GridGain Professional и Enterprise Edition

2018

There was a version of GridGain Professional Edition 2.4

On March 29, 2018 the GridGain Systems company announced release of GridGain Professional Edition 2.4. The framework for continuous training including machine learning and a multilayer perceptron (neural network which allows the companies to start algorithms of machine and deep learning on petabytes of operational data in real time) is added to the version. GridGain Professional Edition 2.4 also allows to increase efficiency of Apache Spark thanks to API for frames of data of Spark (it is added to earlier existing support of RDD in Spark).

GridGain Professional Edition 2.4 (2018)

Framework of continuous training of GridGain

GridGain Professional Edition 2.4 is the first version which completely supports the integrated machine learning and a multilayer perceptron of Apache Ignite. Thus, continuous machine/deep learning will be directly available in GridGain. Optimizing libraries of machine learning for mass parallel computings, the version can accelerate considerably machine learning on large volumes of data. Data processing directly in a cluster of GridGain allows to provide continuous process of training: it is not necessary to move transaction data to separate base before training of model any more. Result — training in real time and also continuous training with smaller complexity and reduced costs in comparison with traditional approaches.

The framework of continuous training of GridGain is a structural element for the intra process applications using hybrid transactional and analytical data processing (HTAP). In such applications the information model constantly studies at a basis of the entering data. Thanks to intra process HTAP calculations of the application of generation can take advantages of training in real time. Different business applications will be able to make more effective decisions in real time — to prevent fraud, to make recommendations in the field of e-commerce, to approve the credits, to manage logistics and service of the transport systems.

Expanded support of frames of data of Spark

Now GridGain can be used for storage and management of frames of data of Spark. Support of frames of data expands support of Spark which was the most complete among in-memory of platforms also earlier. GridGain still includes GridGain RDD API for the GridGain data access as to changeable RDD and also the Ignite (IGFS) file system which allows to use GridGain as in-memory implementation of a distributed file system of Hadoop (HDFS).

Spark can be used for data processing in GridGain as frames of data or RDD and also for preserving of frames of data or RDD in GridGain for postprocessing. These opportunities allow Spark-developer to use GridGain as in-memory storage for access, preserving and data exchange between Spark tasks. GridGain supports ANSI SQL - 99, including indexing of data so Apache Spark can use the distributed SQL of the GridGain platform for performance improvement of independent requests to 1000 times. Spark developers can also use a framework of continuous training for automation of solutions and permanent updating of models for the purpose of improvement of results in real time.

"The companies which aim to automate more intellectual decision making process need to lead to permanent cooperation two sides of a digital brain — machine learning and the system of automatic decision making. With a release of the version of GridGain continuous training of information models in real time on data bulks, with a speed and scale of in-memory of calculations, with smaller difficulties and expenses became possible. It is the first step to that intra process HTAP-applications operated applications of continuous training which, in turn, will provide implementation of initiatives of digital transformation and implementation of omnichannel". Abe Klyaynfeld, president and CEO of GridGain Systems

2017

There was a version of GridGain Professional Edition 2.0

On May 15, 2017 the GridGain Systems company announced release of the version of GridGain Professional Edition 2.0. In a product the architecture of off-heap memory is updated, support of the version of Apache Ignite 2.0 is implemented. also support of languages DDL for work with SQL bases is added.

Among basic functions of GridGain Professional Edition 2.0 there is an architecture of off-heap memory, it goes beyond RAM and supports the systems on the basis of SSD and flash-memory. It allows to reduce pauses in work of Java Garbage Collection. Support of languages DDL within SQL Grid storage, the Spring Data API tools, functionality of .NET and C ++ and other opportunities is added. The release increases performance and simplifies integration for the user scenarios in which distributed computing in memory provides the speed and scaling of the high-loaded applications.

