Cloud data platforms spark partners' consulting business

Migrating to cloud data platforms

There's no single answer to the backlog question, which calls for a combination of technology, tools, processes and people. But the current focus on cloud data platforms highlights customer interest in the technology aspect. In recent quarters, more than 250 customers have asked Hashmap for briefings on its Cloud Data Platform Benchmark Analysis, which evaluated Amazon Redshift, Azure Synapse Analytics, Databricks, Google Big Query and Snowflake.

"We find that organizations really want to understand their options," Kohlleffel said.

Kelly Kohlleffel

The shift from on-premises data warehouses and other data repositories to cloud-based offerings has been rapid, he noted. Hashmap launched in 2012, specializing in on-premises Hadoop deployments. By late 2017, 70% of the company's business had moved to cloud. Today, fewer than 5% of Hashmap's customers run on-premises data platforms.

"There's a great migration taking place right now," Kohlleffel said. "Everyone is trying to get off traditional appliances and traditional architectures or Hadoop environments. They're too costly, too hard to upgrade or maintain."

Ansley Galjour, consulting principal at Pariveda, a consulting firm based in Dallas, also cited the rise of cloud in data analytics infrastructure.

"We advocate the cloud to our clients in nearly all scenarios," she said. "It's 2022, and no one wants to be in the business of managing infrastructure unless that is your business."

Galjour advised clients to take full advantage of using the "undifferentiated heavy lifting" that major cloud providers offer.

Ansley Galjour

Pariveda works with customers to create strategies and roadmaps for moving to the cloud -- or optimizing on the cloud if they are already there. Clients that migrate to the cloud for data and analytics can free up technical staff for more creative, value-added activities versus performing maintenance on in-house infrastructure, Galjour said.

"For most organizations, cloud computing is the inevitable future for data analytics," said Jacob Samuel, vice president, Digital and Analytics Practice at Trianz, a digital transformation consulting firm based in Santa Clara, Calif. "That is our main focus and what the majority of companies are interested in."

Trianz works with businesses to select a wide array of analytics platforms that support everything from data management and visualization to managed services. The company aims to deliver analytics with the fastest possible time to market, Samuel said.

To that end, Trianz invests in reusable frameworks, accelerators and software intellectual property, which speed up projects. The company develops those offerings on its own or jointly with partners. One example: Trianz created Athena Rapid Analytics, a tool built on Amazon Athena. Athena Rapid Analytics lets users query multiple cloud or enterprise data sources without the need for complex extract, transform, load (ETL) tools, according to Samuel.

Despite the cloud's pull, some customers still use on-site technology.

"We have many clients seeking to obtain more advanced analytics capabilities from their on-premises infrastructure," Samuel said.

The tip of the infrastructure iceberg

Even though cloud data platforms are in high demand, they represent only a fraction of the service provider opportunity.

"Those platforms are really one-eighth to one-tenth of the equation," Kohlleffel said.

After the cloud data platform decision, customers must figure out how to transform, integrate, curate and enrich data -- and then deliver useful products using that data. "There are more technology decisions. It doesn't stop with just the cloud data platform," Kohlleffel said.

Pariveda, meanwhile, cited demand for end-to-end data offerings, under the umbrella of what the company calls the "modern data enterprise." That concept consists of three elements: a foundational platform, a value pillar that seeks to balance iterative design with long-term data strategy, and a governance component that focuses on security and compliance.

"We've seen success in moving from an ideation phase for data products -- like AI and machine learning models -- to architecting and building the data pipelines and platform to support those use cases," Galjour said. The ultimate end is creating self-service capabilities, she added.

Enterprises seek self-service analytics to let business users query data and create reports without relying on the IT department. To make that happen, those organizations must first break down information silos and make data readily available to consumers.

That means organizations must use cloud-based ETL for data integration and conversion as well as reporting tools to visualize data across the enterprise, Samuel said. They must also consider data governance frameworks that offer self-service capabilities. Data governance helps assure data accuracy and regulatory compliance when users take on more analytics tasks.

Aruna Mathuranayagam

Aruna Mathuranayagam, vice president and operating unit CTO at Leidos, a technology and engineering services provider based in Reston, Va., also pointed to a range of data platforms and tools customers can use to support analytics and AI. Mature organizations deploy data lakes and enterprise data warehouses, she said. Customers also tap the major cloud providers and their associated marketplaces, which provide numerous analytics and ML services. Homegrown and open source tools are also part of the landscape.

"It's a mixture of tools you find," she said.

The variety of offerings supporting analytics, AI and ML seems likely to keep service providers in demand -- and on their toes.

"We are working with new technologies, new use cases and new workloads," Kohlleffel said. "Literally, on an hourly basis, there's something new you are being challenged with."