ScaleOut Powers Distributed LINQ Query with Data-Parallel Computing for .NET Developers

Print Friendly, PDF & Email

ScaleOut_newlogoScaleOut Software, a leading provider of in-memory data grid software, announced a new query capability called “computational query” that combines Microsoft’s LINQ query with the company’s data-parallel computing technology to offer more powerful and flexible query tools for grid-based applications. Now available with ScaleOut Software’s suite of in-memory data grid products, including ScaleOut StateServer® and ScaleOut ComputeServer™, computational query makes it easy for .NET developers and architects to harnesses the full power of the grid’s data-parallel compute engine and dramatically accelerate query processing. It is also available for use with Java on both Linux and Windows systems.

ScaleOut Software has integrated Microsoft’s powerful LINQ-based query capability into its in-memory data grid, allowing applications to perform fully distributed queries for grid-based objects. However, existing query tools, including LINQ, face inherent limitations. While Microsoft LINQ provides an excellent technique for querying data hosted in a distributed, in-memory data grid, it only examines the values of specific object properties. For example, a standard LINQ query in a financial services application can identify which stocks traded more than 100 times in the last day but they would not be able to identify which stocks met criteria described by a statistical analysis of these trades. This lack of flexibility can lead to excessive data returned to the client for processing, and this places a heavy burden on both the client and the network.

ScaleOut Software’s Computational Query

Computational query overcomes these limitations by running a parameterized, user-defined method within the grid to refine the query results. This new capability makes it easy for applications to perform additional query processing directly within the in-memory data grid; the grid handles all of the details of shipping the user’s code to grid servers and managing its execution. By harnessing the power of the in-memory data grid to perform advanced query analysis, computational query accelerates query processing, reduces the client’s workload, and helps eliminate network bottlenecks.

ScaleOut’s C# APIs offers .NET developers access to a distributed implementation of standard LINQ query which:

  • enables grid-based objects to be queried by selected properties and/or tags
  • uses fast, indexed lookup on each grid server to minimize query times
  • runs on all grid servers to scale query performance for large workloads

ScaleOut’s computational query enhances LINQ query with a user-defined “filter” method which:

  • allows applications to extend query semantics by analyzing all properties of an object with a user-defined computation
  • provides automatic, data-parallel computing that analyzes data in place, thereby reducing both networking traffic and client workload
  • is automatically shipped to grid servers, which manage its execution

We are excited to add computational query to our distributed LINQ query support based on customer feedback asking for even richer query capabilities,” said Dr. William L. Bain, founder and CEO of ScaleOut Software. “We see a growing need for computational query in e-commerce, finance, and IoT applications that depend on in-memory data grids to track fast-changing data. These applications require powerful query tools to make informed, timely decisions before the moment is lost. We see this as an easy but important step that .NET developers and architects can take towards harnessing the full power of data-parallel computing.”


Sign up for the free insideBIGDATA newsletter.

Speak Your Mind