In GridGain Professional Edition 2.0 the architecture of off-heap memory is upgraded. The off-heap memory technology uses page-by-page approach with layer-by-layer distribution of memory which at will can become attached to any persistent-storage without the need for serialization or deserialising of data. GridGain Professional Edition 2.0 also automatically executes defragmentation of data and gets rid of any noticeable pauses in work of Java GC.

Support of languages DDL in the built-in storage SQL Grid is added to GridGain Professional Edition 2.0. Administrators can set and change indexes during operation without reset of a cluster now. Access to function is provided through standard SQL commands, including CREATE and DROP. Support of DDL improves possibilities of SQL Grid storage and expands the set of applications which the organization can accelerate and scale using the platform with the open code.

Integration from Spring Data simplifies interaction with a cluster of GridGain Professional Edition 2.0 at the expense of the used Spring Data API. Being connected to GridGain cluster using Spring Data API, users can execute the distributed SQL queries and simple CRUD transactions.

Ignite.NET provides the system of plug-ins for integration into third-party solutions. Plug-ins expand standard Ignite.NET API due to customization of API or adding of functionality.

Ignite.C ++ allows to execute the code customized by C ++ on remote nodes of Ignite. Developers C ++ do not need to switch to Java if they want to turn off certain updates on nodes of data GridGain – they can make it in C ++.

With integration the cache memory of Hibernate L2, GridGain Professional Edition 2.0 supports Hibernate 5 now.

The GridGain Professional Edition 2.0 functions reflect our vision of simple approach in use to calculations in memory. The updated architecture of off-heap memory in GridGain Professional Edition 2.0 freely expands RAM due to support of systems on the basis of SSD and flash-memory. Users can unroll a multi-tier architecture of memory which balances loading in applications, sensitive to delays, better. Support of DDL strengthens possibilities of ANSI SQL-99, simplifying use of a product by the specialists who got used to work with databases using the standard SQL commands. The release includes professional support and timely error correction. Abe Klyaynfield, president and CEO of GridGain Systems

GridGain is compatible to Hortonworks and Tableau

On April 5, 2017 the company GridGain Systems announced certification of GridGain Professional and Enterprise Edition on compatibility with solutions Hortonworks and Tableau. Along with it the companies stated technology partnership.

It is supposed that these steps of the companies will simplify start of the big data systems constructed on Hortonworks and using in-memory computing technologies for clients. Corporate users of GridGain will be able to visualize data in memory using solutions of Tableau.

Within technology cooperation GridGain received the certificates of Hortonworks and Tableau confirming full compatibility of the GridGain Professional and Enterprise Edition platforms with these solutions.

the Organizations even more often rely on Hortonworks and Tableau for maintaining a key business intelligence and timely decision making. As the companies scale the solutions for analytics of big data, the GridGain platforms allow them to achieve exclusive performance in data processing which only in-memory computing can provide. Thanks to reduction in cost of memory and multifunction advanced design of GridGain, creation of in-memory of infrastructure already began the most profitable method to provide performance and scaling. Possibilities of our ODBC/JDBC API also open access to a wide range of solutions for data visualization, including Tableau, for easy work with data in memory and fast obtaining results of their processing.

GridGain Professional Edition 1.9

On March 13, 2017 the GridGain Systems company announced release of the version of the GridGain Professional Edition 1.9 platform. Basic functions of release - support of the version of Apache Ignite 1.9, the increased performance, the automated performance monitoring, support of a management system for clusters of the containers Kubernetes and expanded support of a data control language (DML).

Apache Ignite 1.9 shows for 20-40% higher performance for the majority of transactions of caching and SQL in comparison with the previous version. One of the main improvements – a possibility of a configuration of level of parallelism for SQL queries in Ignite SQL Grid – improves Apache Ignite performance in the user scenarios with a high load. Besides, the automated benchmarking is added. Users can unroll the binary Apache Ignite 1.9 file and carry out tests in the environment of Ignite.

