HomeFREE DOWNLOAD for development and evaluation

Concept Origins in the Beginning of 90ies

The idea to develop software for very fast database information retrieval based on non-relational document indexing first appeared to entrepreneur Gints Ernestsons back in the beginning of 90ies, now co-founder, chief technology architect and CEO of Clusterpoint Ltd.

Gints Ernestsons started his career in technology as physicist in the Latvian University, then changed profession to the computer programmer in early 80ies. His was software developer on IBM Series 360 mainframe in 1983 (Fortran, Assembler, PL/1, Adabas Natural).  In 80ies Software AG Adabas was the dominating mainframe database system for long years.  Later Gints was managing teams of developers programming database applications with PC based NoSQL RDBMSes (Dbase, FoxPro, Clipper) in late-80ies, before SQL technology became the next dominating relational database technology in the IT industry.  Then Gints pioneered use of SQL server database platform software for use in the largest national databases in mid-90ies, implementing SQL server-based RDBMSes for both private and government sector customers, using the database software platforms from the world's leading database vendors Oracle and Sybase, with hands-on personal experience in setting up, managing and application programming within those environments. 

Gints was also gradually starting to manage groups of database developers in public and private business IT projects since the beginning of 90ies, in mid 900es taking over CIO/CTO level responsibility to build the largest national database systems.   In most projects he was responsible to choose and implement database software platform technologies, accumulating an expert knowledge of SQL and relational database technologies and a good knowledge in networking and computer security: core IT infrastructure technologies of past two decades.

Next 15 years he has managed and supervised setting up software engineering organizations and putting in place database systems in the largest nation-wide scale database projects in roles of CIO and CTO, having overall system architecture design and implementation responsibility for many national databases and public registers in 90ies: the Register of Enterprises (Companies), the Population Register, the National Passport Issuing system, Border Control system, Immigration registers and more than 40 other database systems.

Gints is also a co-founder of several private IT companies since 1992.   The idea of document-oriented DBMS first was concepted by Gints in early 90-ies within Lursoft, one of the first national IT companies he helped to set up.

back

Co-founding LURSOFT 1992

In 1992 Gints was CTO and a co-founder of Lursoft Ltd.  The company has launched very successful business in public database information distribution market and still is the market leader in subscription based Internet database services today, please check the company Web site:

Lursoft logo  www.lursoft.lv

In early 90ies IT sector using PC technologies was dominated mainly by DOS and Novell operating systems.  Dbase, Clipper, Paradox and FoxPro relational databases were widely used.  It was pre-dawn era of SQL systems.

During the development of the large-scale workflow support software information system for the national Register of Enterprises of Latvia, Lursoft Ltd. was set up by Gints and his team of co-founders as the vendor to provide key IT support services, software development and data processing services for the national Register of Enterprises of Republic of Latvia.

Register of Enterprises Latvia logo

Register of Enterprises was struggling to effectively process loosely structured corporate statutes, legislation documents, and other items having a lot of supplementary text in document format.  All IT support systems were primarily designed with a relational architecture, requiring to format and store all data heavily split among multiple tables, in so called normalized format.  The strong encoding of data was the ruling relational standard of that time. 

Gints, using existing relational technology to its full capacity, felt frustrated and saw the need to have a better tool of search of information in unstructured and semi-structured data, piling up on Novell Netware servers for this particular customer.   In fact, this huge information could not be organized by relational systems at that time: they were ridiculously limited by design.  Everyone who had worked with the first generation of table-type relational database files knew that it was not the best format needed for the requirements of the storage of documents. In the relational  database systems in 90ies all the information had to be strictly divided into tables and fields in heavily encoded format. There were severe limitations for data objects maximum size and file size, imposed by hardware and operating systems in 90ies.  For example, the largest document that could be stored in table format at that time systems could not even exceed 64 kilobytes, which by todays standards seem age old concept. The architecture of the relational databases provided fast information retrieval only if using strictly defined indexes.  It can not do fast searches in the textual data of free format such as office documents.  At the time the fastest processors were 25Mhz and 33Mhz and it was very difficult and time consuming to find required contextual information among hundreds of thousands of textual document files residing on PC hard disks or a Novell Netware server disks.  Frequently, simple context based searches took tens of minutes, but searching in large archives of documents  could take even several hours.

It was a very clear need to develop a better software solution to work with databases containing a mix of structured and unstructured data, enabling to manage tens of thousands of documents, used by the Register of Enterprises.   

Lursoft co-founder Gints Ernestsons, being also an agile innovator and computer research pioneer in emerging technologies, when facing this problem, started to look for non-traditional approach how to solve the fast search problem in document data, such as use of inverted indexes and existing mathematical algorithms of building graph indexes.  Being himself a database architect and a skilled software developer, Gints initially choose a relational database management system in 1992 to develop the first tool for full text search in documents as an extra application software for RDBMS.

Gints first vision at that time was to build the complete full text retrieval system doing fast search within textual information, where documents are located in file system, but all all data and indexes are in RDBMS table structure.  He started to work on that idea, developing design and system architecture, and then software for this kind of application, using sample document collections available in the Register of Enterprises: laws, regulations, company statutes etc. back

Fast Document Search System Project 1993

In 1992 a software source code was written by Gints for the prototype application of the document-based databases and search system on the basis of originally designed by Gints inverted index (a design concept frequently used in full text search systems).  He used the FoxPro database management system as the development environment for such a system.  In this computing environment documents were divided into separate words and phrases,  and saved in the system of linked matrix, creating the inverted index. This system had all the basic elements required for a fast full-text search, including inverse index building module, simple ranking algorithm of full text search results relevance positioning, and support of multiple national languages. 

The core prototype source code of the Fast Document Search system was developed by Gints Ernestsons in about 6 month time.  To check viability of the approach, another Gint's company software engineer, working under Gints supervision, agreed to finish the prototype in 1993 having Gints as a coach for his university diploma work.  At the faculty of Automatics and Computer Technology at Riga's Technical University the project "Fast Document Searching System" was successfully presented as the engineering graduation project.  Gints "brainchild" software remarkably got recognition from computer science professors for this novel software: it was showing different approach to data management to students, centered around document-oriented data model vs table data and relations.  Since then Gints is still being invited as a guest speaker giving lectures about emerging database technologies for computer-science students in two leading technology oriented universities: Riga Technical University and Ventspils Augstskola.

Riga Technical University logo

and

Ventspils Augstskola

Due to FoxPro's architectural limitations (basically FoxPro was a scripting language interpreting 4GL database commands) restricting overall performance of this prototype, Fast Document Search System pilot software was not implemented in any commercial software solution by Gints or his company.  However the project was a good test bed to test future concepts, algorithms and prototyping models showing potential of different index types compared to relational.  Gints Ernestsons later used this FoxPro based prototype document search software to set up some of his non-commercial Internet services for public authorities in 1995-1996, for example, providing for public administration the first Internet database of national legislation documents with ultra-fast and relevant search for documents by content.   It was also used internally for LURSOFT needs to access and search legal documents at speed and depth of search coverage impossible with relational technologies. back

Advent of SQL Era in mid-90ies: RDBMS and Enterprise Search Symbiosis

From mid 90ies most IT companies heavily invested into the infrastructure of SQL relational databases.  They gradually replaced so called 4GL-based DBMSes (Dbase, Paradox, Clipper, FoxPro etc.) with databases based on SQL servers.

Gints Ernestsons, in mid 90ies managing the implementation of the complete technical IT project of National Population Register in the role of CIO, became the very first customer for Oracle SQL server DBMS product in the country.  At that time Oracle SQL server was replacing aging Software AG Adabas database system and was a natural choice for a modern client-server database platform for the largest database of the Ministry of Interior, containing identity information about millions of country residents.  Oracle SQL was also later used by Gints for other public databases.
For other projects Gints choose less costly Sybase SQL server. 

Unfortunately SQL server systems were not designed to provide fast search in document-type textual data.  To address search issues for textual data items, SQL server systems usually provided some tools to integrate with enterprise search software.  This approach was also selected by Gints in LURSOFT Newspaper Library, that was a database collection of several millions of full text articles of Latvian central and regional press and media, please see www.news.lv:

News.lv - Lursoft Newspaper Library

To provide fast full text search for this huge document database, the commercial enterprise search platform Verity software (later acquired by Autonomy, UK for $600 million, now the later offered to be acquired by HP for $10 billion).

Verity software purchased by LURSOFT in mid- 90ies.  Verity search solution in LURSOFT was based on a Sybase SQL database server and operated as the additional full-text indexing and searching software system.  Although this solution worked fine for specific implementations, it was heavily integrated with SQL core system and very difficult to maintain and administer.  Still nothing much better existed in enterprise search tech at that time.  Under Gints leadership complete multi-tier application server system KAKTUSS was developed for Lursoft in 1995, that accessed Sybase SQL server with Verity functionality from tens of KAKTUSS application servers running on a large cluster of networked PC hardware in parallel and each servicing web requests incoming from thousands of simultaneous users in real-time and being load-shared and executed across all KAKTUSS application servers in a cluster.   Unfortunately Verity has severe index synchronization problems with core database content, and was poor to support national languages.  It was also too slow for Lursoft's needs as concerned company's mission-critical public 24/7 Internet services, starting to acquire more and more simultaneous users since their launch in 1995.  

Soon performance demands from all KAKTUSS application servers exceeded performance capacity of this Sybase SQL +Verity  enterprise search platform based technologies, that did not scale well in clustered multi-server environment and became a serious performance bottleneck for Lursoft Internet services. back

Siets Server Project 2001

At the end of the 90ies, Lursoft was also running about 40 other Internet-subscription based and public database services.  Many of them were requiring fast and relevant full-text search functionality.  

Gints Ernestsons decided to apply his own previously proved and original concept of the document-oriented database search software - Fast Document Search System - based on the rank-driven inverted index for LURSOFT needs.  The prototype software developed by Gints in the beginning of 90ies outlined the basic engineering principles of the construction and potential of the full-text search based indexing systems.  Emerging Internet Web services and growing data volumes required also new functionality not present even in legacy commercial SQL-based search engine and related enterprise search products (such as Verity): a way to achieve fast sorting of only the most relevant data out of very large data volumes.

That was foundation for a new search engine software project which Gints Ernestsons started in Lursoft with a group of his own hand-picked best C/C++ software developers, which he found and hired, using his good contacts, mostly from Riga Technical University and the Latvian University.  The new search engine project was implemented in Lursoft during 2001-2005.  In 2001 the first draft concept for a new search engine platform was designed in blue-prints by Gints and his developers team.  The new search engine technology was based on combination of one old and two emerging technologies: full text search, XML and clustering.  Directly managing his team of C/C++ software developers as a chief technology architect for a project, the resulting concept of the new full text search engine was soon implemented by Gints and his team into fast, scalable search engine software code capable of running in cluster architecture.

In 2002 Gints and his team of C/C++ software engineers created the first working software platform of the completely new search engine system and named it: Siets Server (Latvian  name 'siets' means 'sieve' in English ).

Siets.lv logo

Its core engine operated on native XML data.  In such a system all objects were stored in the server's index and were formatted as simple XML documents.  It allowed to distribute data and workload among many networked Siets Server computers to achieve linear performance scalability for indexing and at the same time maintain the same very high search speeds across the whole distributed search index.  Such a system has the maximum indexing and search speeds and is portable across different operating system platforms.   It has also basic information sorting facility enabling to rank the most relevant data.

Siets Server was successfully developed according to key concepts outlined above and installed on the Linux platform.

Siets Server search engine software started operating with the real Lursoft Internet database services in the spring of 2002, replacing Sybase SQL server software and Verity search engine software platforms for the Lursoft Newspaper Library database, and outperforming old Sybase+Verity installation by several orders of magnitude. back

Country-wide Internet Search Service: Siets.lv

Based on Siets Server software technology, Gints, working as a co-founder and CTO for Lursoft, set up the largest national Internet crawling and search engine service SIETS.LV.  It was run by Lursoft from 2002-2004.  Under Gints Ernestsons management and supervision, the SIETS.LV national Internet search service commanded about 25% of local Internet search market in 2003-2004:

Siets.lv national search engine screen shot minimal

The success of SIETS.LV was the proof of Gints originally designed document search software technology viability and scalability for very large scale clustered projects.  For example, daily indexing 2x as much national Internet data as the closest competitor and gaining popularity from users by providing 2 x wider and richer national Internet content coverage than any global Internet search engine at that time, and maintaining better index free from global "spam", SIETS.LV soon was among top 10 sites generating the most user traffic.  The complete system was running only on 5 clustered hardware servers handling some 10 million documents from the national Internet index at that time.

Unlike some implementations of indexes in Internet search services based on big-tables that work with a key-value or columnar databases, Gints technology for Internet search did not use that concept, instead using even more easy to understand and manage document-centric approach.  It fits well for most of Internet published data available on the Web (any HTML, PDF or other file on the Internet in essence is a document).

There were numerous publications about this Internet search service SIETS.LV in local media, for example, an article in the local magazine for IT professionals "E-pasaule" (E-world): Technology solutions for Internet Search describes well this project.

Eventually SIETS.LV services were discontinued by Lursoft as Gints Ernestsons business relationship with Lursoft ended in late 2005.  

Based on mutual copyrights agreement between Lursoft and Gints Ernestsons, the old Siets Server software can still be used for some internal projects by LURSOFT today.  Gints retained his copyrights to use the Siets search engine software technology in his own future business ventures, which he did.  

Together with his team of C/C++ software developers Gints started his own business enterprise in 2006: Clusterpoint Ltd.back

Clusterpoint Ltd.: A Vendor of New NoSQL Document-oriented DBMS (2006)

Innovative ideas prototyped in Fast Document Search System project, later advanced in Siets Server enterprise search engine software and proved in Siets.lv Internet search engine technology, outlined to Gints and his team of software developers the true business potential for a new full-blown NoSQL database management system platform.  Over 12 years the idea to develop a full-fledged DBMS system based on similar principles how to handle document oriented data, including a new ranking-based database content indexing approach for XML document structure pioneered by Gints Ernestsons and his team members in previous projects, materialized into "grand vision" and plan to develop the next generation database platform software, that can efficiently solve SQL technology shortcomings and limitations.

Himself being a professional database software programmer for more than 20 years, Gints over the last 30 years acquired expert knowledge on the best industry tools for very large scale database management.  This practice yielded also good understanding about shortcomings and limitations of legacy relational database systems.  Those systems, in Gints opinion, were inviting for innovation and performance-related problem solutions.  There have to be the better, more simple way how to manage huge databases.  It was clear, that in Internet era database management can not be effectively done using just old database management platforms.  Problems with SQL technology could be also traced along the source code origin history for SQL servers.

Most of the source code for DBMS server-based platforms was of legacy origin and rather outdated.  There were only a few completely from ground-zero written and originally developed database server source codes in the last few decades in the global IT industry.  Almost all C/C++ source code for industry leading SQL servers was traced back to the origin of IBM's System R / Ingres / Sybase and more than 30-year old architectures.  Most open source database projects were derivatives of previous undertakings and some older source code almost always served as a platform for next slightly improved products by new vendors.  Gints was thinking they were just adding improvements of old SQL RDBMS architecture, instead of changing architectural design that was required to make the new foundation for better database management.

Idea to write a fresh new source code for a completely new server-based non-relational DBMS  platform, without any legacy limitations, motivated Gints and his team of expert C/C++ software engineers to create a new dedicated business venture to implement this vision in 2006.

Together they have decided, that the source code for the new DBMS platform must be written from ground-zero and in C/C++, using contemporary multiple-CPU and multi-core CPU hardware, capable to use all of RAM, working with ubiquitous networking, and using unlimited cheap disk space, simplifying and automating everything possible in data base management.

The next Clusterpoint team started to look for investors who could provide capital for their new business venture.

They eventually found Imprimatur Capital, London (UK), and early stage venture capital company, managed by the same people who once invested in the future software powerhouse Autonomy (UK).  After Gints met with Andy Bottomley, who previously was a main board member of Durlacher plc, the UK LSE listed investment bank (now owned by a division of Lazard) specializing in high growth companies, who was managing his own new venture capital fund Imprimatur Capital when he first met with Gints, Gints learned about the future-gazing investment approach of the Andy's team of venture capital and checked his previous public references such as this interview where Andy Bottomley outlines similar investing principles:

"Future-gazing - or researching what types of companies might evolve over the next three years in a market that does not as yet exist - is a core part of the Durlacher investment process. Bottomley maintains that the decision to create Autonomy, one of Durlacher's greatest successes to date, evolved from research showing that intelligent search engines would become an essential tool on the Internet.'We evolve investment hypotheses to target opportunities,' he says. "  (Source: Dow Jones Financial News, May 15, 2000.  New breed fills digital economy advice gap).

Imprimatur Capital and Gint's team of software developers found common mutual interest to invest into new NoSQL software technology in 2006.  NoSQL market as we knew it by this term nearly did not exist in 2006.  A new database technology, pioneered by Gints, combined an XML-document based database platform software with an enterprise search software functionality.  It was a classic NoSQL database technology, looking from today's point of view.  The term "NoSQL" was not widely known and used rarely until 2009, when it reappear as designation for emerging database technologies that were not using relational architecture.

Together with Imprimatur Capital in 2006 it was decided to set up a new company and build a new software platform using the best of the design principles and know-how about non-relational database technologies, and, especially, database search,  accumulated by Gints and his experienced team of C/C++ developers. 

In August 2006 Clusterpoint Ltd. was established.  All developers, who were once hand-picked by Gints and did work with Gints since 2001, also became Clusterpoint Ltd. co-founders and together with Gints formed the new company's core R&D team.  back

Clusterpoint Data Base Management System 2007

Clusterpoint DBMS was born as a new database platform product in 2007, nearly a year after the Clusterpoint Ltd. was set up and commenced software development activities in 2006.  

In about 12 months under Gints Ernestsons inspirational management Clusterpoint's enthusiastic and dedicated team of professional software developers produced a new DBMS platform software source code in C/C++ implementing the concept of a new document-oriented NoSQL database management system:

Clusterpoint Server

It was a new, originally developed, well-engineered and complete database server platform software with API.  The resulting software product - Clusterpoint Server - was outcome of the right match between the early stage venture capital resources and many years of hands-on software engineering experience and database design knowledge acquired by Gints and his team members in previous non-trivial IT projects.  Without this previous teamwork, where all members acquired experience and software engineering knowledge, it would probably take many more years to implement a new full-fledged database platform, starting from the new DBMS design blue-prints transformed into the finalized, tested and robust software.  Coherent system design for the future DBMS platform, clear initial concept and vision of founders, and reliable and proved in production software engineering methods used by Gints teamover long years, resulted in the production of Clusterpoint DBMS software in a record breaking time.

The best know-how ideas and concepts from Fast Document Searching System software and Siets Server search engine software were naturally implemented in the new software system and advanced to the next level - a full-fledged DBMS, with innovative and completely original Indexing, Search and Clustering system.

Many enterprise-class design requirements were implemented to make the Clusterpoint DBMS software an easy to use and administer database management system.  Ranked database search engine concept was seamlessly integrated enabling to focus the new database management platform functionality on highly relevant and ultra-fast database search - a key feature for most modern web databases.

As the company name - CLUSTERPOINT implies -  one of the key design goals of the new DBMS software platform was creating fully scalable cluster functionality, which underlies any modern distributed database architecture.  As a result multiple database clustering management options were implemented - from simple database striping (cluster database), to running multiple copies of full database instances in parallel (database mirroring) with automatic database synchronization, and even distributed cluster mirroring.   All security, safety and monitoring features required by customers running 24/7 mission critical applications using Clusterpoint DBMS as their main data storage and database management platform were implemented: enterprise-level user management, security based on groups and roles, audit logs, single-click cluster-wide configuration changes, monitoring and statistics tools etc.  For large enterprise customers SNMP-protocol client software was implemented to enable visibility and monitoring of Clusterpoint DBMS servers in the corporate network.

Please see Clustering and Management.

The Clusterpoint DBMS software was extensively tested in production by selected customers for the next 12 months to bug-fix errors and achieve stability, required by 24/7 mission critical Internet services. back

Proof of Market Concept for Clusterpoint DBMS: 2008-2010

Since 2008 active sales of the Clusterpoint DBMS licenses in local markets proved the concept and maturity of product for the IT market.  The product was sold to leading information services, national business directories, the largest telecommunications and logistics companies in the country, using the classic enterprise software licensing and sales business model by Clusterpoint Ltd.

Clusterpoint database software mission critical customers

All of them are running Clusterpoint DBMS in mission critical 24/7/365 environment, providing core functionality for their Internet services.  To discover where Clusterpoint database technology has been successfully applied please see Customers.

In many cases Clusterpoint DBMS displaced two legacy products at once: an old SQL server and an integrated enterprise search engine.  Clusterpoint was reducing software licensing costs and operational expenses for customers: no need to buy database client access licenses, and no extra integration efforts with an enterprise search software were required.

