Please enable JavaScript.
Coggle requires JavaScript to display documents.
Real-Time Communication in Packet-Switched Networks - Coggle Diagram
Real-Time Communication in Packet-Switched Networks
Introduction
QoS
Network
Perishable: messages may be useless to the application if delivered after the deadline
Messages arriving too early may be harmful to the system as well
Soft real-time application can tolerate some amount of lost messages, hard real-time application have no tolerance
Unsatisfactory Approaches to Real-Time Communication
Low jitter
Low latency
Ability to easily integrate non-real-time and real-time services
Adaptable to dynamically changing network and traffic conditions
Good performance for large networks and large num- bers of connections
Modest buffer requirements within the network
High effective bandwidth utilization
Low overhead in header bits per packet or cell
Low processing overhead per packet within the network and at the end system.
Characteristics of Real-Time Traffic
General Properties
Data rate
Packet size
Loss tolerance
Characteristics
Protect other applications from a misbehaving client
Differentiate negotiated traffic (which should have guaranteed service) from excess traffic
Behavior
Constant bit rate
Variable bit rate
Periodic with variable packet sizes
Hard-Real-Time Communication
General Remarks
Methods
Scheduler-based: analyzes the packets interactions and check if a deadline can be missed. Allows independently bandwidth, delay and jitter allocation.
Rate-based: QoS is bounded to transmission rate or bandwidth. Simpler to implement.
Real-time Scheduling Theory
Round Robin
Priority-based
Schedulability
: a set of real-time connections is schedulable on a link if it can be guaranteed that no messages in those connections will miss their deadlines on that link
Traffic characterization
QoS
Delay
Jitter
Loss bounds
Leaking Bucket: a kind of buffer
Connection-Level Processing
Admission control: calculates which resources are needed to provide the QoS, check its availability and then reserve them.
Scheduler-based methods
earliest due-date for delay
earliest due-date for Jitter
smallest response time
preemptive cut through
Rate-based methods
Hierarchical round robin
Stop-and-go
Weighted fair queueing
Packet generalized processor sharing
rate-controlled static priority
Per-packet processing
Soft-Real-Time Communication
Application-level characteristics
QoS
Delay
Jitter
Maximum tolerable packet loss due to either buffer overflow or exceeding the delay bound
Leaky bucket
Connection-level issues
Multiple traffic classes and grades of service
Providing statistical guarantees on delays and loss
Source-based approach
Bounding approach
Observation approach
Best effort delivery
Synchronization of soft-real-time traffic
Per-packet processing
Priority policies
Laxity-based policies