How Do I Improve Network Performance?

Network optimization refers to making the network performance reach the best balance point we need through various hardware or software technologies! Network optimization is also SEO. Network optimization is mainly divided into two aspects of equipment and services, of which planning, evaluation, and optimization belong to the service industry;

Network Optimization

at the same time,

SmartRoute network optimization diagram

Under normal circumstances, Netcom users access the telecommunication network through the interconnection center (the path in blue in the figure). Due to insufficient bandwidth of the interconnection center, the network delay is too large and the access speed is slow. After using the SmartRoute network optimization service, the traffic of Netcom users accessing the telecommunication network will be forwarded to Etrunk's SmartRoute router first, and then the telecommunication network will be accessed through the Etrunk telecommunication port (green path in the figure). Because Etrunk is directly connected to the backbone network of telecommunications and China Netcom, and has a sufficient network bandwidth, the delay of the line is small, and the number of routing nodes is small, which greatly improves the access speed.

Features

First-line connectivity: As long as you have a single Internet connection (both Telecom and Netcom), you can easily gain the speed advantage of the dual lines of Telecom and Netcom, breaking through the gap between networks, and making your business and information flow unimpeded.

Automatic update: Automatically update routes without manual intervention, eliminating tedious and complicated daily maintenance.

Full control: Professional-level product positioning, customers have complete control over specific routes, and can make in-depth routing adjustments.

Security and stability: When the optimized path is unavailable, the original path is automatically selected, which will not affect the normal network usage, so you can rest easy!

In the current network state, users often encounter problems and dilemmas that have a negative impact on network use and resources, such as broadband congestion, poor application performance, worms, DDoS, and malicious intrusions. Network optimization functions are aimed at existing It can supplement equipment and network problems such as firewalls, security and intrusion detection, load balancing, bandwidth management, network anti-virus, etc., and can perform parameter collection and data analysis by accessing hardware and software operations to find out the reasons that affect network quality. Through technical means or adding corresponding hardware equipment and adjusting the method to achieve the optimal operating state of the network, the network resources can obtain the best benefits. At the same time, understand the growth trend of the network and provide better solutions to achieve network application performance acceleration, security content management, security event management, user management, network resource management and optimization, desktop system management, and monitoring, measurement, and tracking of traffic patterns. Products that analyze, manage, and improve the performance of application transmission over the WAN. The product mainly includes three categories: network resource manager, application performance accelerator, and web performance accelerator. It optimizes the network according to different needs and functional requirements.

Network optimization devices also have functions such as supported protocols, network integration functions (cascade mode, bypass mode), device monitoring functions, compressed data statistics, QOS, bandwidth management, data export, application reports, and continuous work in the event of a failure Or upgrade via the network.

Step 1: Familiarize yourself with the industry

No matter what level you are

1. The network design must adapt to the development of technology and requirements, taking into account the actual needs and long-term considerations, and at the same time have clear stage goals and countermeasures, so that investment is inheritable, the network has scalability and business upgrade capabilities .

2.The network structure is clear, with high reliability, openness and expandability, using optical fiber as the backbone network, and using various connection methods, fewer exchange stations, fewer sites, scattered service nodes, and service nodes have the ability to provide comprehensive services. Ability, network connection is transparent.

3. When formulating a network optimization plan, it should be committed to adopting a comprehensive solution, while paying attention to the comprehensive cost / performance ratio to avoid repeated investment in the future.

Modern telecommunication network architecture can be divided into three levels, namely: backbone network, branch network, and end network. As shown in the figure, a more typical comprehensive package solution for modern telecommunication networks:

Network optimization wireless network

Introduction

Wireless network optimization is to find out the reasons that affect the network quality by means of traffic data analysis, field test data collection, parameter analysis, hardware inspection and other means on the existing network, and through parameter modification, network structure adjustment, and equipment Adjust the configuration and adopt some technical means (using MRP planning methods, etc.) to ensure the high-quality operation of the system, make the existing network resources obtain the best benefits, and obtain the maximum benefits with the most economical investment.

Network optimization is the abbreviation of "wireless network optimization", which means that after the communication network is completed, various optimizations (including software, hardware, configuration, etc.) are performed on this basis; network regulation is the abbreviation of "network planning", which refers to the construction of communication networks Prior to planning network construction based on network construction goals, user needs, and local actual conditions.

