Sudheesh Narayanan, CEO“Most CIOs understand the value of big data analytics and Hadoop, and that is why they invest substantially in building Enterprise Data Lakes (EDL) for big data management,” points out Sudheesh Narayanan, CEO, Knowledge Lens. However, incorporating data lake management best practices to integrate, organize, govern and secure huge volumes of data continues to be an area of concern for enterprises. “A critical challenge entails, as high levels of optimizations and governance are required to manage data lake infrastructures and there is a shortage of associated skill-set to do that,” adds Sudheesh. Additionally, a long-standing lack of capabilities with respect to data catalog management, data exploration, data quality management, and data visualization hinders an enterprise’s data lake initiative.
Addressing the challenges associated with deployment of big data lakes and self-service exploration and analytics on Hadoop, Knowledge Lens provides consultation services for big data lake implementations along with a comprehensive line of products branded as the “Lens Family.” Offering an all-inclusive one-stop-shop for next generation data lake implementation, the firm’s BigLens platform empowers data lakes with machine learning and artificial intelligence. The firm’s BigLens product shortens the implementation timeline to less than two months. “This is a big boost for companies that are venturing into building big data lakes, as they can leverage our expertise to reduce their time-to-market,” emphasizes Sudheesh.
Post-deployment, Knowledge Lens helps its customers handle data lake management tasks that occur in a typical production implementation of Hadoop. Dubbed Migration Lens (MLens), the platform assists enterprises in handling backup, restoration, recovery, archival, compression, and migration. Also featuring a unique in-memory compression capability, MLens facilitates high-speed information transfer to allow petabytes of data movement at lightning speeds. “Usually in a Hadoop ingestion process from multiple servers, there is a staging layer where data is first landed before being pushed to a Hadoop cluster,” states Sudheesh. This is primarily due to the fact that there is no direct connection allowed from the source machine to the distributed nodes on the Hadoop cluster in a secured cluster.
Even in situations where the cluster is fully secured through Kerberos, MLens can directly transfer data into a Hadoop cluster
Despite lack of direct connectivity between the source machines and Hadoop nodes, MLens provides a direct push to clusters, eliminating intermediate landing. “Even in situations where the cluster is fully secured through Kerberos, MLens can directly transfer data into a Hadoop cluster,” adds Sudheesh.
In addition to its consultation, Knowledge Lens provides specialized Hadoop administration service, where a team of certified Hadoop administrators handles the end-to-end management of customers’ big data platforms. To top it off, Knowledge Lens’ industry-specific analytics services span text analytics, machine learning, artificial intelligence, and image processing.
A key aspect of Knowledge Lens’ work culture is the spirit of camaraderie within the organization. Sudheesh says, “Every employee of Knowledge Lens is part of one big family with clearly-defined roles and we stand by each other to share the common dream of performing magic with data!” Knowledge Lens encourages an open R&D culture within the organization. The firm has a dedicated big data and IoT innovation lab, equipped with the latest software and technologies to enable their employees’ quest for building smart and productive solutions.
Moving ahead, Knowledge Lens is all set to launch its new product—PredictiveLens—to exclusively support artificial intelligence and machine learning and to revolutionize data analytics over IoT data in the big data lake domain. Apart from this, “We will be expanding our geographical presence across the U.S. and Europe to enable data-driven decisions through our ‘family of Lens’,” informs Sudheesh.