{"id":658901,"date":"2024-06-04T15:00:12","date_gmt":"2024-06-04T19:00:12","guid":{"rendered":"https:\/\/siliconangle.com\/?p=658901"},"modified":"2024-06-04T12:25:14","modified_gmt":"2024-06-04T16:25:14","slug":"neo4j-integrates-dozens-graph-analytics-functions-data-snowflake","status":"publish","type":"post","link":"https:\/\/siliconangle.com\/2024\/06\/04\/neo4j-integrates-dozens-graph-analytics-functions-data-snowflake\/","title":{"rendered":"Neo4j integrates dozens of graph analytics functions with data in Snowflake"},"content":{"rendered":"

Graph database vendor Neo4j Inc. is teaming up with Snowflake Inc. to make a library of Neo4j\u2019s graph analytics functions available in the Snowflake cloud.<\/p>\n

The deal announced today allows users to access more than 65 graph algorithms using SQL without moving data out of the Snowflake environment.<\/p>\n

The algorithms cover a variety of common graph uses, such as identifying anomalies, detecting fraud, optimizing transportation routes, consolidating data records and offering purchase recommendations. Graph databases are efficient at handling large quantities of data and representing complex relationships between items in a way that is difficult to do with relational tables.<\/p>\n

Customers in common<\/h3>\n

Sudhir Hasbe, Neo4j\u2019s chief product officer, said an example of the graph query would be, \u201cFind user A and what review he wrote, what product he bought and what other products are linked to it.\u201d<\/p>\n

Hasbe said the partnership is a natural fit. \u201cBoth we and Snowflake focus on Fortune 2000 customers,\u201d he said. \u201cWe feel there\u2019s a significant overlap in the customer base.\u201d<\/p>\n

Users can also build knowledge graphs, which capture relationships between entities and are useful in building large language models. The arrangement doesn\u2019t give users full graph database functionality, but that wasn\u2019t the objective, Hasbe said.<\/p>\n

\u201cThis will primarily be algorithms, not full database capabilities,\u201d he said. Users who want full graph functionality \u201chave to get the tools from us to move the data.\u201d<\/p>\n

Gartner Inc. has forecast<\/a> that graph technologies will be used in 80% of data and analytics use cases, up from 10% in 2021.<\/p>\n

SQL support<\/h3>\n

A key feature of the partnership is the use of SQL to build knowledge graphs and run algorithms. \u201cIf we wanted to make it easy for people to adopt the technology, we had to go to where the customer is, and that meant we had to make sure that they could use the interface they already know,\u201d Hasbe said. \u201cWe enable all of these algorithms using the SQL interface so users don’t have to run Python code or build Python notebooks. It’s super simple to learn and get going.\u201d<\/p>\n

Neo4j uses Cypher as its native query language, but it isn\u2019t necessary to access the algorithms in Snowflake, he said. \u201cIf you define a graph, you need Cypher, but you call and run algorithms in SQL,\u201d he said. Data isn\u2019t stored in graph form in the Snowflake database but converted in an in-memory process to run the pre-defined algorithms.<\/p>\n

The serverless process creates ephemeral graph data science environments using Snowflake SQL, requiring customers to pay only for resources utilized during the algorithms\u2019 runtime. Payment can be in Snowflake credits.<\/p>\n

Image: Neo4j<\/h5>\n
\n

A message from John Furrier, co-founder of SiliconANGLE:<\/span><\/h1>\n

Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network<\/strong>, where technology leaders connect, share intelligence and create opportunities.<\/p>\n