Our Design: Simplicity is The Key

Clusterpoint database software technology architecture is based on few simple database concepts already known to majority of developers community: 'database' (data storage) and a 'cluster' (a number of hardware servers together running all particular database nodes). Simplicity for customers and developers was also our key design requirement for other aspects of using our database technology.

Simplicity of database model: Clusterpoint database logically works as a simple XML / JSON data storage where each data object (XML / JSON document) can have any schema-less free-format data structure and custom content.  All data objects must be uniquely identified by id-key, that is used to store, retrieve, manipulate and search data in any Clusterpoint database.  A key can be any XML or JSON existing attribute such as Web link, social security number, bank account number, chemical formula or any other parameter which is a string with unique identification value for the particular data object within the entire database so that customer applications can precisely store, retrieve, update and delete any data objects in a Clusterpoint database, using keys as unique references for locating data objects.  From application developer point of view Clusterpoint database (storage) is operating in nearly the same way as a key-based data store, yet Clusterpoint technology is providing more powerful structured and unstructured data processing functionality within content of data objects stored, addressing the internal structure of customer data objects so that, structured, unstructured and semi-structured data can be effectively manipulated and customer data structure can be handled similarly to that of SQL database programming methods. 

Simplicity of combined structured and unstructured data access: Clusterpoint ability to work with a database structure similarly to SQL and perform high quality enterprise search in all text values are two features that sets our software technology "miles apart" from a simple key-value store.  This Clusterpoint database extra functionality goes far beyond limited database functionality typically found in key-value store database software systems. The resulting Clusterpoint database software is an innovative hybrid database server platform, representing clustered data store functionality, structured database access functionality and enterprise search functionality for unstructured data, all at the same time.  All features, in particular, fast and relevant search that in essential for modern web apps (please see Information Ranking), are implemented for native scalability and high-performance in clusters so, that our customers can start building scalable software applications with rich structured and unstructured data retrieval functionality, easily handling growing databases with out-of-the-box clustering.  All above functionality is supported within a single database platform software: Clusterpoint Server. 

Simplicity of application software scalability: Our customers can start programming natively scalable applications in typical interactive web model fully supported by Clusterpoint Server, without the need to re-engineer their application software later, when databases can be scaled out in massively clustered private cloud configurations, likely with multiple replicated database mirrors, and still be able to work with the same software.  The simplicity of application software scale out ability is a key Clusterpoint database feature too.

Simplicity of system design concepts and terminology: We are also different from most of other database software vendors that we do not use technically sophisticated concepts to describe our database platform.  Often too much use of specific technical terminology developed by engineers for their "baby products" can make database programming and management unnecessarily difficult to understand and follow by customers. We did our best to simplify our database concept and high-level vision as much as possible.  For our customers there is no need to use such technical terms as 'relations', 'primary keys', 'foreign keys', 'buckets', 'forests', 'data blocks', 'key sharding', 'map-reduce' or other geekish, sophisticated terminology to work with Clusterpoint database.

In fact, most of Clusterpoint data storage internal technical complexity, where it is necessary, is hidden behind our core platform software developed as a database server.   Developers and database customers can just work with their logical databases and XML / JSON data objects, without worrying about technicalities or software engineering details driving  the Clusterpoint Server platform software.

Apart from this our customer-centric design  simplification, our product is a well-engineered, developed from ground-zero, fully functional database server software.  It is a fast, robust, scalable and entirely searchable XML / JSON data storage platform.

Simplicity of deployment: It is also very easy to deploy our database technology.  There is no client software necessary to store, access and search data in Clusterpoint DBMS.   Use REST web services or native TCP transport (through our API libraries for specific programing languages) and "talk" with Clusterpoint Server in XML or JSON, data formats nearly universally supported across different IT platforms.

