As enterprises proceed to speculate heavily in advanced analytics and enormous language models (LLMs), graph technology has grow to be some of the popular approaches to constructing the information stack. It allows users to know complex relationships of their data sets which might be often not apparent in traditional relational databases.
However, maintaining and querying graph databases alongside traditional relational databases is sort of laborious (and expensive). Today, PuppyGrapha San Francisco-based startup founded by former Google and LinkedIn employees raised $5 million to fill this gap with the world's first and only Zero ETL query engine. The engine allows users to question their existing relational data as a unified graph without the necessity for a separate graph database and lengthy extract-transform-load (ETL) processes.
The engine was launched in March 2024 and is already getting used by several firms to simplify data evaluation. The eternally free developer edition alone is seeing a 70% increase in downloads month over month.
The need for PuppyGraph
A graph database architecture mirrors sketching on a whiteboard, storing all information in nodes (representing entities, people, and ideas) with relevant context and connections between them. This graph structure allows users to discover complex patterns and relationships that will not be easily apparent in traditional relational databases (query over SQL) and supply algorithms to quickly enable use cases akin to AI/ML, fraud detection, customer journey mapping, and risk management for networks .
As things currently stand, the one strategy to adopt graph technologies is to establish a separate native graph database and keep it in sync with the source database. The task sounds easy, however it becomes very complicated as teams should arrange complex and resource-intensive ETL pipelines to migrate their datasets to graph storage. This can easily cost tens of millions and take months, leaving users unable to execute critical business queries.
Not to say, once the database is ready up, it also must be continually managed, which further increases costs and results in scalability issues in the long term.
To fill these gaps, former Google and LinkedIn employees Weimo Liu, Lei Huang and Danfeng Xu joined forces and founded PuppyGraph. The idea was to offer teams a strategy to query their existing relational databases and data lakes as graphs without data migrations.
In this fashion, the identical data analyzed using SQL queries may very well be analyzed as a graph, leading to faster access to insights. This could be particularly useful in cases where the information is closely linked to relationships at multiple levels, akin to in the provision chain or cybersecurity space.
“The deeper the extent, the more complex the query becomes in a conventional SQL query. This is because each additional level requires a further table join operation, adding complexity and potentially significantly slowing query performance. In contrast, graph query handles these multi-level relationships far more efficiently. They are designed to quickly traverse these connections using paths through the graph, whatever the depth of the connection,” Zhenni Wu, who joined PuppyGraph’s founding team, told VentureBeat.
According to Wu, PuppyGraph completely eliminates the necessity for extensive ETL setups and enables “deployment to question” in nearly 10 minutes. All the user must do is connect the tool to the information source of their alternative. Once done, it routinely creates a chart schema and queries the tables in chart models. Additionally, the engine's distributed design allows it to handle extremely large datasets and complicated multi-hop queries.
It can hook up with all popular data lakes, including Google BigQuery and Databricks, to perform accelerated graph evaluation – while keeping costs down.
“The separation of storage and compute architecture implies that low price is one in all PuppyGraph's biggest benefits. There aren’t any storage costs because the engine queries data directly from the user's existing data lake/warehouse. “It provides the flexibleness to scale computing resources as needed and make adjustments to efficiently handle fluctuating workloads without risking resource contention or performance degradation,” Wu added.
Significant impact within the early days
Although the corporate is lower than a 12 months old, it’s already seeing success with several firms, including Coinbase, Clarivate, Dawn Capital, and Prevelant AI.
In one case, an organization switched from a legacy graph database system to PuppyGraph and was in a position to reduce its total cost of ownership by over 80%. A number one financial trading platform was in a position to perform a 5-hop path query between Account A and Account B across around 1 billion edges in lower than 3 seconds.
Before PuppyGraph, their homegrown SQL-based solution couldn't even query greater than a 3-hop query and had issues with batch timeouts.
With this funding, the corporate plans to speed up its product development, expand its team, and increase its market presence by making the Zero ETL Graph query engine available to more organizations worldwide.
Accordingly GardenerThe graph technologies market will grow to $3.2 billion by 2025, with a compound annual growth rate of 28.1%. Other players on this category include Neo4j, AWS Neptune, Aerospike and ArrangoDB.
