Frequently Asked Questions
Is VoltDB a database?
While VoltDB does have its roots as an in-memory database, since we do a lot more than just store and provide access to data, we consider ourselves to be a lot more than just a database: we consider ourselves to be a data platform.
VoltDB is a tool for storing, accessing, analyzing and, most importantly, ACTING ON data in real time, at the crucial moment it needs to be acted upon. It combines the velocity of stream processing with the state-based consistency of an operational in-memory database and the decisioning intelligence of advanced analytics. So, applications can ingest high velocity data feeds from multiple sources, enrich that data with historical insights, and trigger an immediate and meaningful action
Is VoltDB a BSS?
No. VoltDB is not a business support system (BSS) or a BSS-related application. However, VoltDB does support the building and deploying of powerful BSS and OSS applications via the enablement of sub 10-millisecond decisioning.
What systems does VoltDB integrate with?
VoltDB seamlessly integrates with all of the major database technology systems, including queuing systems such as Kafka, containerization systems such as Kubernetes, and big data storage systems such as Apache Hadoop.
However, while VoltDB was designed to become a part of your tech stack and work well with any layer, it was also designed to be able to empower enterprises to simplify their stacks and eventually run everything through Volt to significantly decrease latency and the potential for downtime.
Why is sub 10-millisecond decisioning so important?
The advent, proliferation, and combination of 5G and IOT has significantly raised the bar around latency. Because there’s now so much more data to deal with than ever before, things now need to happen much faster than ever before. Your applications can no longer afford to send data to a lake or warehouse for processing, because in the time it takes to make that trip, the opportunity to either upsell a customer or prevent fraud has been lost. If 250 millisecond is the new gold standard for something to “happen” (ie, such as a page loading or a financial transaction going through), then the decisioning part of that process needs to happen in under 10 milliseconds. VotlDB is the ONLY data platform capable of such speeds.
ACID? SQL? This sounds like you’re talking about old school legacy technology. Is that what you are?
No. What we are is new-school technology incorporating the best and most currently relevant “old school” attributes. The truth is that most of our customers come to us after experiencing quite a lot of headaches—and heartache—with their legacy databases, which simply can’t keep up with the data volume, variety, and velocity that 5G and IoT throw at them.
Most NoSQL databases are fast in part because they simply don’t do transactions in the traditional RDBMS sense, and went to market being clear on this point. Over time, we’re now seeing NoSQL vendors attempt to retrofit ACID transactions with varying degrees of success, and with many caveats and creative re-definitions of the term. This is because as NoSQL users broaden their use cases, they find that not only do they need ACID transactions, but that getting developers to implement them is a time consuming and risky activity. VoltDB was designed from the ground up to handle large numbers of transactions at a time.
Likewise with SQL. As applications scale and become more complicated, the need for standard data structures and a standard way to manipulate them becomes increasingly urgent. That’s why many NoSQL vendors who started out as pure document or key-value databases are now either erecting SQL facades or coming up with a query language whose name ends in ‘QL’. VoltDB anticipated this and has been SQL all along.
You talk a lot about speed. What about scale?
For us, speed and scale are the same thing. That is, the problems enterprises have with scaling their database-related use cases ultimately come down to SPEED. They can be fast but only at certain data volumes. Or — they can scale easily and affordably but only if they are willing to sacrifice some latency and do everything slower.
VoltDB offers horizontal scalability to support the increasing volume and variety of data, without compromising on latency. Scaling only vertically—ie, by adding CPU cores—very quickly puts your applications into trouble with latency and uptime because of the overhead involved and the extra data center trips your data packets have to make. As an in-memory data platform, VoltDB is optimized for both horizontal and vertical scaling because it’s not bound to disk and can use partitioning to scale horizontally and multi-core support to scale vertically.