26 Aug It’s Not Just Hype – Graph is Transforming Businesses
First, the good news.
In a recent report, Top 10 Data and Analytics Technology Trends That Will Change Your Business, Gartner Research identified graph database and analytics as a key technology. Gartner forecasts that, “Graph analytics will grow in the next few years due to the need to ask complex questions across complex data, which is not always practical or even possible at scale using SQL queries.”
Not only that, graph technology enables other trends described in the same report. Chief amongst these is augmented analytics, which uses machine learning (ML) and artificial intelligence (AI) techniques to transform how analytics content is developed, consumed and shared. Without graph technology augmented analytics would be ill-equipped to offer meaningful insights as graph generates new features based on the relationship analysis to feed the machine learning models and improve their accuracy.
Another study from Gartner, Hype Cycle for Data Management, predicts that graph database management systems (DBMSs), “will increasingly be used to navigate both existing and newly discovered relationships more efficiently than relational processing over the next two to three years.”
Last, and definitely not least, in its “Hype Cycle for AI” study, Gartner included graph technology in two places on the curve, Knowledge Graphs and Graph Analytics.
Gartner forecasts that, “The application of graph processing and graph DBMSs will grow at 100 percent annually through 2022 to continuously accelerate data preparation and enable more complex and adaptive data science.”
Clearly, graph technology is hot.
Previous Generations of Graph Technology Haven’t Delivered.
Now for the bad news.
If you look closely at the Hype Cycle for Data Management mentioned above, you’ll see that Gartner has positioned graph DBMSs at a place on the cycle where it’s heading into the Trough of Disillusionment. Ouch!
Look carefully at the Hype Cycle for AI too, and you will see that, while knowledge graphs are moving towards the Peak of Inflated Expectations, graph analytics is already heading into the Trough of Disillusionment, according to Gartner.
This is because the early generations of graph technology have failed to fully deliver on what they promised.
Scale-out Graph Fulfills the Promise
We are in an age of artificial intelligence that offers the potential for deep learning to uncover new patterns and provide insights that will drive innovation for another 20 years. Deep learning, however, requires data, lots of data and the more examples of a trait or relation that can be identified, the more accurately deep learning can pick out patterns.
Graph technology is the perfect fit for this – its expressiveness is ideal for creating business logic algorithms (network analysis, abnormal pattern matching, community detection) across complex data. However, performance and scale have held back graph’s progress. For example, to find deep information about a user may require traversing across many tables, which in turn requires many joins and/or key value look-ups. Results are obtained slowly, if the query completes at all.
With a native parallel graph, on the other hand, all connections are established in the graph at data loading/updating time – no joins are required and queries are typically completed in seconds, or less. We are in the age of Hybrid Transactional Analytical Processing (HTAP) and have a rich library of algorithms that quickly traverse graphs to find insights in real-time. According to Dan McCreary, a respected authority on data analysis, if you’re not leveraging graph algorithms you are missing out.
Driving Unprecedented Innovation
Graph technology is now in its third generation and the following chart shows how the technology has evolved:
Graph technology has made huge strides. Third-generation graph includes a rich array of functionality that was lacking in earlier versions and is transforming business by providing the data scientists at those organizations with unprecedented abilities to understand the meaning in the connections in their data. Some of the functionality that’s essential includes:
• Supporting deep link analytics (in other words, going beyond three hops) to reveal connections that are necessary for next-generation fraud detection, recommendation engine, artificial intelligence and machine learning and other use cases.
• Scaling to multiple CPUs and across multiple machines for storing and parallel query processing of terabytes of data that yield results in seconds or minutes, instead of hours or days.
• Protecting sensitive data in ways that enable the privacy requirements associated with GDPR and CCPA to be met.
TigerGraph exemplifies third generation graph performance. TigerGraph addresses what Gartner describes as a need for data and analytics leaders to “consider the ecosystem holistically in order to get value most efficiently from your data and analytics landscape” and that “it no longer makes sense to evaluate analytic and operational use cases in isolation.”
Don’t Just Take Our Word For It
A recent study by researchers at the University of California Merced compared the performance of Graph 3.0 and Graph 1.0 technologies – in this case, TigerGraph and Neo4j. The study was a complete implementation of the Linked Data Benchmark Council Social Network Benchmark (LDBC SNB), considered by many to be the reference standard for evaluating graph technologies.
The research compared the different generations of graph technology in their loading, storage, and execution of 46 queries across a range of short-running (OLTP) and long-running (OLAP) inquiries. The results were revealing, if not unexpected.
The researchers found that the TigerGraph solution consistently outperformed the legacy Neo4j technology, by more than 100 times faster in some instances, with that gap expanding as the amount of data increased. Moreover, the Neo4j technology was unable to complete 13 out of 25 queries within the five-hour time limit.
This independent study aptly demonstrates the following three things:
• TigerGraph is the first scalable graph technology to demonstrate both analytics and transactional processing capability.
• Only TigerGraph could scale to 1TB for deep analytical queries.
• TigerGraph shows significantly compressed storage on typed property graph data as compared to Neo4j.
Don’t Settle for Legacy Technology
There’s never been a better time to upgrade to graph 3.0 technology. Businesses who have done so are beginning to outperform their competition. You can read reviews by people just like you who already using our technology here.
Gartner predicts that graph technology will “enable more complex and adaptive data science” but, as we have seen, this will only be possible with third-generation graph technology. TigerGraph’s technology is transforming businesses – don’t be left behind using a legacy system.