In conversation with Amit Zavery: Oracle makes the case for its cloud

Amit Zavery - Oracle’s senior VP of Integration Products in theCUBE

Oracle Corp. barely shows up on most lists of the biggest cloud computing providers. Amazon Web Services and Microsoft Azure dominate, and even Google Inc.’s and IBM Corp.’s clouds far behind the leaders are ahead of Oracle’s.

Oracle is hardly alone in missing the rapid rise of the cloud led by Amazon.com Inc., whose online retail roots only served to blind the traditional computer industry to the threat to their businesses. But as a company with a broad portfolio of corporate applications running in traditional data centers, Oracle has as much to lose as anyone if it can’t become a force in the cloud.

Now, the company has big designs on remedying its late start. On Wednesday in San Francisco, the company took a critical step when it kicked off the first of a 20-city worldwide series of events in coming months intended to attract more cloud software developers to its platform. Any platform, cloud or otherwise, depends on developers to support it and create a self-sustaining ecosystem of programmers, partners and technology suppliers.

In an interview with SiliconANGLE during the event, Amit Zavery (pictured), senior vice president for cloud platform and integration products at Oracle, explained how the company aims to steal a march on the cloud leaders with a different approach focused more on accommodating existing applications and data centers than on a pure public cloud arrangement. Zavery also provided a vision of how cloud computing is quickly evolving. Not least, he addressed the implications of Amazon Web Services’ outage this week. This is an edited version of the conversation.

Q: What’s Oracle’s value proposition for cloud-native developers? Why would they come to Oracle for this kind of help rather than to AWS or Azure?

A: We’ve been in the forefront of the developer community from the day we built a database. We evolved that to all the middleware stuff we were doing, with thousands of developers building applications on Java. So we’ve had a background in the core development community for many years.

The next evolution, no doubt, is how do you provide those kinds of services in the cloud as well? We have a whole portfolio of SaaS [Software as a Service] applications, so we know how to build modern, cloud-based applications. And the experience we have and the work we’ve been doing allows us to help other developers learn from that experience. So that’s our value proposition.

The right stuff

Q: Still, Oracle might not be seen as cool among the new cloud kids. Does that matter?

A: ‘Cool’ is a fashion. Where they need a service and a technology to help them run the business effectively 24 by 7, scalable and secure and enterprise, I think it’s a matter of getting the right stuff rather than the right look. Coolness will be there if everybody likes what we provide.

Q: Do you need to build a new distinct ecosystem of cloud developers, or can you build on what you already have?

A: We have a large ecosystem, but we want to extend it and include things that are the latest and the greatest. For example, with the latest cloud-based developers, we want to make sure we are connected and working with them on a regular basis so we stay ahead of the trend and we understand what the needs are and work with them collectively to provide those kinds of services.

Q: Do you have to make the case to move workloads to the cloud or are people pulling you there?

A: We don’t need to make the case. The case to be made is more what business practices this company is allowed to do [in the cloud], [depending on] regulatory issues, compliance requirements and data residency needs. They still want the benefit of the cloud, but maybe they don’t want to do it in the public cloud.

But in general, everybody has bought into the idea of agility, speed of development, the subscription pricing model, the ability to turn on and turn things when they’re needed or not needed. We’re not seeing customers who say, “I still don’t see why to have a cloud.”

Q: How do you see cloud computing evolving over the next few years?

A: Looking at how you interact with your customers, for example, people evolved from web-based mobile applications to things like bots and chatbots. That’s a pretty big change, and it means your application has to be AI-based. You have to have a technology under the covers that is evolving and learning those user patterns and allowing you to interact and predict queries and how you answer them in an automated way. So AI and chatbots are going to be a big part of cloud computing going forward.

With cloud computing, what you’re getting is the infrastructure as well as scalability which you could not get in a small company. Now you can do those algorithms and run those things very fast. That’s the next way to build what we call intelligent apps.

Close to the edge

Q: Real-time computing and data analysis needs have spurred a lot of talk about how data needs to be worked on at the edge rather than sent back and forth to cloud data centers, which can be expensive and time-consuming. How does Oracle view the “end of cloud computing,” as Peter Levine at Andreessen Horowitz recently put it?

A: We do expect the computing environment not to be as static. With the cloud, where we have data centers spread out globally, we have things such as Cloud at Customer where you have the data center of your choice, including your own, it can hook into the Oracle cloud or any other cloud environment, but it doesn’t have to be running in the Oracle ecosystem instantly.

Q: Will you need to provide even more local processing in remote locations, say, than Cloud at Customer can provide?

A: As long as it’s available close enough to the users, and then you have the ability to get the data moved when it’s needed fast enough, I think we’ll be OK. The big thing is to have data centers geographically distributed with low latencies in between, and you can failover to another zone in the same area. I think we can solve most of the users’ performance requirements.

Q: Speaking of fails, Amazon Web Services had an outage this week in part of its storage service, stemming from a particular set of data centers in northern Virginia. Does that suggest the need to rethink cloud architectures a bit, or at least how the cloud is used?

A: Cloud has become integral to everybody. It does say that if you’re a service provider, you need to have a very high bar. That’s on everybody, including Amazon, Oracle and everybody else, to provide a very reliable service.

Q: Were you surprised something like that happened?

A: Definitely unfortunate.

Q: Do customers who may have assumed they could leave the details to the cloud provider need to exert more control of the details of their cloud use, or are there ways cloud providers or partners need to offer more granular controls to avoid this sort of outage?

A: You need to know which dependencies you have to control and which elements that maybe it’s OK to depend on somebody and if it fails, you know what the impact is. Then you design your architecture based on that.

The next battlegrounds

Q: With a lot of applications depending on a whole lot of other applications, such as Uber using Twilio and other applications and microservices, how much control can you really have?

A: That’s why you have to build the right instrumentation. It can’t be an afterthought.

Q: Oracle is still way down on the list when it comes to cloud market share and mindshare. Does that matter, or are you mostly appealing to existing Oracle customers looking to move to the cloud?

A: The definition of cloud is too narrow. When people talk about cloud, they’re usually talking about infrastructure. Infrastructure today is the easiest thing to do. It’s outsourcing hardware. All it did was replace buying hardware in your own data center and move it to companies like Amazon.

That is not the only thing you should think of as cloud. The second part is the apps — huge cloud-based services, where Oracle is a leader today, such as CRM [customer relationship management], ERP [enterprise resource planning], supply chain management. We compete very aggressively.

And the big battle now is platform as a service … for application development and management. We have a lot of intellectual property understanding as well as capabilities we’ve delivered in the cloud that are much broader than what other vendors provide.

Q: Why has there been so much focus on the infrastructure layer?

A: It’s a proven thing. You have data to talk about, revenues and use cases that are clear. In that one, we are coming from behind, but I think we have done enough now to have a seat at the table in most of the conversations we have today. And we are continuing to invest in and build that infrastructure.

Photo: SiliconANGLE