Many customers started with some prototype testing on their own sample database, before they decided to purchase commercial Clusterpoint licenses.  One of the most frequent response from customers after they did their own initial tests on real data, was feedback that Clusterpoint DBMS outperforms legacy SQL-based solutions by several orders of magnitude, for the same types of database search transactions and the same results ordering.  Please see also our sample use case Calculate ROI.

It was interesting to watch also some customers, who after learning that Clusterpoint DBMS is not an open source product, decided to build their own integrated SQL-based solutions, using open source systems such as Sphinx or Lucene for search. They discovered that to set up and maintain coherent integrated systems built and patched with open source tools did not always turn up to be cheap.   Many of them came back to Clusterpoint, after spending months of their developers time (the most costly aspect in any IT software project today) with little progress and unresolved functionality problems.  Loosely (un)supported open source software can not effectively address details and service quality demanded by enterprise customers.  Need to integrate multi-vendor open source software only makes situation more complicated.  

Clusterpoint in a single platform DBMS server product provided needed main data storage and enterprise search functionality and also provided uniquely flexible information relevancy ranking.  Clusterpoint Server driven database system turned up to be more cost-effective platform product in the long run for our customers.  Customers started to replace their free open source databases and search tools with Clusterpoint, despite the extra licensing costs.

To learn more about business benefits and advantages of Clusterpoint database technology please see Customer Advantages. back

Clusterpoint Ltd.: 2011 and Beyond

Clusterpoint Ltd. is looking for a strategic partnership with a venture capital company or an established IT technology vendor to speed up the growth of Clusterpoint database platform software business internationally.  

Today Clusterpoint DBMS core database server software represent an ultra-fast and original C/C++ software source code which is cross-platform and portable, and delivers productive management of industry standard XML-only format databases.  The software copyrights and source-code are fully owned by Clusterpoint Ltd.  The software is distributed to customers in binary code only.  It compiles and runs natively under Linux, FreeBSD and MacOS operating systems. There is Windows version available as a virtual appliance, however, it is possible to port the Clusterpoint database software to run natively under Windows too (upon customer request).  Other vendor hardware specific UNIX ports of  Clusterpoint Server are also possible, if the customer is ready to provide hardware and OS for porting and testing to our team.

In short - we have a robust, reliable, technically well-engineered NoSQL database platform product, that solves customer problems in IT data management in uniquely productive way.  It is not an open source software, although we may launch open core community edition at some point (without enterprise features), if our investors convince us to do so.  We have a fully protected by copyrights source code at the moment.

Yet Clusterpoint Ltd. alone lacks sales "muscles" and brand awareness to conquer the global database software market.

We are interested to find a well-connected partner with knowledge, interest and capital resources to start a new Clusterpoint database technology based marketplace internationally, that requires significant resources beyond early stage risk capital investments used for the initial product development and the proof of market.

We provide 2 key opportunities to our eventual partner:

  • to challenge one of the world's most lucrative and profitable technology business sectors: enterprise database software, with the unique high-value core IT infrastructure software product - Clusterpoint DBMS platform, that simplifies customer database design, management and search, increasing productivity of customer business;
  • to become a market leader in emerging document-oriented NoSQL database technologies, delivering to customers uniquely relevant software product: high-performance industry standard XML-only database management system with  enterprise search functionality in a single platform, that natively scales out in modern private cloud IT architecture and with its innovative Information Ranking makes database search ultra-fast and always relevant.

Together with a strategic partner we can grow this our database software business internationally faster, compared if we do it at our own slower pace.  Provided our partner resources, we can set up a strong international management and sales team or use our partner's existing sales network, to establish market presence in different geographies all around the world.

We have enthusiastic and committed technology development team that focuses its efforts on advancing  of Clusterpoint DBMS software and maintain it a technically strong and well-supported product. 

We believe that technically we can address the challenges of the global database software market and together with the right strategic partner for sales capacity development we can quickly gain mutual benefits and establish a profitable business relationship.

Please see also Venture Capital Relationship.  In case of interest to learn more please do not hesitate to make us a phone call on +371-29479905, asking for Mr. Gints Ernestsons, or just send and email to Clusterpoint Management.

Enquiries about Clusterpoint DBMS software distribution partnership are also welcome.  Please send email to Clusterpoint Sales Email. back