[ad_1]
A pair years again, we posed the query as as to if SAS prospects would decide up on its transfer to the cloud. On the time, SAS was on the homestretch of rearchitecting Viya to a cloud-native structure, and some months later, introduced that it was planting its stakes with Microsoft Azure. The partnership prolonged to joint expertise improvement, go to market, and gross sales. Viya is clearly SAS’s lifeline to the long run – with the corporate planning on an IPO in a pair years, it’s going to want a development narrative to entice Wall Road.
What’s occurred over the previous 18 months?
SAS Viya on Azure remains to be a fraction of SAS’s core enterprise, however development is beginning to decide up. For calendar This fall 2021, Viya on Azure gross sales have been up 30% over a yr in the past as the corporate closed 20 seven-figure offers, one in every of them topping out at $6 million for a three-year legacy SAS 9.4 migration to Viya. Over the previous yr, the most important was an identical $15 million multi-year migration engagement.
SAS has been constructing integrations with Viya providers to slot in natively with Azure analytics, streaming, and machine studying providers. Clearly, with such an bold sport plan, it is at present a piece in progress. SAS at present provides Viya plus a dozen basic providers on Azure Market, with materials outcomes on Microsoft’s backside line: SAS is now one in every of Azure’s prime fifteen companions primarily based on income.
Let’s drill down on the options and the place SAS is in working Viya into the Azure material. Generally, you may entry SAS fashions and mannequin administration from varied Azure analytics providers and vice versa, though not surprisingly, the diploma of integration at present varies by service.
It begins with Viya integration into core Azure infrastructure, together with Azure Lively Listing for single sign-on; Azure Key Vault for certificates administration; Azure monitor for logging; and Azure Coverage for governance. Integration with Azure Kubernetes Service (AKS) permits Viya providers to run on Azure K8s clusters, in order that, as an illustration, SAS Occasion Stream Processing (ESP) can elastically scale with out having to reinvent the wheel. Different fundamentals embrace the power for builders to model code from SAS Studio in GitHub. And on the database stage, SAS Viya has connectors to all Azure databases; this isn’t distinctive as a result of, like another analytics suite, Viya has connectors to different widespread databases corresponding to Oracle.
Among the best developed integrations is with chatbots; you may construct your chatbot in SAS Dialog Designer and run it in Groups. SAS fashions can now be registered to run in Azure Machine Studying, whereas Azure Synapse Analytics can run SAS scoring fashions in-database. For anybody who follows SAS, there is a small irony right here. Working the fashions in Synapse requires working Spark, a change from the LASR engine that was initially SAS’s most well-liked high-performance analytics compute engine within the days earlier than Viya. You may register fashions developed in Azure Machine Studying (Azure ML) in SAS Mannequin Supervisor for lifecycle administration, and in flip register and deploy SAS fashions as containers in Azure ML.
There are different connections within the works. They embrace working SAS Clever Decisioning (managing selections and monitoring their effectiveness) with Microsoft Energy Apps and Dynamics 365 with connectors that at the moment are usually out there.
In the long term, the payoff of SAS integrations on Azure can be gated by the breadth of the Viya platform. Not surprisingly, the event of Viya has adopted the 80/20 path, with the crux of SAS capabilities in statistics, danger administration, decisioning, and so forth lined. However SAS has a wealthy trove of options in domains corresponding to fraud and anti-money laundering; provide chain; danger administration; optimization, and so forth. There may be additionally SAS’s broad portfolio of vertical options, and on Azure, SAS is specializing in 4 key verticals together with banking, healthcare, manufacturing, and retail.
Why Azure?
SAS’s technique is a chance that departs from the same old script of third-party expertise suppliers, particularly for these whose information and analytics platforms and options compete with counterpart choices from every of the Massive Three.
For example, assume Informatica Clever Information Administration Cloud’s Information Integration Service and Azure Information Manufacturing facility; they each compete, however Informatica’s providing is on the market on Azure Market. Or MongoDB Atlas vs. Amazon DocumentDB, with Atlas out there on AWS market. Coopetition is the rule; third events compete and accomplice with the key clouds, with their core differentiator being that they ship the identical expertise no matter cloud. SAS departs from that script. Positive, you may run Viya anyplace – on-premises or on any cloud K8s cluster. However if you need SAS on SaaS — delivering Viya as a managed cloud service – all roads result in Azure.
So why go all in on Azure? Partly, it is to offer a extremely useful bridge to the cloud for its present put in base. However there’s another excuse, we suspect, for getting shut with Azure:
It is the shortest path to interesting to a brand new era of consumers. It is a era that received impressed over the previous decade by Tom Davenport’s extolling of information science as the sexiest job of the twenty first century.
Over 40 years, SAS could have constructed an analytics portfolio virtually unequalled in depth and breadth, however which means comparatively little to grads popping out of information science college applications accustomed to working with open supply languages, applied sciences, notebooks, and AutoML cloud providers. Although SAS has made peace, and works with all that open supply stuff, that could be a secret to up and coming information scientists that is been stored all too nicely.
With a targeted relationship with Microsoft, SAS may enrich Azure analytics with property corresponding to mannequin life cycle administration; useful depth in disciplines corresponding to decisioning, danger administration, and optimization; and naturally, its deep portfolio of modeling libraries. A few of that is right here as we speak; we wish to see SAS carry extra of the richness and breadth of its capabilities into Viya, and in impact, conceal in plain sight as they function as extensions to Azure Synapse or Machine Studying. We’d proffer that Azure supplies the onramp to a brand new buyer base to make the most of SAS capabilities with out having to make massive enterprise software program commitments.
So, this is our suggestion to SAS: clear up your Viya on Azure branding as a result of it is downright complicated. Proper now, you name it SAS Viya on SAS Cloud, implying that prospects need to run on a separate cloud. Why not borrow a web page from Databricks, which at present enjoys an obscene valuation, and name it one thing like Azure Viya?
[ad_2]