Simplicity of high-performance data management: Please note that our database technology is seemingly very simplistic in use and appearance, but our software engineering efforts that underlies Clusterpoint Server implementation in the software source code is neither simple, nor trivial.  Actually Clusterpoint Server is driven by powerful and functionally versatile high-speed data management software that is written in fast C/C++ source code.  It was written from ground-zero and accumulates about 80-man years of our teamwork (we are professional C/C++ developers, please see Clusterpoint Team).  The software undergone many years of testing, bug-fixing and its reliability is proved by our customers running it in 365/24 production since 2008.  We have managed to build cutting-edge clustering, database management and enterprise search engine functionality into a single cohesive software platform, engineered for maximum performance and scalability and designed to run on inexpensive networked hardware in the so called private cloud architecture.  Code is optimized to run close to machine-level speed.

Simplicity of underlying computing infrastructure: We believe that our engineering efforts also resulted into a database software platform that requires only modest computing resources to operate and the software is better designed for green computing needs than any legacy database platform.

Clusterpoint software can work at high speed on any commodity IT infrastructure.  When developing Clusterpoint Server source code, we took advantage of the best industry software engineering standards and methods such as multi-core processing, multi-threading, data and index in-memory caching, map-reduce type of interprocess communications in distributed cluster architecture, server software virtualization and many other software-based optimization techniques that delivers the most performance out of the modern underlying hardware networked together for the total aggregate computing power.   Clusterpoint is the software system that is natively designed for this cloud-like architecture, without legacy software dependencies.  You can use Clusterpoint Server database platform as a fast and cost-efficient "machine" that enables to scale out any databases to any imaginable size just with a simple hardware clustering, maintaining high performance.  Clusterpoint technology enables to do clustered database management with the least cost of ownership:  using ubiquitous commodity computers and networking equipment with modest computing power and therefore low power consumption.  

Simplicity of integration: Clusterpoint database platform is remarkably easy to integrate from software developer point of view.  You can actually learn it in just a couple of hours and start programming for Clusterpoint databases, using your own favorite programming language.   There is nearly no learning curve in Clusterpoint software deployment and use.  Clusterpoint Server does all database clustering and scaling functionality behind the scenes, where this complexity is hidden from application software.  We automate on a database server level all database content indexing, search and cluster management tasks. We unlock application software from database structure complexities, cluster configuration, data sorting tasks and database replication tasks. It is all done by Clusterpoint Server software, transparently from application, there is no need to program this logic into application software code.

Simplicity of API: Our database server has also very simple API that is completely open and based on web standards.  We use only industry-standard and open data formats in our database storage: XML and JSON, and we have open Clusterpoint API that is web-based and uses only standard Internet protocols (http/https, REST, TCP/IP).  Our API works similarly to Amazon AWS services, so that our customers can easily develop their own enterprise cloud database services and even provide them to their customers.  Our open API and use of industry standards enables our customers to design, manage and run their XML- or JSON- only databases in fully cross-platform and application software vendor-independent way. 

Simplification of application software code: In particular, one of the greatest ROI benefits coming from our database model simplification is in application software development area. Clusterpoint database makes any web application software code shorter, more simple and more fast.  Clusterpoint DBMS enables our customers to cut their application software development time from months to weeks, saving significant development efforts and shortening time to market for their new products and new Web services.

All of the above simplicity factors have been engineered into Clusterpoint core database server technology.  The resulting product - Clusterpoint Server delivers cost-savings in a number of areas.  It saves licensing costs compared to multi-layered solutions.   There is no need to integrate Clusterpoint based application software with distributed file systems, map-reduce platforms, big-table columnar databases or enterprise search platforms when using Clusterpoint.  Clusterpoint database technology delivers great value for money: a long-term scalable data storage infrastructure, empowered with great database search facility out-of-the-box.  It saves tons of human maintenance efforts over IT system life-cycle due to much simpler database architecture and data model. Technology provides also excellent interoperability with existing customer software applications, eliminating entire layers of IT complexity compared to integrated solutions typically used to drive applications that are created around legacy SQL "world" concepts.  Using Clusterpoint Server as a primary database storage technology results into substantial customer cost-savings: target ball-park figure 50%.  It results also into employees productivity increase.  Clusterpoint database users are enjoying much better, faster and more relevant access to accumulated database information, driving up your customer satisfaction rates and bringing more revenue to your business.