Representation of architecture of GridGain Professional Edition, (2016)

Kubernetes is open source solution which automates deployment, scaling and management of container applications. Integration of Apache Ignite with Kubernetes simplifies implementation of an Ignite-cluster in the container Kubernetes, allowing the solution to manage resources and to scale an Ignite-cluster. For example, if the user specifies that the cluster of Apache Ignite should support at least five nodes, Kubernetes will automatically provide accomplishment of this requirement at any time.

Expanded support of languages - Data Manipulation Language

In the Apache Ignite 1.9 platform which is compatible to ANSI SQL-99 support of DML includes API Ignite.NET and Ignite.C ++ now. It expands possibilities of use of DML for users of Apache Ignite which work with .NET or C ++. The streaming mode providing accomplishment of DML transactions in certain scenarios – for example, package inserts and updates or preliminary data loading is added.

Expanded support of .NET and C ++

In Ignite.NET, the version of Apache Ignite for .NET, .NET TransactionScope API which allows to work with the distributed Ignite-transactions is added, relying on standard interfaces in .NET. Ignite.C ++, the version of Apache Ignite for C ++, provides support of API for the most popular continuous requests now. At the expense of it users can in C ++ applications monitor the changes of data happening in the distributed Ignite cache.

Apache Spark integration

Apache Ignite is integrated with Apache Spark 2.1 now that provides use of the ring Ignite Shared RDD databases in the applications working with the current version of Apache Spark.

GridGain Professional Edition 1.9 had additional opportunities for deployment and management that simplifies use of in-memory computing in the most important and high-loaded applications. The convenience of integration, professional support and timely error correction in GridGain Professional Edition does calculations in operational memory by the most reasonable solution for the organizations solving the most complex problems of data processing. Abe Klyaynfield, president and CEO of GridGain Systems

2016

GridGain Enterprise Edition на базе Apache Ignite

The software of GridGain Enterprise Edition based on Apache Ignite provides scaling of applications with a high volume of data processing to petabyte level and acceleration of transactions by 1000 times in comparison with processing on a disk without replacement of the existing databases. It offers high-speed ACID transactions, streaming in real time and effective analytics within uniform complex data access and level of their processing. The solution ensures functioning as existing, and new applications in the distributed, mass and parallel architecture on the available, standard hardware in the industry which is easy for scaling by adding of additional nodes in the computing system. The GridGain platform does not require, or requires the minimum changes in the application or levels of the database for architecture on the basis of RDBMS, NoSQL or Apache Hadoop.

GridGain Professional Edition 1.8

On December 19, 2016 the GridGain company announced release of release of the version of GridGain Professional Edition 1.8. In release support of In-Memory SQL Grid storages is organized.

In version 1.8 GridGain Professional Edition the full support of the latest version of Apache Ignite 1.8, a possibility of the distributed data warehouse of In-Memory SQL Grid is implemented. The In-Memory SQL Grid storage will help users to accelerate and scale the SQL applications without changes, or with the minimum adjustments of the code. Users can execute requests to data in GridGain, using SQL via drivers of ODBC/JDBC, to modify data by means of the DML commands without the customized code.

In the platform geospatial support is upgraded. In-Memory SQL Grid will provide more convenient integration of the GridGain platform and the improved functionality for the bigger volume of the user scenarios, including cloud applicaions and Internet of Things (IoT). GridGain Professional Edition 1.8 supports the Redis protocol and includes support of caching of .NET Entity Framework 2nd-Level Cache and ASP.NET Session Caching.

In-Memory SQL Grid gives possibilities of the distributed data warehouse in the GridGain platform that allows to accelerate and scale considerably SQL applications without changes, or with the minimum adjustments of the code. In-Memory SQL Grid is scaled horizontally, differs in fault tolerance and compatibility with ANSI SQL-99. The storage completely supports the SQL and DML commands, including Select queries, UPDATE, INSERT, MERGE and DELETE.

In-Memory SQL Grid allows to work with the GridGain platform, using standard SQL commands using the GridGain JDBC or ODBC drivers without customization of the code. In addition to support of Java, .NET, C ++, possibilities of the JDBC and ODBC drivers provide the connection to GridGain or Apache Ignite based on SQL from the platform in any programming language, including Ruby, PHP, Python and others.

The improved geospatial functions which are also supported by In-Memory SQL Grid adapt the GridGain platform for transfer of calculations in RAM to bigger number of the user scenarios in the field of IoT founded on a geolocation – for example, in the field of transport, logistics, mobile technologies, retail and wearable devices.

Support of the Redis protocol

Redis is the high-performance not relational distributed data warehouse in memory on the open-source platform which can be used as the database, a cache or the module for the translation of messages from one protocol in another. Integration with Redis into GridGain Professional Edition 1.8 allows to store and take the distributed data in GridGain cache, using any client program, compatible to Redis. It gives the chance of convenient migration from the solution for caching of Redis on the full GridGain platform with the minimum changes of protocols of communications of clients.

Caching of .NET Entity Framework 2nd-Level and ASP.NET Session Caching

The updated platform provides support of caching of .NET Entity Framework's 2nd level which can be easily integrated with the existing applications. It provides increase in productivity as results of requests to data are kept in the distributed failsafe cache with a possibility of shared access.

The processed component for management of ASP.NET sessions which is provided by Ignite.NET provider allows to distribute a session between a cluster of machines for the failsafe balanced load.

The GridGain Professional Edition 1.8 is an important step in development of our platform for data processing in memory. A number of new features simplify interaction with GridGain for users and improve work of the platform for the user scenarios with a high load, including cloud applicaions and Internet of Things. Being the most advanced platform in the world for transfer of calculations in RAM, GridGain Professional Edition 1.8 allows a bigger number of the companies to implement in-memory computing for achievement of goals in acceleration of data processing and scaling. Abe Klyaynfield, president and CEO of GridGain Systems

GridGain was included into AWS

On October 17, 2016 GridGain Systems announced placement of the GridGain Enterprise Edition platform in app store of Amazon Web Services (AWS) Marketplace.

The publication in AWS Marketplace will help the organizations of any scale to deploy quickly in AWS the distributed, mass and parallel solution of GridGain for transfer of calculations in RAM of the computer. In September GridGain Systems announced release of the solution GridGain In-Memory Data Fabric on a cloud platform of Microsoft Azure.

GridGain Professional Edition 1.7

On September 13, 2016 the GridGain Systems company declared availability of version 1.7 of the GridGain Professional Edition platform.

As a part of the upgraded product the version of Apache Ignite 1.7, the corrections which did not enter into Apache Ignite yet and the licenses LGPL.

The basic upgraded GridGain Professional Edition 1.7 functions: support of non-collocated joins in SQL and expanded support of management of clusters of GridGain Professional Edition. The release of Apache Ignite 1.7 is also integrated with the free GridGain Web Console tool for management of the GridGain installations and monitoring of performance.

Users of GridGain Professional Edition will get access to regular technical releases with error corrections. GridGain also transfers these developments to the Apache Ignite project - they will come out in the following release of this platform. Updating of GridGain Professional Edition 1.7 is available on a subscription to clients of GridGain Standard Professional Support. The GridGain Professional Edition platform is intended for the companies which are interested in support, timely error correction and simplicity of installation when using Apache Ignite.

GridGain Professional Edition 1.7 expands possibilities of use of SQL due to support of fast complex requests to large-scale and to distributed databases. Besides, the platform provides the improved management system for the software installations, crucial for business. These innovations clearly demonstrate why Apache Ignite and GridGain remain leaders of revolution of In-Memory Computing, developing the platform which is most actively growing at open-source for data processing in RAM in the world. Abe Klyaynfield, president and CEO of GridGain

Non-collocated Joins

Support of non-collocated joins solves a problem of processing of SQL queries in distributed systems. This function helps to execute ANSI SQL-99 requests to the data located on several distributed GridGain cluster nodes: if data on sales for 2014 with breakdown by types of clients are located on one node, and 2015 – on another, users can formulate a request to both databases, for example: "What total sales with breakdown by types of clients for 2014 and 2015 were?"

