Unravel Data will introduce new capabilities for operationalizing big data applications in production at scale with the launch of Unravel 4.4.

Unravel 4.4 introduces new innovations and capabilities to simplify operations whilst ensuring big data applications are highly reliable and predictable under the most demanding of eEnterprise big data and analytical workloads.

“We were the first in the category to deliver automated configuration tuning recommendations. Our latest release builds on that, leveraging our Machine Learning engine to extend automated tuning recommendations to an entire cluster for all tenants,” said Kunal Agarwal, CEO, Unravel Data. “Users get an end-to-end, granular view that shows how each application is performing across all infrastructure dependencies. Combined with other new capabilities in this release, we are helping eliminate the painstaking, trial-and-error configuration, troubleshooting and remediation workflow that’s long plagued big data Operations teams and now made it possible to manage large scale deployments in an agile and dependable manner.”

In order to accommodate growing enterprise big data deployments, Unravel 4.4 incorporates several major innovations. These include:

- Automation. Unravel 4.4 features new auto-tuning capabilities and sessions frameworks, enabling users to tune jobs automatically to a desired goal (such as speedup, SLA, and resource efficiency). Unravel leverages artificial intelligence operations (AIOps) optimization techniques to converge operation metrics with recommendations and actions to deliver a desired business outcome.

- Capacity Planning: Big data clusters tend to accumulate lots of data and operators are constantly under pressure to add more capacity to meet demands. Traditional approaches to capacity planning are very labor-intensive and exacerbate the problem through the use of ad-hoc spreadsheet models and reliance on tribal knowledge of cluster administrators. With the addition of capacity planning reporting, Unravel 4.4 provides an intuitive and data-driven approach to visualize and predict the growth of the cluster and make an informed decisions around capacity and resource requirements.

- Cluster Optimization: Unravel has long provided actionable insights and recommendations on a per-application basis. The platform was the first in the industry to provide recommendations for tuning precise configuration parameters. Unravel 4.4 extends this capability to provide configuration tuning recommendation at the entire cluster level. With these global recommendations, cluster administrators can now improve the performance of all jobs and applications running on the cluster.

- NoSQL Support: This release adds support for NoSQL and HBase. Unravel 4.4 provides an APM-centric view for HBase usage and helps identify anomalies and outlier issues (for example, table/region hotspotting) that could adversely affect the application, while providing remediation techniques.

- Small Files Analysis: This release offers a new reporting feature that gives users a comprehensive view of entire directories and a granular look at small files. This feature allows users to reduce resource utilization by these small files, freeing resources for larger workloads.

Unravel 4.4 will be generally available in September 2018.