Clusterpoint is also a power-saving computing technology, compared to legacy SQL database platforms, please see Calculate ROI.

For more advantages about using Clusterpoint database system please see the section Advantages. 

Database Engineered for Search

Search is the key functionality for most databases.  Databases are mission-critical part of nearly all corporate IT applications today.

Here is a problem: when your database grows big or complex, a good and relevant search starts to become difficult, slowing down performance and often producing overwhelming result sets. 

Customers have to use enterprise search software to achieve desired relevance and results ranking and fast performance. It is rather complex and costly solution to set up an integrated and scalable solution combining database management system platform (DBMS) and a search platform software.  It is often increasing application development time and requiring extra software development and maintenance.  Search software has its own proprietary indexing system that in most cases does not scale out well.  It has also full text search index synchronization problems with real-time database content updates.  It often requires cumbersome application software modifications to rank search terms for relevance in a customized way.  The concept of addressing fast and relevant database search with an integration of two different software platforms, essentially is locking in customers into proprietary database and search software programming formats, adding to overall IT system complexity.  It makes difficult and costly to migrate resulting integrated platforms among vendors to gain the best cost advantages.  Such systems also do not scale performance very well when the size of a SQL databases grows beyond few tens of millions of records per computer. Scalability needs to be addressed mostly by application partitioning (adding extra complexity) or through really expensive high-end hardware.

We asked ourselves what if we could make an open standards based database platform with rich enterprise search functionality, including fast full text search mechanism, built into the very core of a database server software?  What if we could provide also a database configuration mechanism for flexible search results ranking for relevance, and scalability in clusters for both database and search functionality, without any application software changes, out-of-the-box?  That is how our vision for Clusterpoint Server was born.

Based on this vision and design principles, our concept for Clusterpoint DBMS was implemented into a single, cohesive, engineered for speed, database software platform product: Clusterpoint Server. 

Undivided XML/JSON Data Objects

When we decided to engineer a new database software platform, there were mostly skeptics criticizing our endeavor and they often doomed our undertaking as a dead-cert failure.  Today global database software market is dominating by few industry "behemoths". Skeptics postulated that they would never allow some unknown company "out of nowhere" to compete in enterprise database software market, occupied by few dominating SQL RDBMS platform vendors.

After working with SQL and other relational database systems for almost a quarter of a century, it became obvious for us, that relational architecture is aging and can not handle growing amount of electronic data, especially, data what is produced by web applications on the Internet. 

Today there is an emerging consensus by growing number of IT professionals that legacy SQL technology can not efficiently solve all database management problems.  Michael Stonebraker, a research pioneer who about 30-years ago co-created the Ingres and Postgres technology that underlies many leading relational SQL databases today, including Microsoft SQL Server, Sybase and Informix, was among the first well-known database industry celebrities who honestly stated that RDBMS "should be considered legacy technology" (Source: Computerworld, Sept. 06,2007).

One can take a look at any relational database architecture with its multi-table entity-relations diagrams (see the picture on the right side) and immediately see the complexity behind that outdated computing technology.

Complexity of relational databases design requires complex software to manage this complexity. Legacy SQL database design and maintenance is time-consuming, taking great efforts from analysts, developers, DBAs etc.  Another inconvenience is that column indexes must be used for fast search, and this indexing system must be custom designed into the very structure of each database, often requiring to modify and adjust application software to search limitations imposed by column indexing, restricted by relations between tables and requiring to address in software inflexible, rigid database structure and search logic. 

For the design of the Clusterpoint database we have thrown out legacy SQL architecture and constraints.

We decided to create a NoSQL database platform that is designed for Web and stores all data only as industry standard schema-less XML / JSON documents from API side, while storing them in XML internally.  Any customer data objects can be stored, managed and searched for in the Clusterpoint database in either free format XML or JSON data objects, using Clusterpoint API.

Clusterpoint Server software is data structure agnostic database platform.

