The Intelligent SQL RDBMS for Application Modernization
Splice Machine is an integrated platform that can scale, modernize, and extend the functionality of custom applications that drive competitive advantage for your enterprise. It is an intelligent SQL RDBMS that enables companies to be agile, data-rich, and intelligent without the expense of re-writing their applications, and the assurance that the database will not become a bottleneck when faced with an unexpected crisis or opportunity.
Full ANSI SQL Syntax
Splice Machine’s full ANSI-SQL capabilities makes it the perfect platform for migrating applications from traditional RDBMS databases to a scalable and cost-effective data platform that enables enterprises to leverage the investment they have made in SQL-trained resources. Applications interact with Splice Machine over standard JDBC or ODBC connections. The platform offers native PL/SQL support to facilitate migrations of applications that use this procedural language extension. Splice Machine 3.0 supports specific DB2 extensions that make it easier to migrate from DB2 with minimal SQL rewrite. Examples include DB2 Trigger syntax, error codes, etc.
Splice Machine supports constraints, referential integrity, primary keys, triggers, secondary indexes, custom stored procedures, and functions.
Hybrid Storage
Splice Machine provides the flexibility to store data based on the format that optimizes the way users need to access it. Row-based storage is write-optimized for single-record lookups and short scans, whereas columnar storage is optimized for large table scans, large joins, aggregations, and groupings. In addition to its internal row-based representation, Splice Machine supports external ORC and Parquet columnar tables and object stores like S3 or ADLS. By allowing companies to use lower-cost, higher latency storage options, companies can obtain significant cost savings due to the hybrid storage capability. It also speeds up large OLAP use cases that benefit from the columnar layout.
Splice Machine also provides the ability to store semi-structured including XML and JSON files or unstructured data directly in database tables or through integrated Spark functionality.
Full ACID Compliance and Multi-Row Transaction Support
To power mission-critical applications, Splice Machine’s patented distributed transaction system maintains ACID properties across multiple tables, records, constraints, triggers, and indexes to enable both low-latency reads and write-scalability. Splice Machine uses a snapshot isolation design that implements Multi-Version Concurrency Control (MVCC) to create a new version of a record every time it is updated, instead of overwriting the old data value in place. This means that users have access to the data that was available when they began reading, even if that data has been subsequently updated by a writer. With each transaction having its own virtual “snapshot,” transactions can execute concurrently without any locking. This leads to very high throughput and avoids troublesome deadlock conditions.
Distributed ACID Transactions
Download this white paper to learn about our implementation of distributed ACID transactions, and how it enables Splice Machine to deliver high concurrency and low latency for both OLTP and OLAP workloads.
Hybrid Transactional and Analytical Processing (HTAP)
Splice Machine has a unique “Dual Engine” architecture that provides outstanding performance for concurrent transactional (OLTP) and analytical (OLAP) workloads. The Dual Engine architecture gives you the best of multiple worlds in a hybrid database: the performance, scale-out, and resilience of scale-out NoSQL, the in-memory analytics performance of Spark, and the enterprise sophistication of a cost-based optimizer.
The Splice Machine SQL parser and cost-based optimizer analyze an incoming query and then determine the best execution plan based on the query type, data sizes, available indexes and more. The Splice Machine cost-based optimizer with advanced statistics executes compiled byte-code for OLTP-type lookups, inserts and short-range scans on HBase and lightning-fast in-memory processing of analytical workloads on Spark. Splice Machine implements “sketches” to compute cardinalities, which can produce results orders-of-magnitude faster and with mathematically proven error bounds. The combination of the cost-based optimizer and the right statistics can reduce query response times from hours to minutes and seconds by rewriting subqueries, choosing indexes, ordering joins and selecting distributed join algorithms.
Resource Isolation
Splice Machine isolates the resources allocated to OLTP and OLAP workloads from each other, so each can progress independently of each other. For analytical workloads, the Splice Machine provides the flexibility to allocate resources (CPU, memory etc.) at the user or group level. Resources can also be distributed across all the running queries or a minimum or maximum threshold can be specified. Combined with a multi-version concurrency control (MVCC) locking mechanism, this ensures that the performance of transactional workloads even when large reports or analytic processes are running.
Scale-Out Power
Splice Machine allows you to seamlessly scale-out from gigabytes to petabytes as data volumes change. Splice Machine’s massively parallel scaling architecture can easily handle workloads that would overwhelm traditional databases. With Splice Machine Cloud Manager, configuring a new cluster is as easy as using a few sliders to set compute units for OLTP, OLAP, and ML processing, allocate storage, and schedule backup frequency and retention.
In-Database Machine Learning
Splice Machine’s ML Manager provides native data science functionality with in-database analytics, integrated Jupyter notebooks, and native workflow management and deployment capabilities with MLflow. It has never been easier to enable machine learning in a production application and provide the governance capabilities to audit models in production.
Hybrid Implementation
Splice Machine can be deployed on bare metal or Kubernetes whether your infrastructure is on-premises, AWS, Microsoft Azure, or Google Cloud (soon). Splice Machine on Kubernetes abstracts away the underlying infrastructure and makes hybrid cloud and multi-cloud deployments straightforward.
See What's New in Splice Machine 3.0
Learn more about the latest release of Splice Machine. In this hands-on webinar, you will get a preview of the functionality included in Splice Machine 3.0.