributed shared memory (20)

Symmetric multiprocessing and Microkernel
Symmetric multiprocessing and MicrokernelSymmetric multiprocessing and Microkernel
Symmetric multiprocessing and Microkernel
ย 
Chapter 10
Chapter 10Chapter 10
Chapter 10
ย 
Parallel Processing (Part 2)
Parallel Processing (Part 2)Parallel Processing (Part 2)
Parallel Processing (Part 2)
ย 
Pram model
Pram modelPram model
Pram model
ย 
Parallel processing Concepts
Parallel processing ConceptsParallel processing Concepts
Parallel processing Concepts
ย 
Lecture 6
Lecture  6Lecture  6
Lecture 6
ย 
Lecture 6
Lecture  6Lecture  6
Lecture 6
ย 
Lecture 6
Lecture  6Lecture  6
Lecture 6
ย 
Multiprocessor Scheduling
Multiprocessor SchedulingMultiprocessor Scheduling
Multiprocessor Scheduling
ย 
Shared memory Parallelism (NOTES)
Shared memory Parallelism (NOTES)Shared memory Parallelism (NOTES)
Shared memory Parallelism (NOTES)
ย 
message passing vs shared memory
message passing vs shared memorymessage passing vs shared memory
message passing vs shared memory
ย 
Wiki 2
Wiki 2Wiki 2
Wiki 2
ย 
Overview of Distributed Systems
Overview of Distributed SystemsOverview of Distributed Systems
Overview of Distributed Systems
ย 
22CS201 COA
22CS201 COA22CS201 COA
22CS201 COA
ย 
Parallel Processing Presentation2
Parallel Processing Presentation2Parallel Processing Presentation2
Parallel Processing Presentation2
ย 
Multiprocessor
MultiprocessorMultiprocessor
Multiprocessor
ย 
Distributed Shared Memory Systems
Distributed Shared Memory SystemsDistributed Shared Memory Systems
Distributed Shared Memory Systems
ย 
Non-Uniform Memory Access ( NUMA)
Non-Uniform Memory Access ( NUMA)Non-Uniform Memory Access ( NUMA)
Non-Uniform Memory Access ( NUMA)
ย 
Cache memory
Cache memoryCache memory
Cache memory
ย 
Computer architecture multi processor
Computer architecture multi processorComputer architecture multi processor
Computer architecture multi processor
ย 

Recently uploaded

Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...roncy bisnoi
ย 
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Christo Ananth
ย 
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...SUHANI PANDEY
ย 
chapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringchapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringmulugeta48
ย 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...ranjana rawat
ย 
Thermal Engineering-R & A / C - unit - V
Thermal Engineering-R & A / C - unit - VThermal Engineering-R & A / C - unit - V
Thermal Engineering-R & A / C - unit - VDineshKumar4165
ย 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...Call Girls in Nagpur High Profile
ย 
UNIT-III FMM. DIMENSIONAL ANALYSIS
UNIT-III FMM.        DIMENSIONAL ANALYSISUNIT-III FMM.        DIMENSIONAL ANALYSIS
UNIT-III FMM. DIMENSIONAL ANALYSISrknatarajan
ย 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)simmis5
ย 
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Bookingdharasingh5698
ย 
University management System project report..pdf
University management System project report..pdfUniversity management System project report..pdf
University management System project report..pdfKamal Acharya
ย 
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptxBSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptxfenichawla
ย 
Extrusion Processes and Their Limitations
Extrusion Processes and Their LimitationsExtrusion Processes and Their Limitations
Extrusion Processes and Their Limitations120cr0395
ย 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdfSuman Jyoti
ย 
Online banking management system project.pdf
Online banking management system project.pdfOnline banking management system project.pdf
Online banking management system project.pdfKamal Acharya
ย 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfJiananWang21
ย 
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
ย 
Call Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort ServiceCall Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service9953056974 Low Rate Call Girls In Saket, Delhi NCR
ย 

Recently uploaded (20)

Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
ย 
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
ย 
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
ย 
chapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringchapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineering
ย 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
ย 
Thermal Engineering-R & A / C - unit - V
Thermal Engineering-R & A / C - unit - VThermal Engineering-R & A / C - unit - V
Thermal Engineering-R & A / C - unit - V
ย 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
ย 
UNIT-III FMM. DIMENSIONAL ANALYSIS
UNIT-III FMM.        DIMENSIONAL ANALYSISUNIT-III FMM.        DIMENSIONAL ANALYSIS
UNIT-III FMM. DIMENSIONAL ANALYSIS
ย 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)
ย 
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
ย 
University management System project report..pdf
University management System project report..pdfUniversity management System project report..pdf
University management System project report..pdf
ย 
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptxBSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
ย 
Extrusion Processes and Their Limitations
Extrusion Processes and Their LimitationsExtrusion Processes and Their Limitations
Extrusion Processes and Their Limitations
ย 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
ย 
(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7
(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7
(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7
ย 
Online banking management system project.pdf
Online banking management system project.pdfOnline banking management system project.pdf
Online banking management system project.pdf
ย 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdf
ย 
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
ย 
Roadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and RoutesRoadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and Routes
ย 
Call Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort ServiceCall Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
Call Girls in Ramesh Nagar Delhi ๐Ÿ’ฏ Call Us ๐Ÿ”9953056974 ๐Ÿ” Escort Service
ย 

