CLUSTERPOINT: Distributed Storage
Clusterpoint delivers cohesively engineered document-oriented database, distributed cluster computing and search software in a single cost-efficient database server and API platform.
Use Clusterpoint to design and operate instantly responsive real-time databases, capable to drive distributed demand of web and mobile applications and to provide millisecond database search, querying and analytics on semi-structured data from billions of XML and JSON documents.
You can develop application software for Clusterpoint database running on just one cluster node (hardware server). Then you can add cluster nodes to scale out your database volume and add replicated copies for safety and load-balancing, all without changing a single line in your application software source code.
Clusterpoint operates on a pool of inexpensive commodity hardware installed with operating system and Clusterpoint Server software. Software aggregates all cluster hardware computing power into a single masterless shared-nothing cluster database architecture. Data is stored as machine-readable schema-free XML and JSON documents and all database operations automatically support cluster-wide MapReduce, there is no need to program M-R tasks explicitly.
Our customers get horizontal scalability, high-availability and awesome database querying speed out-of-the-box.
They also get a unique capability to query their databases using full text search queries that uses ranking index. The last feature is a unique in Clusterpoint architecture and it delivers a great end-user experience where the database search performance and quality is rivaling or exceed what customers expect from the Web search.
You can find Clusterpoint API documentation in the section Developer's Guide.
All productivity features in one DBMS platform and one API
Clusterpoint DBMS cohesively unifies key enterprise data management techniques into one software platform. Previously these features were pre-dominantly used as isolated software components made by different product vendors. They required sophisticated integration through customer application software. Their maintenance costs were high, often resulting from resolving incompatibilities between different architectures, from corupt indexes and from the need to learn and master how to patch together several technologies until a reliable deliverable.
With Clusterpoint database software there is no more need to spend time integrating different vendor software to achieve the same data management tasks. Our customers can start working radically more productive spending their development and maintenance efforts on business application software, without painstaking integration of components.
- Built-in MapReduce Clusterpoint API automatically performs cluster-wide Map-Reduce for all database operations and no complex M-R programming is required from a developer;
- Fault-tolerance & Replication Clusterpoint supports fault-tolerant replication into multiple load-balanced copies;
- High Avaiability Clusterpoint guarantees high-availability no-single-point-of-failure operations;
- Linear Scale Out Ability Masterless clustering for scale out ability to petabytes;
- Enterprise-grade Search Clusterpoint delivers built-in enterprise search features for ultra-fast full-text search across distributed semistructured machine-readable document data such as XML or JSON;
- Ranking for Superior Relevance Clusterpoint provides a unique customizable ranking index that enables to scale out data and index while maintaining sub-second latencies, enables search databases across all their content using web-style full-text search queries and instantly retrieve the most relevant results sorted by their contextual match to the query terms.
These features together enables to use Clusterpoint as a novel high-performance distributed database storage software to service distributed web and mobile demand at previosuly unachiveable productivity levels. Our customer TCO gains on a 3-year period who migrated from SQL database products to Clusterpoint DBMS typically are over 60%, mostly in saved human costs and reduced software development time.
Distributed NoSQL database with the focus on search productivity
Clusterpoint database operates as a generic and linearly scalable cluster database storage for XML documents. It can grow to billions of XML data objects, while maintaining search and analytical query speeds within low sub-second latency range. Clusterpoint database also supports JSON over REST API.
Clusterpoint software was engineered for speed and performance in parallel computing architecture, with the focus on database search, hence its name.
Clusterpoint database architecture is modeled along the Web architecture itself. Any Clusterpoint database when distributed across large number of hardware servers (cluster nodes), operates very similarly to the Web "database".
The Web consists from billions of HTML documents linked by unique URLs pointing to them. The Web can be considered as the world's largest distributed document-oriented database. In fact, the Web has long been optimized through web server software to deliver the best latencies possible when accessed by users.
Clusterpoint database software does about the same for XML. Similarly to the ease of accessing and searching worldwide Web HTML pages today, Clusterpoint database works with XML documents. They can be stored, searched and accessed following the same distributed computing paradigm of the Web.
