Salesforce this week launched a Zero Copy Partner Network initiative through which it will apply data virtualization to eliminate the need to copy data into the Salesforce Data Cloud.
Announced at a World Tour Event in New York, Salesforce already provides data virtualization capabilities spanning multiple platforms from Amazon Web Services (AWS), Databricks, Google Cloud and Snowflake. The Zero Copy Partner Network adds Microsoft to that list of Salesforce data virtualization partners along with Accenture, Cognizant, Deloitte Digital and PwC to provide data management consulting services.
Patrick Stokes, executive vice president of product and industries marketing for Salesforce, told conference attendees organizations will soon be able to make data accessible to applications using Salesforce Data Cloud, a repository for centralizing storing customer data, without having to copy that data from the location where it is stored. “You can now virtually mount data in DataCloud” says Stokes.
That capability then makes it possible for Salesforce to extend its metadata framework to create workflows that are automatically triggered as data is updated in real time. Salesforce this week also pledged to extend zero copy data access to any open data lake or lake houses based on the open source Apache Iceberg framework. The Heroku arm of Heroku of Salesforce also committed to making it possible to use Zero Copy tool with relational data storied in an open source PostgresQL database.
Data virtualization is hardly a new concept, but as organizations look to drive digital business transformation initiatives at scale it’s not practical to constantly copy data into a central data warehouse. However, by working with other providers of data warehouses and data lakes Salesforce is laying a foundation for removing much of the friction historically associated with moving data between platforms.
That approach also minimizes data governance and compliance issues, because data continues to reside on the platform where it was originally created.
In addition, a prompt sent via the Einstein Trust Layer that Salesforce has developed for accessing multiple large language models (LLMs) can now include virtual data residing on an external platform.
It’s not clear to what degree the members of the Zero Copy Partner Network might eliminate the need to extract, transform and load (ETL) data, but today the demand for data engineering expertise far exceeds the available supply. As such, many organizations are turning to data virtualization techniques to make data accessible to a wide array of distributed applications.
One of the most challenging issues any Digital CxO encounters is making sure the organizations is making decisions based on the most relevant and current data available. A recent Salesforce survey, however, finds 80% of respondents cite data silos are a concern, with 62% noting their organization isn’t yet equipped to harmonize data systems to fully leverage artificial intelligence (AI). Only an estimated average of 28% of apps are connected, and 95% of respondents admit integration issues are impeding AI adoption.
It may be a while before every application in an enterprise is integrated, but as the walls between those applications continue to come down, the faster it will get to realize digital business transformation ambitions.
