BEZNext
Cloud Optimization Delivered

Organizations planning to move Data Warehouses and Data Lakes to the cloud have many cloud data platform choices. Most of the platforms are capable of delivering required performance, but at a wide range of costs. Organizations have limited budgets and selecting the cloud data platform that can meet the business performance goals at the lowest cost is a big challenge.

Different lines of business may be best served by different cloud providers and different cloud data platforms. There are many performance and financial uncertainties moving to the cloud.

BEZNext’s approach to this challenge begins with an assessment of performance, resources and data usage to characterize the customer’s business workloads on-premises and then on each of their cloud platforms. For new applications, BEZNext collects and analyzes data during the DevOps process. Their proprietary modeling and optimization technology enables them to predict the minimum configuration and budget required to consistently meet the performance goals for each business workload on each platform. These predictions consider the variability of resource demands at different hours of the day, expected increases in the number of users and data volume, as well as the impact of new applications to be deployed.

“We apply our technology to optimize decisions during DevOps, prior to deployment and optimize FinOps decisions for production Hybrid Multi-Cloud environment.”

In an interview with CIO Applications, Boris Zibitsker, CEO of BEZNext, shares his insights on the evolving cloud technology landscape, the pain points that enterprises face today, and BEZNext value-proposition.

Could you give us a quick overview of BEZNext?

BEZNext helps its clients select the appropriate cloud data platform for on-prem Data Warehouse and Data Lake workloads, and cost effectively manage Hybrid Multi-Cloud environments.

During the DevOps process, we help customers identify performance anomalies and their root causes and assist with cloud selection before deployment of new applications.

After moving workloads to the cloud, we offer solutions to optimize FinOps decisions in complex Hybrid Multi–Cloud environments.

What are some of the challenges that enterprises face with respect to selection of the appropriate cloud data platform and managing complex Multi- Cloud environments?

There are so many cloud data platforms. Our modeling technology predicts the scalability, minimum configuration required to meet business performance goals and cost / performance for each cloud data platform.

Many customers do not take into consideration the dynamics and workloads’ performance, resource utilization and data usage profiles changes.

The workload and volume of data change, deployment of new applications and changing pattern of workload seasonality causes major upsets in performance and costs.

One must consider differences in DBMS optimizers, use of indexes, materialized views, workload management affecting the scalability and cost of running workloads on various platforms.

And how is it that you go about addressing these challenges?

BEZNext’s looks deeply at the customers’ current environments combining on-prem and multi-cloud configurations. We automate continuous data collection, aggregation and building hourly workload profiles, characterizing performance, resource and data usage by each business workload on-prem and each cloud data platform.

We then apply our proprietary modeling and optimization technology to choose the minimum configuration, the best workload management parameters and predict the budget required to meet the performance goals for each business workload on each cloud data platform.

These predictions consider the variability of resource demands during the day, expected increases in the number of users and growing data volume, as well as the impact of new applications to be deployed over the next 12 months.

For new applications, we collect and analyze data during the DevOps process.

At each step of the process, the difference between the predicted performance of the modeled configuration and the SLGs determine the size and direction of the next configuration change. Steps are repeated until the predicted performance is better than the SLG.

Our modeling results answer several critically important questions:

• What is the minimum cloud configuration required to provide performance comparable to that of the original, on-premises production workloads?

• What is the initial cost of the cloud configuration, and what will it cost in the future to support workload growth in both the number of requests and the volume of data?

• How will new applications going through the DevOps process perform on different cloud platforms?

• What will it cost to support a new application after deployment, and what will it cost to support an expected increase in the number of users and volume of data on different cloud platforms?

For hybrid multi-cloud environment, our software answers the following questions and generates recommendations, including:

• When and how much should resources adjusted?

• When and how to should workload management parameters be reset?

• Where to best deploy new applications?

• What are the differences between predicted and actual costs and performance for each workload on different platforms?

• What causes the differences?

• How to implement continuous dynamic capacity management to meet SLGs at lowest cost?

We developed this approach to help IT organizations select the appropriate cloud data platform and cost effectively manage complex hybrid multi-cloud environments.

Who are your typical customers? Can you share with us a customer success story?

BEZNext has successfully worked with larger customers across many industries, including Financial Services, Retail, Power and Energy. Any organization planning to move Data Warehouse and Data

Lake workloads to the cloud and interested in reducing risk of performance and financial surprises can take advantage of our offering.

Many customers concerned with the selection of the appropriate cloud data platform typically conduct industry standard TPC type benchmark tests or organize Proof of Concepts tests with their own representative queries and data. After major efforts and 6 – 9 months they realize their tests cannot determine the minimum configuration and cost required to meet their SLGs. Using our software and services they get the results in 2 – 3 weeks.

Our goal is to provide a plan for cloud deployments that aligns the performance necessary to meet stringent business requirements with the associated costs of running that cloud configuration. BEZNext takes into consideration different pricing models and changes in daily and monthly seasonal activity of the workloads. The planning process also needs to factor in growth expectations for both the amount of data being managed and the number of users accessing that data. New applications might also affect operations.

Modeling and optimization results are used to compare the cost/performance of Vantage, Snowflake, Redshift, BigQuery, Oracle Autonomous Data Warehouse, SAP Hana Cloud and other cloud data platforms running in AWS, Azure and other Cloud environments.

The BEZNext modeling and optimization technology can provide comprehensive recommendations for Cloud Selection within 2-3 weeks versus 6-9 months typically required to conduct customized POC tests.

For organizations with Hybrid Multi-Cloud platforms, we also offer Dynamic Capacity Management solutions to optimize resource allocation and workload management to continuously meet SLGs for each workload with the lowest cost.

What differentiates BEZNext from its competitors?

We are an independent software and professional services vendor developing recommendations optimizing DevOps and FinOps decisions.

Our recommendations provide performance and cost expectations. After the implementation we automatically compare actual performance and financial results with expected. If the results are significantly different, we determine the root causes of that and recommend the corrective actions. It enables organizing a continuous dynamic capacity management of Hybrid Multi-Class environment.

Optimization of DevOps and FinOps decisions during cloud data platform selection and management of the Hybrid Multi- Cloud platform with setting realistic expectation and automatic results verification is setting us apart from competitors.

What does the roadmap look like for BEZNext?

We are working toward automation of modeling and optimization of the DevOps and FinOps decisions. We are also working to automate dynamic capacity management for Hybrid Multi-Cloud environments based on closed loop feedback control.