"Net Excellent" also refers to those who are engaged in wireless network optimization. Because they need to travel for a long time, the practitioners are generally young college graduates, more men, and shorter working hours. Most people will choose to leave after the contract expires. In general, there are more people in the industry within 5 years, and fewer people over 5 years. Therefore, the net salary is higher and belongs to high-paying occupations. Among them, the classification is obvious. About four thousand. The working hours of Wangyou are relatively stable. As in other industries, there are regular weekends and holidays, but due to the nature of work, weekends and overtime often occur, but they will also be adjusted during the project. The work is DT test and analysis, CQT test and analysis, antenna feed adjustment (RF optimization), background parameter modification and adjustment, KPI monitoring and improvement, poor cell processing, etc. Its main purpose is to understand the current wireless network coverage. The network optimization professional method is optimized to enable users to talk normally while maintaining mobility while using mobile phones, without short tones, dropped calls, and unconnected problems. At present, 3G is popular in China, and there are increasing numbers of network optimization companies, and there is a shortage of network optimization personnel. Therefore, network optimization is also a rising industry.

Network optimization is a profession that refers to wireless network optimization. It also refers to people engaged in this industry. Wireless network optimization engineers are divided into 2G and hot 3G network optimization. Network optimization is a profession that requires both technicality and experience to coexist. The higher the level of technology, the more work experience, and the more natural remuneration. Many of them need to be learned and accumulated in the project. As far as the market is concerned, the benefits of netizens are uneven. Some work hours are not short, but the level of technology is limited. They always do some junior work. Wages are also stagnant. Some work hours are not long. It is growing very fast, and wages are growing naturally.

In short, the job of network optimization will be very hard at the beginning, and the quality requirements are high, not only must have professional knowledge and technology, but also have a certain psychological and stress resistance. Fresh graduates can do a few years of exercise, but they should be prepared and determined based on their own personality characteristics.

Network optimization optimization scheme

There are many methods for network optimization. In the initial stage of network optimization, the network adjustment plan is often formulated through the analysis of OMC-R data and the results of drive tests. After using the process of Figure 1 for several cycles, the network quality has improved significantly. However, it is difficult to find and solve the problem only by using the above methods. At this time, it is usually combined with user complaints and CQT test methods to find the problem, combined with signaling tracking analysis method, traffic statistics analysis method and drive test analysis method to analyze and find the source of the problem . In actual optimization, especially the analysis of OMC-R traffic statistics report, supplemented by the No. 7 signaling instrument for A interface or Abis interface tracking analysis, is the most commonly used method for network optimization. The most important step in network optimization is how to find problems. Here are some common methods:

1. Traffic statistics analysis method: OMC traffic statistics is an important way to understand network performance indicators. It reflects the actual operating status of wireless networks. It is the main basis for optimizing the basic data for most of our networks. The classified parameters are processed to form a report that is convenient for analyzing network quality. Through the indicators in the traffic statistics report (call success rate, call drop rate, handover success rate, traffic per time slot, wireless channel availability rate, voice channel blocking rate and signaling channel availability rate, call drop rate And blocking rate, etc.), you can understand the traffic distribution and changes of wireless base stations, and then find abnormalities. Combined with other methods, you can analyze the unreasonable network logical or physical parameter settings, unreasonable network structure, and traffic volume. Problems such as unevenness, frequency interference, and hardware failure. At the same time, it is also possible to formulate unified parameter templates for different regions in order to find problems faster, and by adjusting the parameters of specific cells or the entire network, the indicators of each cell of the system can be improved, thereby improving the system of the entire network. index.

2. DT (driving test): When the car is driving at a certain speed, test whether the signal strength in the car meets the requirements of normal calls, and whether there is congestion, interference, or dropped calls. Generally, the duration of each call is set in DT according to the needs. It is divided into long calls (the duration is unlimited, until the call is dropped) and short calls (usually about 60 seconds, based on the average user call duration). Condition adjustment time), in order to ensure the authenticity of the test, the general speed should not exceed 40 km / h. The drive test analysis method mainly analyzes the data and measurement coverage of the air interface. Through the DT test, you can know: the distribution and coverage of base stations, whether there is a blind spot; the handover relationship, the number of handovers, and whether the battery is switched properly; whether the downlink has the same frequency Adjacent channel interference; whether there is an island effect; whether there is a ping-pong effect; whether there is a distance effect; whether the sector is misaligned; whether the antenna downtilt, azimuth and antenna height are reasonable; analyze the call connection to find out the call is unreachable and dropped The reasons provide basis for formulating network optimization schemes and implementing network optimization.

3 CQT (call quality test or fixed-point network quality test): Select multiple test points in the service area, make a certain number of dial-up calls, and reflect the network quality from the user's perspective. The test points are generally selected in places where communications are concentrated, such as hotels, airports, stations, important departments, office buildings, and assembly venues. It is an important supplement to DT testing. It is usually a simple form of field strength test that can also complete the testing of more complex areas such as deep indoor coverage and high-rise buildings where DT cannot test.

4 User complaints: Learn about network quality through user complaints. Especially when the network optimization has reached a certain stage, it is difficult to find individual problems in the network through drive tests or data analysis. At this time, the problems discovered through the calls of ubiquitous users make us better understand the status of network services. Combined with a field strength test or a simple CQT test, we can discover the source of the problem. This method has the characteristics of timely discovery of problems and strong pertinence.

5. Signaling analysis method: Signaling analysis is mainly to track and analyze the data of the A interface and Abis interface of the site in question. By analyzing the data collected on the A interface, you can find that the switching office has incomplete data (missing the switching relationship), signaling load, hardware failure (finding problematic relays or time slots), and uneven traffic (some data definition errors, Link is down, etc.). Through the collection and analysis of Abis interface data, it mainly analyzes the LAY3 signaling recorded by the measuring instrument. At the same time, according to the signal quality distribution diagram, frequency interference detection diagram, and reception level distribution diagram, combined with the occupied length of the signaling channel or voice channel The analysis can find out the problem of excessive path loss on the uplink and downlink, and can also find problems such as cell coverage, some wireless interference, and hidden hardware failure.

6. Analysis of automatic drive test system: The automatic drive test terminal installed on the mobile vehicle can monitor the road coverage and communication quality throughout the process. Because the terminal can automatically return a large number of signaling messages and measurement reports to the monitoring center, it can detect problems in time and analyze the locations where the problems occur, which has a strong timeliness. The method used is the same as 5.

In practice, these methods are mutually reinforcing and mutually reinforcing. The optimization of GSM wireless network is to use the above-mentioned methods to pass the performance statistics test data analysis formulation and implementation around the indicators such as the connection rate, call drop rate, congestion rate, voice quality and handover success rate, and super idle cells and worst cells. The optimization scheme system adjustment re-establishing the optimization goal the spiral loop of performance statistics test rises to achieve the purpose of significantly improving the network quality.

Network optimization wireless network optimization factors

There are several key factors to consider, including:

· Network infrastructure capable of handling a growing number of mobile devices · Visibility of overall network traffic, usage, and performance, including application performance per device Capabilities · Have policies to support the necessary mobile strategies while ensuring the security and reliability of their performance ^[2]

Network optimization related books

Basic Information

Title: Network Optimization (2nd Edition)

Author: Xie Venus, Xingwen training, Wang Zhenbo

Pricing: 19 yuan

Publication date: 2009-7-1

Publisher: Tsinghua University Press

Book Introduction

This book systematically introduces the basic models and algorithms of network optimization, including the basic ideas of constructing these algorithms, and some specific implementation skills and complexity analysis of corresponding algorithms on computers.

The book is composed of 7 chapters: Chapter 1 is an overview, Chapter 2 introduces some basic knowledge about algorithms, and Chapters 3 to 7 discuss tree problems, shortest path problems, maximum flow problems, minimum cost flow problems, and matching, respectively. Question. Each chapter also arranges some exercises.

This book can be used as a textbook for senior students and graduate students in mathematics, applied mathematics, operations research, management science, systems science, information science, computer science, and engineering, and it can also be used as a reference for scholars and technicians in other related disciplines.