As a result, database design becomes simple and application software development becomes even more simple.Key benefits of storing and managing data only in native XML / JSON data formats:

• Industry standard data formats: supported by most other enterprise software systems; easy to export/import and to convert to other formats;
• XML is the most humanly readable machine data format, if database data is de-normalized (a 'heresy' in relational world!);
  
• Both XML and JSON are pure text data formats, that facilitates reuse of corporate IT knowledge: databases become self-explaining;
  
• Real-life data objects can be described with XML / JSON in all of their natural complexity (records, tables, hierarchies, multi-values, relationships etc.; XML also supports cross-links and combination of all other object types can be stored in a single XML document as binary encoded tags);
  
• Application software code for XML / JSON objects can be reused, without multi-table database structure specific  transaction integrity programming into application software; which is one of the most time consuming and error-prone tasks in relational systems;
  
• Database structure can be changed for new application functionality on-the-fly, simply rewriting document by document stored in the database, without entire database structure migration; eliminating one of the most complex processes in database management: complete database migration to support data structure for new application version;
  
• XML- and JSON- documents are cross-platform data objects and could be accessed and re-used in any programming language; JSON (JavaScript Object Notation) is a lightweight data-interchange format, which can be used even directly in web client software through JavaScript;
  
• Rapidly develop new application software functionality and test new application software versions on a few modified for testing XML / JSON objects in production database; while previous application software continues to work along new version software;
• Database administration on system level does not require learning of particular application specific data structure, for system administrators any database is just a collection of XML / JSON documents located in the same name directory.

With innovative Clusterpoint Indexing system, Clusterpoint Search and Clusterpoint Information Ranking we have also solved the perception problem from our customers that XML / JSON is text and being a text, it can be difficult and slow to search. 

Key Software Engineering Principles

Before we started the development of software code for Clusterpoint Server database platform, we drafted a list of key design requirements that would make sense for simplifying database management and database application software:

• Cross-platform data storage: open API, open network protocols and open data formats
• Totally searchable database: index created automatically for the entire database content, any data structure
• Universal database: storage for any types of data structure objects (records, tables, trees, texts etc.)
• Massively scalable database: unlimited data storage, using abundance of cheap storage and low-cost CPU networked in clusters
• Very fast database: sub-second search response time, including user-friendly free-format Internet-style search (full text search)
• Client-server system: reliable IT industry standard for modern multi-tier data center server systems
• Long-term data storage: lasting for many decades from data storage format longevity point of view
• Safe, secure and reliable database: secure access, database mirroring (replication) in multiple copies, native virtualization
• Cost-efficient database ownership: requiring only basic IT skills, nearly no learning curve, use of inexpensive low-cost hardware
• Simple administration: using web-based database management tool (centralized for all databases, one-click cloud-wide changes)

Based on the above wish-list for an ideal database design requirements, we wrote Clusterpoint Server in C/C++ source code to match our design concept. 

Distributed Data Store Architecture

Clusterpoint Server is a distributed XML/JSON database server architecture NoSQL DBMS software with linear scalability to address database growth problem and performance issues.

Clusterpoint technology is a classic transactional, ACID-compatible, data management software technology that delivers in a single software platform a distributed cluster data store functionality, full-fledged XML/JSON database server functionality and enterprise search engine functionality, all at the same time.

The system was designed to take advantage of cost-efficient clustering: contemporary hardware architecture where expendable servers with a lot of RAM and plenty of disks are networked in clusters to provide massive computing and data storage capacity.  Trendy name for this system concept is cloud-architecture, in public cloud or in 'private' cloud, the later being your own corporate IT infrastructure, that uses cloud principles.

We develop, sell and support NoSQL database platform product for building and managing your own private cloud-architecture databases.

For Clusterpoint Server we developed a set of innovative indexing and search facilities:  Clusterpoint Index, which enables to configure Clusterpoint Search for the best search relevancy without application software changes, and the indexing engine that automatically indexes entire database content for easy Internet-style full text search, and enables to apply customer own Information Ranking rules for configuring database search relevancy.

Both indexing and search are linearly scalable in a cluster of private-cloud hardware architecture supported by Clusterpoint design concept.  
Clusterpoint database platform provides configurable database Clustering options,  that are supported by Clusterpoint Server out-of-the-box: without extra software and without special programming for it in customer application software.

Client-Server DBMS with Open API

To integrate it into contemporary multi server data center IT systems architecture, Clusterpoint database server software was engineered to support industry proved client-server database architecture.  The core engine runs as a database server and can be accessed through Clusterpoint API.

Clusterpoint Server uses REST principles for messaging between customer applications and server software.  

Its messaging system is very simple, uses only web protocols and XML / JSON data objects, and underlies Clusterpoint API.  

Development of application software for Clusterpoint databases does not require special client software: any favorite programming language can "talk" to any Clusterpoint Server out-of-the-box, over standard web http infrastructure and in standard XML / JSON or, using Clusterpoint native API libraries for those languages, fast TCP/IP data exchange, no matter where the server is located.  

This our API simplicity provides also easiness and simplicity of integration into existing software applications.  With few lines of existing web application software code changes developers can start accessing Clusterpoint Server instead of SQL server for data access and search, replacing only SQL cursor parsing source code with the same structure XML-/JSON-object parsing and maintaining all visual output element formatting unchanged.

Clusterpoint API is completely open, published protocol for cross-platform software development and facilitates database usage portability and interoperability among different platforms, without vendor specific protocols or data storage standards implementations.

Data Structure Agnostic Database

With Clusterpoint you can start doing things impossible with traditional relational architecture databases. 

One of the biggest advantages of Clusterpoint is that you can store completely different objects into the same database (which we simply call a 'storage').  

Clusterpoint Server automatically indexes all elements in your database content for ultra-fast access and search, creating Clusterpoint Index - complete data index for any custom database objects.  There is no need to manage multiple tables, column indexes and relations in Clusterpoint architecture.

In our architecture, Clusterpoint Server empowers you to store any custom XML / JSON objects, having different data structure, different size, different content, and still can provide relevant information retrieval, sorting, filtering - as if you would be using relational database.  The general data storage architecture for Clusterpoint Server software is illustrated in the picture on the right side.

We decided to use open technology and open data formats that are widely used on Web and in web technology applications, to interact with Clusterpoint Server software, and to store, manage and search information in Clusterpoint databases.  

Although from our API we offer both XML / JSON data format usage, internally we store data only in XML.  

XML was a natural choice as our database format: it is robust and well-supported data format by all IT industry, by virtually all popular software platform and computing environments.  

Another reason for choosing XML is its more flexible accommodation of some specific data formats, that JSON is not supporting.  It is possible to store all JSON objects into XML, but not in reverse way.  So we decided to encode and handle Clusterpoint database structure in XML within Clusterpoint Server internally, providing open API options to access it either with XML or JSON messaging, depending on the preferred development environment and its support for one or another standard.

Focus on High-Performance Search

Clusterpoint Server empowers you to add super-fast Internet-style search queries to your database applications and always retrieve the most relevant data first.

Clusterpoint DBMS indexes entire database content for structured, unstructured (full text) and semi-structured search.  

In Clusterpoint architecture the entire database information is being indexed up to the very last detail (string, word, number, date, tag etc.), and index is being updated in real-time along document writes. 

Clusterpoint software hides real complexity of this full database content indexing and clustering behind database infrastructure. 

To make database SEARCH both scalable and high-speed in massively clustered database architecture, Clusterpoint database technology have a built-in customizable Information Ranking mechanism for flexible positioning, grouping and ordering of data for the maximum relevancy and ultra-fast SEARCH performance.

Information Ranking is a freely programmable database indexing mechanism how to instantly get the most meaningful (relevant) data upfront, e.g., on the 1st web page.

The resulting data retrieval system outperforms SQL databases by factor of 10-1000 at fast, relevant, high-precision search, including free-format Internet-style search (full text search with relevance scoring, stemming, results snippets, autocomplete, facets)

With Clusterpoint technology, you can solve two major database usability problems existing today:

• Performance of search.  It often disappoints users when a database size grows too large.  Quickly finding relevant information in large size legacy databases is difficult and expensive.  With Clusterpoint you can make lightning fast custom search queries with complex data ordering and grouping requirements, where search is performed across a distributed cluster database in a very simple querying format: as Internet-style ad hoc search query using any known content within a database.  Outperform legacy SQL systems by a factor of 100 to 1000, achieving sub-second response times for your Web application.  Please see more in sections Clusterpoint Search and Performance.
  
