Get Graph Insights Faster

Your data is never sent to any LLM when you are using our agentic query and schema creation features. The LLM you chose to work with will only be sent the labels of your data.  For example, if you were to look at customer data, we would only be sending the words “Name”, “Address”, &  “Phone Number”, but not the Names, Addresses, or Phone Numbers themselves. 

Rocketgraph has the ability to bookmark your favorite queries, as well as record your query and prompt history, so if you ever have a query you want to save, just bookmark to come back to it later.

No. The closest description to Rocketgraph is that it is an in-memory graph analytics platform designed for high-performance at all scales. This was an intentional choice during the initial development of Rocketgraph. Our team found that building a database-centric graph engine built in structural limitations to performance, particularly with whole graph search or unbounded search queries. 

Because we are not a database, we can be an unobtrusive addition to any existing analytics environment. With specialized connectors to Oracle, Neo4j, Snowflake, and many more, Rocketgraph is built to  act as an accelerator to your existing workflow. 

Rocketgraph thrives on big data & query complexity. For starters, we handle a slew of use cases that every other graph software handles. Everything from Cybersecurity to Pharmaceutical research. Graph tools are built to be able to analyze relationships. We can analyse the same relationships but at a much larger scale than other solutions. Allowing you access to fast query responses on queries 100’s of times larger than other solutions are capable of.

Rocketgraph is frequently used on-premise by organizations who have either 1/ An existing on-premise footprint which would make cloud deployments difficult or 2/ Security requirements that limit the ability to run it in the cloud. Aurora is well suited to organizations who already have extensive data resources in the cloud or who want the ease of a fully managed cloud graph analytics platform.

Yes it can. In fact, some of our technology partners integrate with us because we have worked with them to create knowledge graphs which enhance their offerings.

Yes. In fact, we provide a dataset with Rocketgraph which can be used to try this out, and we have produced a series of two minute tutorial videos located here to show this process.

In theory, yes. But if you are already using Neo4j, Rocketgraph is well-suited for complimentary work. For example, Neo4j is quite good at bounded graph searches, whereas Rocketgraph excels at full graph scans or unbounded searches.

Yes. In fact, we have specific features to support path finding with Breadth First Search. We recently published a blog on this topic which can be found here.

We have closely integrated AI with the user-experience so analysts can get started with graph more quickly and produce more insights faster than they could previously.

The answer is more complex than a simple yes or no. But we have run a benchmark comparing the two which shows that Rocketgraph performs better than Neo4j on larger graphs and more complex queries. The benchmark can be found here.

First it’s important to set the context on how we evaluate the size of a graph. In our experience the best way to estimate the amount of work that needs to be done to search a graph is with the total number of edges. Nodes typically represent entities, like people, phones, computers, etc. whereas edges represent connections. So while we might have a graph with thousands of nodes, there could be tens of millions of connections. Rocketgraph has been used for graphs ranging in size from a few thousand edges to hundreds of billions.