6.distributed shared memory

  • 1.
  • 2. Basic Concepts Of DSM: ๏‚ž A DSM system provides a logical abstraction of shared memory which is built using a set of interconnected nodes having physically distributed memories.
  • 3. Types of DSMs ๏‚ž Hardware level DSM ๏‚ž Software level DSM ๏‚ž Hybrid level DSM
  • 4. Advantages of DSM ๏‚ž Simple abstraction ๏‚ž Improved portability of distributed application programs ๏‚ž Provides better performance in some applications ๏‚ž Large memory space at no extra cost ๏‚ž Better than message passing systems
  • 5. Comparison of IPC paradigms ๏ƒ˜ DFS ๏‚— Single shared address space ๏‚— Communicate, synchronize using load / store ๏‚— Can support message passing ๏ƒ˜ Message Passing ๏‚— Send / Receive ๏‚— Communication + synchronization ๏‚— Can support shared memory
  • 6. Hardware architectures ๏‚ž On chip memory ๏‚ž Bus based multiprocessor ๏‚ž Ring based multiprocessor ๏‚ž Switched multiprocessor
  • 7. On chip memory In this CPU portion of the chip has a address and data lines that directly connect to the memory portion. Such chips are used in cars appliances and even toys
  • 8. Bus based multiprocessor All CPUs connected to one bus (backplane) Memory and peripherals are accessed via shared bus. System looks the same from any processor.
  • 10. Ring DSM architecture ๏‚ž Shred memory : ๏‚— Private areas ๏‚— Shared areas
  • 12. Switched multiprocessor: ๏‚ž Multiple clusters interconnected by a bus offer better scalability ๏‚ž Example : Dash system
  • 13. NUMA Multiprocessor ๏‚ž Non-uniform memory access (NUMA) is a computer memory design used multiprocessing, where the memory access time depends on the memory location relative to the processor. ๏‚ž Under NUMA, a processor can access its own local memory faster than non-local memory. ๏‚ž The benefits of NUMA are limited to particular workloads, notably on servers where the data is often associated strongly with certain tasks or users.
  • 14. UMA Multiprocessor ๏‚ž Uniform memory access (UMA) is a shared memory architecture used in parallel computers. All the processors in the UMA model share the physical memory uniformly. ๏‚ž In a UMA architecture, access time to a memory location is independent of which processor makes the request or which memory chip contains the transferred data. ๏‚ž Uniform memory access computer architectures are often contrasted with non-uniform memory access (NUMA) architectures. ๏‚ž In the UMA architecture, each processor may use a private cache. Peripherals are also shared in some fashion. ๏‚ž The UMA model is suitable for general purpose and time sharing applications by multiple users. It can be used to speed up the execution of a single large program in time- critical applications.
  • 15. D/f bw two Multiprocessor: BUS BASED MULTIPROCCESORS RING BASED MULTIPROCCESORS ๏‚ž They are tightly coupled with the CPUโ€™S normally in a single rack. ๏‚ž It has separate global memory ๏‚ž Machines here can be much more loosely coupled and this loose coupling can affect their performance. ๏‚ž It has no separate global memory.
  • 16.
  • 17. DSM design issues ๏‚ž Granularity of sharing ๏‚ž Structure of data ๏‚ž Consistency models ๏‚ž Coherence protocols
  • 19. Thrashing: ๏‚ž False sharing ๏‚ž Techniques to reduce thrashing: ๏‚— Application controlled lock ๏‚— Pin the block to a node for specific time ๏‚— Customize algorithm to shared data usage pattern
  • 20. DSM structure ๏‚ž Organization of data items in the shared memory
  • 21. Consistency models //Important. ๏‚ž Refers to how recent the shared memory updates are visible to all the other processes running on different machines
  • 22. Strict consistency ๏‚ž Strongest form of consistency. ๏‚ž All write operations are visible to all processes.
  • 23. Sequential consistency ๏‚ž All processors in the system observe the same ordering of reads and writes which are issued in sequence by the individual processors
  • 24. Causal consistency ๏‚ž Weakening of sequential consistency for better concurrency ๏‚ž Causally related operation is the one which has influenced the other operation
  • 25. PRAM consistency ๏‚ž Pipelined Random Access Memory consistency ๏‚ž Write operations performed by different processes may be seen by different processes in different orders ๏‚ž All write operations of a single process are pipelined ๏‚ž Simple, easy to implement and has good performance.
  • 26. Processor consistency ๏‚ž Adheres to the PRAM consistency ๏‚ž Constraint on memory coherence ๏‚ž Order in which the memory operations are seen by two processors need not be identical, but the order of writes issued by each processor must be preserved
  • 27. Weak consistency ๏‚ž Use a special variable called the synchronization variable. ๏‚ž Very difficult to show and keep track of the changes at time to time.
  • 28. Properties of the weak consistency model: ๏‚ž Access to synchronization variables is sequentially consistent. ๏‚ž Only when all previous writes are completed everywhere, access to synchronizations variable is allowed. ๏‚ž Until all previous accesses to synchronization variables are performed, no read write data access operations will be allowed.
  • 29. Release consistency ๏ถ Uses two synchronization variables (Acquire and Release) ๏ถ Release consistency model uses synchronization mechanism based on barriers.
  • 32. Entry consistency โ€ข Use acquire and release at the start and end of each critical section, respectively. โ€ข Each ordinary shared variable is associated with some synchronization variable such as a lock or barrier. โ€ข Entry consistency (EC) is similar to LRC but more relaxed; shared data is explicitly associated with synchronization primitives and is made consistent when such an operation is performed
  • 33. Scope consistency ๏‚ž A scope is a limited view of memory with respect to which memory references are performed
  • 34. Comparison of consistency models-1 ๏‚ž Most common: sequential consistency model .
  • 35. Comparison of Consistency models-2 ๏‚ž Based on efficiency and programmability
  • 36. Coherence protocols ๏‚ž Specifies how the rules set by the memory consistency model are to be implemented
  • 37. Coherence algorithms ๏‚ž Maintain consistency among replicas
  • 38. Multiple Reader/ Multiple Writer algorithm ๏‚ž Uses twin and diff creation technique
  • 39. Write Protocols for consistency ๏‚ž Write Update (WU) ๏‚ž Write Invalidate (WI) protocols
  • 40. Replication versus migration strategies ๏‚ž Replication strategy: ๏‚— No replication ๏‚— Replication ๏‚ž Migration strategy ๏‚— No migration ๏‚— Migration ๏‚ž Non Replicated and Non Migrating Block- NRNMB. ๏‚ž Non Replicated, Migrating Block- NRMB. ๏‚ž Replicated, Migrating Block- RMB. ๏‚ž Replicated Non Migrating Block-RNMB.
  • 41. Case studies ๏‚ž Munin ๏‚ž Linda ๏‚ž Midway
  • 42. Munin
  • 46. D/F Munin and Midway: