为什么你的 SQL Server 需要 Redis.pdf

110

7页

2次

2022-07-28

免费下载

WHITE PAPER

Why Your SQL Server Needs Redis

Roshan Kumar, Redis

CONTENTS

How pung Redis in front of SQL Server can help 2

Redis performance lets you do more with less 3

Redis is all pervasive, even more so with Windows and .NET 3

Deployment Opons 3

Fully managed Redis Enterprise in your Azure Virtual Network (VNet) 3

Redis Enterprise as soware on your Azure server instance 3

Redis use cases boost SQL Server performance 4

1. Caching 4

2. Session stores 4

3. Counng and gamicaon 5

4. Metering and rate-liming calls to legacy servers 5

Other use cases 5

Embrace Redis today 5

SQL Server has been around for a few decades and is one of the most stable and widely used relaonal database

management systems (RDBMS) in the market. Just like any other RDBMS, SQL Server encounters performance issues as

the data it stores grows to larger volumes. SQL Server experts recommend performance tuning that includes seng up

the right indices, correctly sizing TEMPDBs and ensuring maximum degrees of parallelism (MAXDOP), as well as properly

conguring virtual machines (VMs), disk IO, main memory, etc.

Even with all of this ne tuning, SQL Server requires a heavy investment in the physical infrastructure in order to perform

a few thousand operaons per second. However, to stay ahead of the compeon, most enterprises need to operate at

the scale of a few million operaons per second with sub millisecond latency. Today’s compeve business environment

also drives companies to innovate and bring new services to market faster than ever before. But they must do so without

aecng any legacy applicaons.

Figure 1. SQL Server has limited opons to scale

How pung Redis in front of SQL Server can help

If you are innovang new soluons, the two most important quesons you face are:

1. How can we scale our applicaons with limited eort and cost?

2. How can we introduce new features, applicaons and soluons faster?

Redis helps you meet both challenges. First, as an extremely lightweight, in-memory database, Redis performs many more

operaons per second (usually in the order of 100-1000 mes more) with fewer computaonal resources. Second, Redis

enables you to get your apps to market faster. Its built-in data structures allow you to store data in whichever format

your program requires. In many cases you can avoid the burden of translang from a nave data structure to a relaonal

database. If you are using Redis as a scaling layer (such as caching), you can insert Redis between your data access layer and

the database without disrupng your program stack.