Hadoop Salesforce

Salesforce Wave Analytics Cloud Platform Providing More Options for the Hadoop Cluster Users

Salesforce customers who are using their Wave Analytics platform now have a seamless way to gain access to data stored with on cloud-based bid data warehouses or in on-premise Hadoop clusters. The launch of connectors in partnership with Google, Cloudera, New Relic and Hortonworks among others has enabled faster big data integration and support.

The Wave Analytics Cloud was launched in October 2014, designed as a cloud-hosted analytics platform to provide the non tech-savvy users using Salesforce sales, marketing and service applications to gain access to the big data resources that would improve customer satisfaction and retention, thereby boosting sales.

A Few Existing Gaps

This system, based on a NoSQL platform gained positive acclamation, particularly because it enables better self-service, and because of its schema-less approach to data storage. While Salesforce invested years in to the research and development of the Wave platform, there still exists some gaps – this is no surprise, since first generation products never meet all user requirements.

For instance, Wave had an inherent gap in the way external data could be introduced to the platform for analysis, particularly from Google’s Bigtable and Hadoop. Even though the tag at the time of release was that Wave is a ‘self-service’ platform, data uploads to Wave from external repositories like Hadoop necessitated the use of highly complex ETL processes.

It was this functionality hitch that was addressed with the launch of connectors towards the end of May. Even though on-premise Hadoop big data must first be uploaded to the cloud before being accessible to Wave platform users, partnering vendors have created automated procedures to enable extraction of data and subsequent loading into Wave. However, some data may require transformation prior to extraction and loading to make it useful for Wave users’ purposes.

Closing The Gap Between BI Users And Hadoop

The message from Salesforce is clear: the company delivered “native” applications by using a Java-based program which is installed onto the Hadoop clusters to push data to the cloud. The idea is to create a connection for ‘the last mile’.

Many vendors today are providing powerful big data platforms to be utilized by IT, data scientists, developers and other professionals. However, there wasn’t a way to connect them to the final user of the data – the personnel in sales, service and marketing, who are the real consumers of Salesforce and Wave technology. That was the challenge.

With this new scheme, everybody in the enterprise can gain the value that big data stored in Hadoop clusters provides to make interactions with customers more meaningful, driving sales and profitability higher.

Salesforce developed the Java-based program in conjunction with its partners to enable consumers tap into that data they have stored in Hadoop clusters using Hortonworks or Cloudera. Google’s partnership is intended for an “expanded product integration” according to the Salesforce’sBusinessObjects expert, Keith Bigelow. In the meantime, the basic integration package is available for consumers using Google Bigtable and other related big data products.