  
• Relevancy of search. Many application software systems have an underlying legacy database which can be accessed by only a limited number of queries and reports, pre-programmed and optimized for performance by expert programmers.   Clusterpoint's enterprise search functionality, including free-format Internet-style ad hoc querying system, allows users themselves to create specific, customized queries from any known content in the database and still always get the most relevant data out of the database, even if contains billions of objects and is massively clustered over tens and hundreds of servers.  Information ranking is used to instantly bring the most useful database information on the first Web page, sorted, grouped and exactly positioned by customer own ranking rules, with follow-up pages containing less ranked (less important) results.  Information ranking mechanism is freely programmable by customer own data ordering preferences and search logic requirements in Clusterpoint database architecture, where customer can set up and fine-tune its own search relevancy ranking logic at database configuration level, no application software changes are required.  In fact, Information Ranking is a game-changing characteristic of Clusterpoint database technology, enabling our customers to start building disruptive software applications, requiring linearly scalable, data-structure agnostic, high-performance computing enabled database infrastructure that delivers guaranteed sub-second quality search in any database, without performance degradation when database size grows big.

 Please read more about Information Ranking.

Virtualization of Database Servers

Taking the best practice methods of the latest virtualization software technologies, each database storage is serviced in RAM by the fully functional Clusterpoint database server software instance. 

There is no need to run virtualization software on a hardware: each Clusterpoint Server can be managed to run into separate RAM address space, servicing different customer databases and different applications, with their own users and access rights. 

At the Clusterpoint Manager level (application for managing, configuring and controlling database servers, storages and clusters) there can be  multiple administrator accounts, who can see, manage and configure only their particular storages, run and stop server software only for their storages etc.

This feature empowers you to run many different Clusterpoint databases on the same hardware securely and resource-efficiently.

You can also run Clusterpoint software on different operating system platforms, or run it under a virtual machine or a virtual appliance software environment using elastic cloud compute architectures such as Amazon EC2 or its open source analog Eucalyptus.   

Please contact us if you want to test-run Clusterpoint databases using public cloud services, for example, using Amazon AWS services.  We did help a number of our customers to do so with Amazon and we could quickly set up a test-bed for Clusterpoint DBMS evaluation also for your needs on Amazon cloud, provided that you supply us your sample database and key data retrieval requirements.  Please simply let us know about your interest sending us email on Clusterpoint Technical Support.

Multi-lingual Database Software

Clusterpoint DBMS delivers the product engineered for global IT market: with full support for national character encodings and transformations between code pages. 

High-speed conversion between UTF-8 and national encodings is performed on the server if requested by Clusterpoint API.

Clusterpoint’s database engine is designed to provide character conversion automatically on the server, taking advantage of fast C/C++ code. 

Among key national-languages specific features supported in fast Clusterpoint Server code are a number of linguistic functionality, usually requiring some extra configuration of legacy database systems or customizable software code in application software for a particular national language.   This built-in Clusterpoint feature set is especially benefiting web applications, which ideally must be multi-language from ground-up.

Here are main multi-lingual features of Clusterpoint Server:

• The Clusterpoint DBMS platform enables to store and query data in multiple languages in the same storage in UTF-8;
• Customers can use multiple language terms in a single search query without setting up customized database versions for each particular language;
• Query in any language; combine multiple terms in different languages in a single search query;
  
• Store information in > 160 languages into a single Clusterpoint database, and even in a single database document;
• Custom modules for language stemming rules from linguistic and grammar point of view are available for all major languages, enabling automatic expansion of search queries with all relevant by grammar rules word forms; the system can be customized for any language;
• Statistics based default stemming is supported for any language, provided good enough stemming rules based on the accumulated index elements such as word forms from the actual database data;
• Customizable stop-words and delimiter characters can be configured for each database storage, depending on the default language, type of application, database content and customer needs;
• A custom synonym vocabulary can be submitted for search terms automatic expansion, when requested by Clusterpoint API; it is useful to add abbreviations to text data, similar word forms etc.;  the synonym vocabulary is a simple comma delimited text file which can be freely edited by customer;
• customizable language specific wildcard options can be configured, setting up limits for automatic search expansion for fuzzy search (spell check against actually present data elements), limits for wildcard expansion when requested and limits for string pattern lookups; we provide general mechanism to configure default values fitting to any language.
  