Super-fast database indexing method engineered for real-time Big Data access
The Web-like architecture might seem too simplistic and trivial solution for a corporate database, where data is stored as documents. We all know that essentially any XML is just a text document data format. When data was stored as text documents in legacy systems, it was slow and clumsy process. Legacy tabular data organization in SQL and its indexing system software was never designed for the efficient text handling.
We engineered an innovative hybrid data indexing that delivers high-performance advanced search, querying and analytics facility for semi-structured XML and JSON data content stored into the Clusterpoint database.
This advanced index actually "turns" Clusterpoint's simple XML data storage into the powerful database management system.
Technology delivers real-time web-like FULL-TEXT search and SQL-like query through a single API, maintaining low sub-second search latency even in billions of document-oriented XML data objects. Ranking index also enables massive scale out ability to very large Big Data volumes in a distributed cluster architecture.
Easily integrate with any existing systems using open API
We made interoperability with any Clusterpoint database a dead-simple task. Clusterpoint API uses simple messaging based on open web standards and protocols. You don't need complex ODBC/JDBC drivers, proprietary client side software or sophisticated middleware gateways to work with Clusterpoint. Connect and "speak" to Clusterpoint database directly from your favorite programming language.Any Clusterpoint database can be set up to operate as a distributed XML data store that runs across a pool of networked hardware servers, connected into a cluster. Like all those variable HTML pages existing on the Web, custom free format XML documents can be stored into the Clusterpoint database. Clusterpoint is a schemaless storage, exactly like the Web is. Your XML documents will be stored unchanged in Clusterpoint database.
There is a clear guiding principle in using Clusterpoint database - what you store into Clusterpoint database as your XML data objects, is what you get out from the database. We don't change anything in your data content. Your data just needs to be correctly formatted in XML. Documents in JSON can be stored and retrieved over REST API.
Reduce your database TCO using cost-efficient Rack&Stack commodity hardware
One can scale out Clusterpoint databases in the most cost-efficient way: using inexpensive Rack&Stack COTS hardware, that might be white label boxes or even self-assembled from components systems.
There is no extra software required for Clusterpoint database operations but a bare-bone operating system such as Linux, MacOS or FreeBSD. Clusterpoint Server needs only underlying operating system and its local file system.
When installed with Clusterpoint software (same software on each hardware node), you have set up entire Clusterpoint database infrastructure on a cluster. Provided that you have enough hardware, all Clusterpoint databases can be sharded across multiple cluster nodes, and replicated into multiple copies running in parellel. In the result you are running masterless, high-availability cluster database infrastructure with multiple load-balanced and fault-tolerant replicas.
It simplifies database hardware provisioning and enables upgrades through incremental steps, rather than through fundamental replacement. Hardware management largely becomes independent from Clusterpoint database services uninterrupted provisioning.
Develop scalable database software applications from day-one
Clusterpoint database software cost-efficiently delivers unlimited scale out ability and replication of your database services in a data center. Clusterpoint delivers all of that valuable functionality out-of-the-box, without the need to change a single code line in your application software code. You need to write application software only once and the resulting code can be later run in large database cluster to work with billions of data objects. There is no application code changes required to run many-times replicated, load-balanced, high-availability fault tolerant Clusterpoint databases.
Similarly, use enterprise search features out-of-the-box, they are part of Clusterpoint API. No need to integrate complex external search software tools, saving your time and money. Clusterpoint's rich enterprise search options provide the same scalability as XML data store. Full-text search indexing is executed in real-time along XML document CRUD updates and structured data indexing. It all works cohesively in Clusterpoint database, delivering native scalability and consistency for your application software from day one.
It simplifies and future-proofs our customer application software code, enabling to focus development solely on business functionality and on delivering of great user experience. Clusterpoint database software will take care about operational database management for scalability, availability and feature rich real-time access.Contact us to learn more about Clusterpoint use as a distributed data storage!