|
contents | software | |||||||
| xkoto Launches GRIDSCALE for Microsoft SQL Server xkoto announces GRIDSCALE for Microsoft SQL Server, a product that improves application performance, scalability and reliability in SQL Server environments. GRIDSCALE for Microsoft SQL Server is an active-active solution that virtualizes the database infrastructure. It enables businesses to distribute application loads horizontally across multiple instances of SQL Server, resulting in better reliability and performance than that of more costly and complex legacy clustering solutions. With GRIDSCALE for SQL Server, businesses can eliminate planned and unplanned outages, rapidly scale application load across SQL Servers, and implement cost-effective disaster recovery solutions. With GRIDSCALE, database administrators can perform hardware upgrades and software updates without having an impact on system performance or uptime. Plus, GRIDSCALE eliminates the limitations of clustering configurations, such as shared storage, partitioning and mirroring. xkoto underwent extensive performance and functional testing of GRIDSCALE for SQL Server in one of Microsoft's technology centers. Results showed near-linear scalability for read-intensive workloads, such as those used for online commerce applications. GRIDSCALE also received high marks on availability tests, which simulated hardware failures, network outages and site-wide disaster. Unlike traditional clustering, which relies on multiple databases tied to shared storage, GRIDSCALE employs a shared-nothing architecture with a dynamic number of database servers for continuous availability. Each database server operates completely independent of the others, maintaining its own complete and consistent copy of the database. The GRIDSCALE server appears as a single, fully functional database to all applications, which are unaware that any one of several database servers may be used to fulfill their requests. GRIDSCALE for SQL Server initially supports ODBC, JDBC, and .NET connectivity. write your comments about the article :: © 2008 Computing News :: home page |