Whether your database is the popular open-source MySQL or another implementation of SQL, let’s face it—SQL was simply not designed to handle today’s large-scale Web applications with millions of users. As your data volume and number of transactions increase, the response time increases exponentially to the point where your system may no longer meet its performance objectives. Buying a bigger server to scale up only buys you some time until you are forced to scale out your database servers to improve capacity and performance.
Using master-slave replication to spread the workload across multiple servers is the simplest approach to the problem. This solution can speed performance to a point, allowing read-intensive processing to be offloaded to slave servers. However, a single master server for writes can quickly create a bottleneck. Using sharding to carve up your SQL database, so that each server houses a subset of the total database, is complex, costly, and risky. It also makes future application changes difficult, and only works until you hit the next wall in capacity and performance—requiring more sharding.
Solution: Clustrix Clustered Database System
The Clustrix solution allows you to quickly and cost-effectively achieve massive scalability and linear growth for your Web applications without giving up robust relational functionality and transactional consistency guarantees. With minor changes to your applications, you gain a massively scalable, relational, fault-tolerant database to serve you needs now and into the future. The only alternatives to scale your database are sharding and converting to a key-value store architecture—both require major application changes and produce a database that falls short in many ways.
A Clustrix Clustered Database System is delivered as a complete and easily-deployed CLX appliance consisting of CLX 4000 Series nodes configured as a cluster. To grow capacity and performance on an incremental, as-needed basis, simply add another node. With each additional node, transactions per second, I/O throughput, capacity and cache scale linearly. Whether there are a few or hundreds of nodes in a cluster, the Clustrix cluster appears to applications as a single database.
For Other SQL Users
Challenge: Scaling Capacity and Performance
Whether your database is the popular open-source MySQL or another implementation of SQL, let’s face it—SQL was simply not designed to handle today’s large-scale Web applications with millions of users. As your data volume and number of transactions increase, the response time increases exponentially to the point where your system may no longer meet its performance objectives. Buying a bigger server to scale up only buys you some time until you are forced to scale out your database servers to improve capacity and performance.
Using master-slave replication to spread the workload across multiple servers is the simplest approach to the problem. This solution can speed performance to a point, allowing read-intensive processing to be offloaded to slave servers. However, a single master server for writes can quickly create a bottleneck. Using sharding to carve up your SQL database, so that each server houses a subset of the total database, is complex, costly, and risky. It also makes future application changes difficult, and only works until you hit the next wall in capacity and performance—requiring more sharding.
Solution: Clustrix Clustered Database System
The Clustrix solution allows you to quickly and cost-effectively achieve massive scalability and linear growth for your Web applications without giving up robust relational functionality and transactional consistency guarantees. With minor changes to your applications, you gain a massively scalable, relational, fault-tolerant database to serve you needs now and into the future. The only alternatives to scale your database are sharding and converting to a key-value store architecture—both require major application changes and produce a database that falls short in many ways.
A Clustrix Clustered Database System is delivered as a complete and easily-deployed CLX appliance consisting of CLX 4000 Series nodes configured as a cluster. To grow capacity and performance on an incremental, as-needed basis, simply add another node. With each additional node, transactions per second, I/O throughput, capacity and cache scale linearly. Whether there are a few or hundreds of nodes in a cluster, the Clustrix cluster appears to applications as a single database.