3 Cloud Trends to Prepare for in 2016
With the rapid changes to SaaS applications and cloud platforms taking place today, the area of cloud integration is now in constant flux. Years ago, cloud integration used to be seen as a tool that accomplished a simple use case, such as replicating SaaS data to an on-premise database for analytics. However, with the innovations taking place in the cloud with regards to analytics, big data, and application development, the very nature of cloud integration is changing. Here are three major changes in cloud for 2016 that will impact the way in which companies will need to think about their cloud integration strategies.
Central IT Takes Over Cloud Analytics Initiatives
Ever since the announcement of salesforce.com’s Wave Analytics in October 2014, with the tagline “Analytics for the Rest of Us”, the cloud analytics race began. In the very same month, Birst announced its partnership with SAP HANA, touting an architecture that provided more instant analytics to the end user. Barely a year later, in October 2015, Amazon Web Services announced its QuickSight cloud BI service, targeted towards data analysts.
A few other occurrences in the cloud analytics world happened in 2015 that are interesting to note:
- It was revealed in January 2015 that a basic Salesforce Wave license costs about $40,000, in addition to other site licenses and per user costs
- In April 2015, Domo announced their latest funding round along with an interesting tidbit –customers would have to contact Domo directly if they wanted any data integrations set up so they could maintain the secrecy of their cloud analytics stack
- A few days later, Tableau and Birst, once bitter rivals, formed a partnership that allowed Birst users to directly connect to Tableau. Birst would provide a central enterprise data repository from which to create new datamarts for business users, while Tableau brought its data discovery and visualization capabilities to the table
- Tata Consulting Services (TCS) and Tableau announced an alliance where TCS would focus on developing “large scale delivery capabilities” for Tableau’s data visualization functionality
- These occurrences prove that although cloud analytics was originally targeted towards the self-service needs of business users within LOBs, many of these deployments are more complex than originally thought. The need to contact vendors directly to set up data integrations, enterprise IT-style pricing, the realization that data visualization only forms one layer of the cloud analytics stack, and the involvement of global System Integrators all points towards signs that Central IT will be the major driver of cloud analytics projects in 2016, rather than the individual LOBs.
Your 2016 Cloud Takeaway
Organizations looking at cloud integration solutions to integrate the variety of data sources required for an enterprise-class initiative should choose a cloud integration platform that has a unified data fabric across data sources, transformations, and integration patterns.
Big Data Processing Moves to the Cloud
With the growth in big data, Spark is replacing MapReduce as the data processing standard. The industry is moving more towards real-time streaming data and experimenting with machine learning use cases. To successfully accomplish Big Data use cases in production at scale, all associated infrastructure for provisioning, deployment, logging, and monitoring, as well as ingestion technologies for streaming data and workflow management tools need to be on a single cloud platform. Cloud leaders such as Amazon Web Services have perfected the art of minimizing latency and optimizing clusters for the processing of large datasets. The recent January 2016 announcement by Chinese e-commerce giant Alibaba launching 20 new online services within its AliCloud offering related to big data only underscores the importance of having a cloud-centric big data strategy. This trend will only accelerate in 2016.
Your 2016 Cloud Takeaway
Using cloud integration as a centerpiece for big data processing in the cloud is a strategy that enterprise architects should consider. Since the future of big data is real-time, organizations should realize that there are different real-time streaming technologies out there, such as Kafka, Storm, and Spark Streaming to name a few. Each streaming technology serves a different use case, and a cloud integration platform needs to connect to a variety of these streaming technologies.
“Hybrid Integration” Gets Redefined Again
The “hybrid integration” term has been one of the most confusing terms out there. In recent years, middleware and integration providers have used the term to mean many things. While it is quite clear that a hybrid integration scenario is one involving cloud and on-premises data sources, the specific approach taken towards solving the challenges involved with integrating these apps will define how successful enterprises are in deriving value out of their investments.
Your 2016 Cloud Takeaway
The ever-changing definition of hybrid integration will continue to evolve due to the explosion of varieties of data within the internet of things (IoT), new open source big data technologies, and the emerging area of microservices. At the end of the day, it is important that companies realize that it is in their best interests that all their apps, databases, and infrastructure move to the cloud. A hybrid integration strategy that supports this end goal will ensure that innovation moves at a much more rapid pace.
Is Cloud Integration Best for Your Organization?