If you need some additional language specific feature, we can quickly implement it in our fast C/C++ server source code, which can improve your application performance and make application software less complex. 

System Architecture & Components

Clusterpoint DBMS can be downloaded as an installation package that includes the core Clusterpoint database platform software and Clusterpoint pre-packaged applications.  Overall Clusterpoint DBMS system architecture and components packaging is illustrated in the picture below.

The core software is Clusterpoint Server, which installs as the same copy on all hardware servers (colored in blue).   The software colored in blue combines all data storage services, all clustering functionality and all enterprise search functionality (indexing service and search facility), all within a single, scalable, client-server database software platform.  

This server-based software platform is available to outside "world" through open API based on industry standard web messaging (and supports all TCP/IP, http / https / REST, XML / JSON standards). 

One can illustratively say that the core Clusterpoint DBMS database engine is being driven by seamless operation of those key server components like a wheel is driven by ball-bearings.

There are also other pre-packaged software applications in Clusterpoint installation: Clusterpoint Manager, Crawler Application and Searcher Application which are packaged together with Clusterpoint Server software (please see applications colored in orange and yellow in the picture). 

To operate Clusterpoint databases, customer uses own client software and customer XML databases (colored in green), interacting with Clusterpoint Server through Clusterpoint API.

Normally we use Clusterpoint Server and Clusterpoint DBMS terms in the same context, assuming that the core software Clusterpoint Server will always come with some pre-packaged Clusterpoint applications, for out-of-the-box usage.

Clusterpoint Manager is a centralized administration and management application for a cluster.  It is used by a database administrator (DBA) or by a system administrators to set up, run, control and monitor all Clusterpoint databases and their daily operations across the enterprise IT network.  This is probably the most important Clusterpoint DBMS pre-packaged application.  

It is also a recommended application, if you want to conveniently use Clusterpoint Server software in cluster environment, perform routine DBA tasks, backups, restores, user account management, log file and audit management etc.  Clusterpoint Manager application comes free-of-charge with every Clusterpoint Server installation package and installs on a web server as as module, which can be accessed for log-in and used for  centralized management of Clusterpoint Server database IT infrastructure across all your servers.

Administrators can use any Internet browser software to centrally log in and manage all Clusterpoint servers and databases installed and operated enterprise-wide.  They can also easily manage their corporate Clusterpoint DBMS software infrastructure remotely from any location with Internet access, using  fully encrypted and secure SSL tunnel over the Internet to the Clusterpoint Manager utility (if this type of access is configured on corporate firewalls).  

There is also a command line management utility, that allows to manage Clusterpoint Servers from scripts, OS shells and in batch processing without Clusterpoint Manager.  It can be downloaded from our Web site free of charge.

Please read more about Clusterpoint DBMS management and administration functionality in section Management.

Clusterpoint Crawler and Clusterpoint Searcher are two optional sample demonstration applications (colored yellow in the picture above), pre-packaged for a quick Clusterpoint DBMS capacity and functionality testing by our customers.  Interested customers may use those 2 applications to crawl some existing document-type data (for example, a corporate Web site data or local file system documents), store collected documents in a Clusterpoint DBMS storage as XML documents, and perform resulting database search and performance tests, without spending any time for application software programming, out-of-the-box.

Using pre-packaged Clusterpoint applications, our customers can see and inspect how Clusterpoint databases are working, what is system architecture, how Clusterpoint databases are storing documents, how customer XML structure can be configured for relevant search etc.

Please see more about Clusterpoint DBMS demonstration applications in section Clusterpoint Applications.