Know the disruptive technology running into LeanXcale
LeanXcale can offer all this set of functionality thanks to a never seen technology, as a result of 20 years of investigation of some of the most brilliant minds in the database field, leading by the former UPM teacher Ricardo Jimenez-Peris.
Run on a PC or in 100s of servers, always get a linear performance. Process up to millions of transactions per second. Store up to petabytes. Query up to billions of rows instantly, thanks to Iguazu Tech.
Hybrid Transactional Analytical Processing
Real-time analytics: analytics queries directly from operational data. No ETL, no data movements, high performance.
A novel active-active replication algorithm: No more updates lost nor downtime due to passive replication. No more bottlenecks nor unacceptable costs due to obsolete active replication.
Our novel non-intrusive data migration algorithm allows to move data while is being updated and keeping full ACID consistency and without disrupting operations. Grow or shrink according to current needs with zero downtime. Minimizes operational costs by reducing HW resources to actual needs
Dual SQL and Key-Value Interface
One database two interfaces: Key-value for ultra-efficiency and semistructured data, SQL for simplicity and query-power.
Ultra-Efficient Storage Engine
Designed to work efficiently in multi-core and many-core HW. Ultra-NUMA efficient.
View and join information among all databases portfolio. LeanXcale can query MongoDB, HBase, Neo4J and your RDBMS in their native API in a single view combining the information to create a single resultset.
Integration with Data Lakes
When defining metadata and parsing of HDFS files, they become read-only SQL tables. SQL queries can query and correlate operational data and data lakes.
Compute parallel aggregations in insertion time without conflicts. Aggregate analytical queries become costless single-row queries.
Efficient for Range Queries & Random Updates:
As efficient as B+-trees for range queries (used by relational DBs). As efficient as LSM-trees for random updates/inserts (used by key-value data stores).
Costless Multi Version Concurrency Control:
Novel MVCC with almost zero overhead. Avoids stop-the-world obsolete version cleaning. Avoids resource waste for multi-versioning.
Hardware acceleration exploiting SIMD instructions. Acceleration of scans with predicates. Acceleration of aggregation. Acceleration of sorting
Efficient Distributed Secondary Indexes
They scale as distributed indexes. They are as efficient as centralized indexes.
Traditional Dictionary Compression
Efficient storage for columns with predefined sets of values. Efficient predicate evaluation over these columns
Novel Dictionary Compression:
Efficient storage for columns with dynamic sets of values & high redundancy. Same performance as traditional predefined dictionaries.
Encryption using hardware acceleration. Encrypted storage. Encrypted transmissions.
FPGA Accelerated Query Processing:
FPGA acceleration of query algebraic operators. Yielding analytical query acceleration of 10-100x.
Continuous Dynamic Load Balancing
Take advantage of the capacity to mo migrate data with no impact: balancing dynamically your load maximizing performance.
Compression of columnar storage (RLE, max-min, etc.). High levels of compression for columns with redundancy.
Hybrid Row-Columnar Storage
Query acceleration through columnar technology. Non-intrusive background columnification