Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Navigation  Back up to About Overview 
[+] Expand All
[-] Collapse All

Table of Contents

About This Guide
Documentation Conventions
Related Documentation
Obtaining Documentation
Documentation Feedback
Requesting Technical Support
Performance, Planning, and Tuning
Overview of Performance, Planning, and Tuning of Steel-Belted Radius Carrier
Steel-Belted Radius Carrier Overview
Document Organization
Performance Theory
Performance Theory: Threads
Thread Models
Thread Priority
Locking Threads to a Processor
Performance Theory on Contention
Managing Contention
CPU Utilization
Case 1: More CPU versus More GHz per CPU
Case 2: Virtual CPU versus Physical Cores
Performance Theory on Memory Data Structures
Understanding Performance Metrics
Metrics and Definitions
Components of a Sample Use Case
Average Peak Rates
Spike Rates at Peak
Utilities to Measure the Performance of an Existing System
prtdiag and sosreport
cpuinfo on Linux
Taskset on Linux
Disabling a CPU on Linux
System Statistics
Performance Planning
Understanding SBRC Call Models
Understanding the Memory Utilization Levels
Memory Utilization on Standalone
Memory Utilization on SSR Cluster
Understanding the CPU Requirements
Standalone or Front-End Applications
Back Ends
Load Balancing on RADIUS Front-End Applications and Downstream Proxy Devices
Stateful and RADIUS-Aware
Network Configuration for SSR
Why Latency Kills Performance: D-D and S-D
Subscriber Scaling
Why Are Interims Hard, and When Can They Be Thrown Away?
Handling Downstream Latency and Traffic Spikes
Thread Control
Flood Control
Understanding the SBRC Network Architecture
SBRC Architecture Overview
Understanding “Geodiversity” Between the SSR D Nodes
NDB Cluster Timeout Dependencies
High Throughput Configurations
Risks to Uptime within the Architecture
Debugging Performance and Functionality Problems
Determining the Effective Performance and Peak Capacity
Calculating Performance Scaling on an Existing System
Debugging Performance Issues
CPU Bound
I/O Bound
Contended Systems
Notes on Debugging Performance Issues
Special Topics in Performance
Multifactor Authentication over EAP
Reauthentication Cache
Cipher Suites and Key Sizing
Cipher Suite Scaling
Reducing the Number of Round Trips
Certificate Management
Multiple TLS/TTLS-Only Front-End Applications
EAP-SIM/AKA and authGateway
External Authentication and Accounting
Proxy Radius
Smart Static Accounting
Spooled Accounting
IP Pools through the SSR
Accounting-Off or Accounting-On with AutoStop Enabled
SBRC System-of-Systems Performance Reference
Hardware Used
CPU Utilization
Test Results Summary
Pap Authentication, Accounting Start, Accounting Stop
Pap Authentication, Accounting Start, Accounting Stop—Sessions Preloaded
Accounting Only—Sessions Preloaded, One Start and One Stop, Four D Nodes
Accounting Only—Sessions Preloaded, One Start and One Stop, Two D Nodes
Standalone: Auth/start/stop CPS
Standalone: Accounting for 200 Threads
Standalone: Authentication Only
LDAP Authentication Only
Oracle 11g
LCI Queries against the SSR through SBRC
IP Address Allocation
IP Address Allocation with Eight Threads, Two D Nodes
IP Address Allocation with Eight Threads, Four D Nodes
TTLS Plus Storing Resumption Context
4D node system: M5000
Standalone: M9000
Standalone: T3
SSR and Standalone Performance: E4870 and X5687 CPUs

Modified: 2017-09-26