IBM makes mondo multi-cloud management move

IBM is wasting no time putting to use the assets it acquired from network automation company NS1 earlier this year. NS1’s technology is set to be incorporated into a new IBM Hybrid Cloud Mesh product designed to make the lives of CloudOps teams a little easier. The key ingredient? Paprika. No, wait, That’s wrong. Automation! The key ingredient is automation.  

Andrew Coward, GM of software defined networking at IBM, who is rapidly acquiring a reputation in the industry as the baron of automation, told SilverLinings that application deployments have hitherto followed a sequential path: first the DevOps team writes the app, then it hands it over to the CloudOps team to put the necessary network puzzle pieces into place. But because every cloud is managed differently, it can take months for a CloudOps team to manually sequence everything properly to ensure an app connects appropriately to each part of the company’s network. 

“No application is an island, right? So, every application you deploy today must talk to other things,” Coward explained. “So, if you move an application that's on premise and you put it into AWS, well, it's likely that a database hasn't moved with it. And you're going to need to talk to that database that's left on the private cloud.” 

IBM's goal with its new Hybrid Cloud Mesh product is automate the process of managing application connectivity across multi-cloud environments. It claims to accomplish this by allowing CloudOps teams to go into the cloud ahead of time, while the DevOps team is still cooking up the app and set the rules for all of their network infrastructure. They can then model the performance of the app and make changes accordingly.  

“So, what is your company allowed to do? What's allowed to transit typically between AWS and Azure or IBM Cloud and private cloud and kind of create these rule sets, and then go roll out a virtual infrastructure essentially inside the cloud.” 

Coward said this will drastically slash the CloudOps task timeline from months down to “a few hours.” And reduce friction between DevOps trams and their colleagues in CloudOps, who typically end up being blamed for app rollout delays and performance issues.  

Coward says that the new product’s modelling and “what if” capabilities also deliver a significant advantage within telco cloud operations, by helping service providers move away from best effort cloud service and closer to the deterministic performance available from global MPLS wide area networks.  

Automation is the focus of an astonishing level of activity in the cloud infrastructure industry right now, with every major player working on new capabilities, especially in the area of AI. We asked Coward what sets IBM apart: “What IBM does really well is work out how people do their job and then automate a business process around that. We start with the job, and architect around that.”    

Hybrid Cloud Mesh is initially being offered as an early access program. IBM said commercial availability will follow in the second half of the year.


Want to learn more about data strategies for the cloud? Register for our Cloud Data Center Strategies virtual event here.