Robust and Scalable Distributed Architecture
LongDB's modern Lambda architecture avoids the complexity of the traditional Lambda architecture and data redundancy, and supports both streaming and batch data processing on the same platform. Users can access the platform using mainstream programming languages and common interfaces like JDBC/ODBC. They can also connect to the platform using seemlessly integrated native spark data source, manipulating Spark Data Frames using Scala and Python.
LongDB cluster can grow from a few nodes to thousands of nodes (TB to EB) to handle data services with varying scales of requirements.
Cost-effective and Versatile HTAP Engine
LongDB is an HTAP (hybrid transactional and analytical process) engine based on mature open source ecosystem components like Apache Hbase and Spark. It is a distributed relational database. It is also capable of storing and processing semi-structured and unstructured data with high efficiency. Different from in-memory databases where data processing remain in main memory, LongDB's cost-based optimizer can intelligently determine the access paths, execution strategies, and parallel algorithms based on incoming workload patterns. What's more, it will adjust data placement overtime based on the learned patterns and put hot data in memory while leaving warm and cold data in different storage layers. This will not only improve query performance but also reduce the cost of operations and maintenance (TCO, total cost of ownership). This is achieved because of the years of experiences from the founding team members working on industry-leading MPP database system and optimizers.
High-performance Computing Architecture
LongDB contains a high-performance distributed computing architecture that can push computation to distributed data shards. It performs massively parallelized predicate evaluation, aggregation and function operations. The job scheduler can also achieve high parallelism to increase the system resource utilization. By utilizing system resources efficiently and effectively, the platform can handle OLTP and OLAP workloads simultaneously while achieving a good overall balance.
High Integration and Compatibility
LongDB platform supports standard protocols for access, for example, ANSI-SQL, JDBC, ODBC, REST. A wide range of existing tools can seemlessly conenct to LongDB and immediately get value out of a powerful platform. This will eliminate costly and unnecessary rewrites when customers want to migrate their applications to the new platform. What's more, we provide a set of easy-to-use tools to make sure the data and application migration is smooth. Since the platform is built on some of the most widely used open source components, it will also make the platform co-exist with clients' current big data environment.