table of Contents

Preface I

Preface III Chapter 1 Introduction 1

1.1 Example of network optimization problem 1

1.2 Graph and Network 2

1.2.1 Basic concepts of directed graphs and networks 2

1.2.2 Basic Concepts of Undirected Graphs and Undirected Networks 5

1.3 Graph and Network Data Structure 6

1.3.1 Adjacent Matrix Representation 6

1.3.2 Association Matrix Representation 7

1.3.3 Arc Table Notation 7

1.3.4 Adjacency Table Representation 8

1.3.5 Star Notation 8

1.4 The concept of computational complexity 11

1.4.1 Combination Optimization Problems 11

1.4.2 Polynomial Time Algorithm 13

1.4.3 Polynomial Problems 16

Chapter 18 Algorithm Basics 19

2.1 NP, NPC and NP-hard concepts 19

2.1.1 Problems, Examples, and Input Scale 19

2.1.2 Determination Question 21

2.1.3 Nondeterministic Polynomial Problem Class (NP) 22

2.1.4 NP Complete Problem Class (NPC) 25

2.2 Algorithm Design and Analysis 29

2.2.1 Greedy Algorithm

2.2.2 Dynamic Programming

2.2.3 Linear Programming Method--Full Unitary Module Matrix

2.2.4 Dichotomy

2.2.5 Network Search Algorithms

2.3.Summary

Exercises 38 Chapter 3 Minimal Trees and Minimal Tree Graphs 41

3.1 Basic Concepts of Trees

3.2 Minimum Tree Algorithm

3.2.1 Kruskal's Algorithm

3.2.2 The Prim Algorithm

3.2.3 Sollin Algorithm

3.3 Minimal Tree Graph 49

3.4 maximum branch 53

Exercise 56 Chapter 4 Shortest Path Problem 58

4.1 Mathematical description of the shortest problem

4.2 Acyclic network and positive cost network: labeling algorithm 60

4.2.1 Bellman Equation 60

4.2.2 Acyclic Networks

4.2.3 Positive Expense Network

4.3 General Cost Network: Label Correction Algorithm 65

4.3.1 Bellman-Ford Algorithm 65

4.3.2 General Label Correction Algorithms

4.3.3 Floyd-Warshall Algorithm

Exercise 70 Chapter 5 Maximum Flow Problem 73

5.1 Mathematical description of the maximum flow problem

5.1.1 Flows in the Network

5.1.2 The Maximum Flow Problem

5.1.3 Augmentation Road Theorem

5.2 Augmentation Road Algorithm

5.2.1 Ford-Fulkerson Labeling Algorithm

5.2.2 Residual Networks

5.2.3 Maximum Capacity Augmentation Algorithm 82

5.2.4 Capacity Rescaling Algorithms

5.3 Shortest Augmentation Algorithm 83

5.3.1 Distance labels

5.3.2 Shortest Augmentation Algorithm 85

5.3.3 Complexity Analysis

5.4 General Pre-Flow Propulsion Algorithms

5.4.1 General pre-flow advancement algorithm

5.4.2 Complexity Analysis

5.5 Highest Label Pre-Flow Advance Algorithm 94

5.5.1 The highest label pre-flow advancement algorithm

5.5.2 Algorithm Complexity Analysis

5.6 Maximum Flow Algorithm on Unit Capacity Networks

5.6.1 Maximum Flow Algorithm on Unit Capacity Networks

5.6.2 Maximum Flow Algorithm on Simple Networks with Unit Capacity

Exercise 98 Chapter 6 Minimum Cost Flow Problem 102

6.1 Mathematical description of the minimum cost flow problem

6. 1. 1 Minimum Fee Flow Problem 102

6. 1.2 Special Cases and Extensions of the Minimum Cost Flow Model

6.2 Loop elimination algorithm and minimum cost path algorithm

6. 2. 1 Loop elimination algorithm

6.2.2 The least cost path algorithm 108

6. 3 Primal-Dual Algorithms

6. 3. 1 Duality Problems and Complementary Relaxation Conditions 111

6. 3. 2 Primal-Dual Algorithms

6.4 Flaw Algorithm

6. 5 Relaxation Algorithms

6. 6 Network Simplex Algorithms

6. 6. 1 general idea of the algorithm 128

6. 6. 2 Methods to deal with degradation 131

6. 6. 3 Initial Basic Feasible Solutions

6.6.4 Cases with bounded capacity 133

Exercise 136 Chapter 7 Matching Problems 141

7.1 Mathematical description of the matching problem

7. 2 Two Cardinality Matching Problems 144

7. 2. 1 Augmentation Road Algorithm 144

7.2.2 Applying the Maximum Flow Algorithm on a Simple Network

7. 3 Non-bipartite cardinality matching problems 147

7. 4 Bipartite Weighting Matching Problem 151

7. 5 Non-bipartite weighting matching problem 152

Index of practice questions 162 and English keywords 165

References 170