Kyvos Insights Case Studies Delivering Cost-effective and Scalable BI at a Leading Telecom

Edit This Case Study Record

	Delivering Cost-effective and Scalable BI at a Leading Telecom Kyvos Insights

Delivering Cost-effective and Scalable BI at a Leading Telecom

Kyvos Insights

Technology Category	Analytics & Modeling - Big Data Analytics
Applicable Industries	Telecommunications
Applicable Functions	Business Operation
Services	Data Science Services System Integration
Challenge	The global telecommunication company was facing several challenges with their existing MicroStrategy BI tool. As the volume of their data increased massively, and more dimensions, metrics, and measures got added to their MicroStrategy reports, response times increased from seconds to several minutes, even for smaller and less complex reports. Transferring billions of rows of raw data from the data lake to MicroStrategy for computations caused the application to time out in most cases. Disconnects between the data collected by different departments such as marketing, finance, product development, and others made cross-functional analysis extremely difficult. Building cubes on MicroStrategy consumed 50% of resources, resulting in prohibitive costs. They also faced inconsistent response times on MicroStrategy queries and an inability to meet increasing business requirements as they could not accommodate additional dimensions and measures in reports without impacting performance. Read More
About Customer	The customer is a global telecommunication company that was using MicroStrategy to deliver business insights to its 10K+ employees. The company deals with massive volumes of data, with more and more dimensions, metrics, and measures being added to their MicroStrategy reports. However, as the volume of their data increased, they faced challenges with response times, memory overload issues, and prohibitive costs. The company needed a solution that could handle their growing data volumes and deliver the desired performance. Read More
Solution	The company decided to implement an OLAP solution directly on their big data platform to deliver the desired performance. After evaluating several vendors such as AtScale and open-source platforms like Apache Kylin, they chose Kyvos as their BI acceleration platform. Using its Smart OLAP™ technology, Kyvos helped them build cubes on massive data volumes. Kyvos pre-aggregated and materialized the entire cube structure for speed, just like MicroStrategy cubes but at a much larger scale. Since the cubes were stored on the data lake and not in memory, it was easy to build cubes that were much larger than before. First, they migrated their existing MicroStrategy cubes to Kyvos, especially the ones that took a huge amount of platform memory and resources. Cube replication was easy as Kyvos uses the same principles and best practices to build cubes as MicroStrategy Developer. Then, they built new cubes on unoptimized raw data on the data lake. With Kyvos, it became easy to scale up architecture as they could increase compute power or get faster responses just by adding another node to the cluster. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Big Data Analytics

Applicable Industries

Telecommunications

Applicable Functions

Business Operation

Services

Data Science Services

System Integration

Challenge

The global telecommunication company was facing several challenges with their existing MicroStrategy BI tool. As the volume of their data increased massively, and more dimensions, metrics, and measures got added to their MicroStrategy reports, response times increased from seconds to several minutes, even for smaller and less complex reports. Transferring billions of rows of raw data from the data lake to MicroStrategy for computations caused the application to time out in most cases. Disconnects between the data collected by different departments such as marketing, finance, product development, and others made cross-functional analysis extremely difficult. Building cubes on MicroStrategy consumed 50% of resources, resulting in prohibitive costs. They also faced inconsistent response times on MicroStrategy queries and an inability to meet increasing business requirements as they could not accommodate additional dimensions and measures in reports without impacting performance.

About Customer

The customer is a global telecommunication company that was using MicroStrategy to deliver business insights to its 10K+ employees. The company deals with massive volumes of data, with more and more dimensions, metrics, and measures being added to their MicroStrategy reports. However, as the volume of their data increased, they faced challenges with response times, memory overload issues, and prohibitive costs. The company needed a solution that could handle their growing data volumes and deliver the desired performance.

Solution

The company decided to implement an OLAP solution directly on their big data platform to deliver the desired performance. After evaluating several vendors such as AtScale and open-source platforms like Apache Kylin, they chose Kyvos as their BI acceleration platform. Using its Smart OLAP™ technology, Kyvos helped them build cubes on massive data volumes. Kyvos pre-aggregated and materialized the entire cube structure for speed, just like MicroStrategy cubes but at a much larger scale. Since the cubes were stored on the data lake and not in memory, it was easy to build cubes that were much larger than before. First, they migrated their existing MicroStrategy cubes to Kyvos, especially the ones that took a huge amount of platform memory and resources. Cube replication was easy as Kyvos uses the same principles and best practices to build cubes as MicroStrategy Developer. Then, they built new cubes on unoptimized raw data on the data lake. With Kyvos, it became easy to scale up architecture as they could increase compute power or get faster responses just by adding another node to the cluster.

Impact #1	Query response times ranging from 0.25 to 0.65 seconds on billions of rows
Impact #2	Instant responses, even for thousands of concurrent users
Impact #3	No compromise with performance

Impact #1

Query response times ranging from 0.25 to 0.65 seconds on billions of rows

Impact #2

Instant responses, even for thousands of concurrent users

Impact #3

No compromise with performance

Benefit #1	Incremental cube build time reduced to 2 minutes from to 30 minutes in MicroStrategy

Benefit #1

Incremental cube build time reduced to 2 minutes from to 30 minutes in MicroStrategy

Download PDF Version

Overview

Delivering Cost-effective and Scalable BI at a Leading Telecom

Operational Impact

Quantitative Benefit