Chinese American distributed OLAP startup Kyligence has accelerated its core engine, added real-time analytics, and put storage tiering in place to support more and bigger datasets.
The Apache Kylin open-source project was based on an internal eBay development — in Shanghai of all places — to speed online analytical processing (OLAP) and has helped to displace Hadoop. Kyligence was started up in 2016 by people who worked on the eBay Kylin project, and its cloud-native Kyligence Cloud software works on petabyte-scale data warehouses and lakes as well as cloud storage in AWS, Azure and GCP. It supports streaming data with Apache Kafka and business intelligence indexing with Elastic Search.
Luke Han, Kyligence co-founder and CEO, provided an announcement quote: “With Kyligence Cloud 4.5, we have expanded our scope to include faster ad hoc and streaming analytics. This will enable data teams to provide optimal performance across three distinct analytics approaches with a single product. All this is accomplished using the same AI-augmented engine and governed by the same semantic layer.”
Kyligence Cloud v4.5 adds the Clickhouse database to accelerate the OLAP engine, introduces tiered storage and provides so-called Real-Time to run analytics in real-time as well as the existing historical batch data support.
Tiered storage provides a cloud-native way to deploy, maintain and accelerate different types of queries using a single, multimodal AI-augmented analytics engine. It negates the need to deploy different analytics engines in different customer data pipelines. Clickhouse is a high-performance, open source OLAP database and its contents can be processed using the Kyligence engine.
The existing Kyligence Unified Semantic Layer provides consistent definitions and views of data across Kyligence data sources. It enables the ability to conduct analysis using unified key performance indicators, across an enterprise. The software has been extended in version 4.5 to support Microsoft’s Power BI business intelligence tool.
The multi-modal analytics, non-ETL, multi-data source area is a hotbed of development — with ChaosSearch, Dremio, Firebolt and Starburst also active in the space.
Kyligence was founded by CEO Luke Han and CTO Yang Li, both based in Shanghai, and it has an office in San Jose. In fact it says it is dual-headquartered. There are subsidiary offices in Seattle and New York.
There have been four funding rounds following a 2016 seed funding infusion — $8 million A-round in 2017, $15 million B-round in 2018, $25 million C-round in 2019, and then a $70 million D-round earlier this year.
That’s a total of $118 million in funding in five years — quite a clip. The investors include Redpoint Ventures, Cisco, China Broadband Capital, Shunwei Capital, and Eight Roads Ventures (the proprietary investment arm of Fidelity International Limited). Kyligence serves a global customer base that includes McDonald’s, Xactly, China Merchants Bank, and Huawei.
You can get a Kyligence Cloud product datasheet here.
Kyligence Cloud 4.5 will be generally available at the end of August 2021 on Azure and AWS, with support for Google Cloud Platform to follow.