Management of GridGain Professional Edition and Apache Ignite with the GridGain Web Console tool

Tools of a configuration, management and monitoring in GridGain Web Console at deployment of software of GridGain became available to users. The supported functions:

• configuration of clusters and cells;
• import of schemes of databases practically from any RDBMS system;
• SQL queries in a free form;
• monitoring and management of the software installation.

GridGain Professional Edition 1.6

GridGain Professional Edition is the platform for transfer of calculations in RAM.

On June 29, 2016 GridGain Systems announced release of the version of GridGain Professional Edition 1.6. Development accelerates transactions in 1 thousand times in comparison with data processing on a disk.

The upgraded version got native support of the database management system Apache Cassandra, the new ODBC driver, transactions without interlock (deadlock-free) and the cloud management console. These functions simplify integration into instruments of data analysis, increase performance, provide access to the new tool for a configuration and management of the GridGain and Apache Ignite installations, increase performance of databases for users Cassandra.

With support of Apache Cassandra, GridGain expands possibilities for acceleration of work and scaling of the most popular databases both SQL, and NoSQL Apache Hadoop. A system Apache Cassandra is optimized for processing of the simple predetermined requests according to the data which are stored on a disk. However in it the possibility of data processing is not provided in RAM and support of transactions. Using integration with GridGain, users Apache Cassandra had an opportunity:

• Accelerate request processing in 1 thousand times due to transfer of calculations from a disk in RAM;
• Use compatibility with ANSI SQL-99 for processing as structured, and unstructured requests with a full support of ODBC and JDBC;
• Perform the transactions which are completely fulfilling the requirements of ACID for a read and write of data in the Cassandra bases;
• Use advantages of the built-in support of Apache Spark, Apache Hadoop and strimingovy applications.

The GridGain In-Memory Data Fabric platform on the basis of Apache Ignite ensures scalable functioning with the applications processing large volumes of data and also thousandfold acceleration of transactions in comparison with data processing on a disk without replacement of the existing databases. The solution supports high-speed transactions with fulfillment of requirements of ACID, streaming in real time and effective analytics within uniform complex data access and level of their processing.

GridGain ensures functioning of the applications as existing, and upgraded in the distributed, mass and parallel architecture on the available, standard hardware in the industry. The GridGain In-Memory Data Fabric platform does not require, or needs the minimum changes in the application or levels of the database for architecture on the basis of RDBMS, NoSQL or Apache Hadoop.

Basic functions of release:

• The simplified integration Apache Cassandra of page of GridGain offers integration with Apache Cassandra in the ready-made solution now, simplifying use Cassandra as permanent storage for any version of GridGain or Apache Ignite.
• New ODBC driver. Innovation allows to be connected to GridGain or Apache Ignite from the platform in any programming language. Users can also analyze the data which are stored in GridGain or Apache Ignite using Tableau or any other standard tool of analytics now.
• The updated management of interlocks. The new feature of transactions of deadlock-free provides deployment of the platform on several large-scale commands without interlock. This function provides automatic recognition of interlocks both for optimistic, and for pessimistic transactions, guaranteeing smooth operation even in situations of the high competition.
• Cloud GridGain Web Console tool. The web management console for GridGain and Apache Ignite simplifies media control due to functions of simplification of configurations, SQL queries, monitoring and control.

GridGain In-Memory Data Fabric is an effective technology for the companies which need to process and analyze large volumes of data in real time in transaction, analytical or hybrid user scenarios. In new release we simplified integration of GridGain with any environment due to control function by interlocks, the ODBC driver and the cloud web console which increase performance for bigger number of users. We are especially glad to emergence of the direct support Apache Cassandra giving to users Cassandra new opportunities – processing of unstructured requests and ACID transactions, compatible with requirements, with preserving of advantages of data processing rate in RAM. Abe Klyaynfield, president and CEO of GridGain