Virtualization Technology News and Information
VMblog's Expert Interviews: Stratoscale Talks Lessons Learned from Fortune 1000 CIOs on Accelerating the Move to Hybrid First

As things shift towards Edge Computing and Hybrid-First IT, and as these things continue to gain traction, I decided to reach out to one of the industry's experts, Ariel Maislos, CEO of Stratoscale, to learn more and find out what Fortune 1000 CIOs are doing to address the many challenges and business demands being faced in this modern era of computing. 

VMblog:  Edge computing is gaining significant buzz in the last few months.  What have you heard from CIOs on their view of Edge Computing and the business value it brings to their organizations?

Ariel Maislos:  IoT and AI are definitely at the center of attention for many CIOs we speak with. It's a major business objective and CIOs are looking for effective technologies to support this growing business demand. The obvious challenges that arise are storage, data analytics implementations, performance and cost. In some recent conversations, however, we've started witnessing a different flavor in the conversation. Data gravity challenges of the public cloud, as well as networking constraints that put a significant cap on the ability to transport data to off-prem environments, drive CIOs to consider edge solutions that will enable them to run cloud-native applications on-prem and will offer equivalent agility and simplicity advantages of the public cloud. Essentially CIOs are starting to evaluate options that will enable them to process data effectively and quickly at each site, as close as possible to where the data is generated. Once processed, the data can than be moved to the public cloud for further analysis and archiving.

VMblog:  The transition towards a hybrid strategy often requires a deep transition in IT skills and DNA.  From your discussion with CIOs from large enterprises, to what extent are they experiencing these challenges and what is their strategy to mitigate these challenges?

Maislos:  We recently had a conversation with a CIO of one of the largest US media companies, who very nicely articulated the "DNA challenge": In the days of the traditional data center, the bulk of IT teams were comprised of IT professionals (virtualization, networking, storage, etc.) and only a small percentage (if any at all) was allocated to cloud architects and DevOps teams. But, as enterprises shift more workloads to the public cloud, the talent composition needs to drastically change and the percentage of DevOps and architects required to drive cloud initiatives and operations needs to grow significantly. A true hybrid setting means that the local data center also needs to shift to a cloud-like environment, meaning that IT staff that continue to run the data center will also need to acquire quite a few new skills such as automation, scripting and orchestrated service delivery. The focus on mitigating the challenge is not so much on replacing the staff, but rather offering effective re-tooling and training to help them catch up with the cloud era and prepare the overall IT organization for a deep cloud transformation.

VMblog:  How has the shift to cloud-native applications impacted the nature of IT tasks?  What are the most frequent challenges IT is facing to answer business demand?

Maislos:  If I had to sum it all up in one word, it would probably be "Now". The cloud era introduced agility, speed and simplicity that developers embraced very quickly. Instead of building huge applications, which would take years to develop, test and deploy, cloud services and new software development methodologies shortened the process to months and many times even weeks. Business executives no longer accept lengthy IT projects. They want everything now. For IT teams, the implications are quite deep. They are expected to rapidly adopt agile capabilities and offer more and more services that would support the development process and deployment. These services include a wide variety of managed databases, which offer automatic provisioning and scaling, managed big data (such as Hadoop and Spark), container orchestration, load balancing, serverless, etc. In addition to these services, the IT framework is also expected to change. From a ticket / event based mode of work, developers expect a self-service platform, SDKs and APIs that would enable them to consume services directly and on-demand, removing IT as middlemen.

VMblog:  Expanding on the previous question, to what extent are CIOs challenged with various siloed environments?  What are their strategies around this issue?

Maislos:  As CIOs are now expected to support business initiatives which are primarily cloud-driven, they are challenged with transforming their local infrastructure so that it resembles, as much as possible, the public cloud environments developers have grown to appreciate. The current situation, where legacy applications run in traditional virtualized environments and cloud-native applications run off-prem, results in a huge discrepancy between environments. These vastly different and siloed types of  infrastructure pose significant management and monitoring challenges for IT organizations, and prevent the flexibility of running workloads across different types of infrastructure. The ability to move workloads across infrastructure has been proven to be one of the main drivers for adopting hybrid strategies, according to a recent survey we conducted among hundreds of IT executives and professionals in mid-2017. Depending on their characteristics, requirements and stage, applications may need to run on-prem or off-prem, and this requirement tends to change a few times during the application lifecycle. Today this is a major hurdle, which not only has cost implications, but also other critical implications that affect the business.

VMblog:  Now that the public cloud has been around for quite some time, what are the common challenges in running and managing public cloud workloads over time?

Maislos:  We can identify three areas of concerns: data, cost, and control. Data and cost are often tied closely together due to the data gravity effect. Let's start with data. In many enterprises, data is perceived as a major business driver. AI and IoT are just two examples. Often, enterprises choose to store data close to the application, which means cloud-native applications may require vast amounts of storage, which tends to be quite expensive. This is essentially the definition of data gravity. But it's not just the cost of storing the data off-prem it's also the cost of moving the data and the growing concerns of cloud lock-in, once the data is in the cloud. But cost has another aspect - predictability. From our discussions with CIOs, when they examine their cloud costs over time they not only see a significant rise, which could be expected given the growth of the business, but it's often unpredictable. Though the on-demand consumption model and the ease-of-use of the cloud bring unprecedented benefits, controlling the costs over time tends to be a difficult task for CIOs and CFOs. The ability of developers to open a cloud account and start working independently of IT, may also create management, control and policy enforcement challenges for the IT organization.

VMblog:  How does Stratoscale help enterprises mitigate these challenges, as they accelerate their transition towards a hybrid-first strategy?

Maislos:  The idea behind Stratoscale is to deliver cloud services on-prem, on top of any infrastructure, which enables enterprises to unify their private and public environments and run cloud-native applications on-prem. There are many opinions and views as to the value and implementation of hybrid. We, at Stratoscale, strongly believe that the true value is flexibility and simplicity. The public cloud introduced game changing benefits that completely changed the ways applications are developed and deployed. The bottom line is ease-of-use, automation and scaling. Stratoscale's hybrid solution delivers these benefits to the on-prem environment by offering self-service and multi-tenant cloud-services including managed SQL databases (MySQL, PostgreSQL) and NoSQL (Cassandra), managed big data (Hadoop, Spark), managed container orchestration (Kubernetes) and many more. Stratoscale currently supports AWS APIs, enabling developers to work seamlessly with both environments. We plan to add support for Google Cloud APIs and Azure APIs in the near future, as well as continue to add more popular cloud services to boost the on-prem environment.


Published Thursday, September 28, 2017 7:33 AM by David Marshall
Filed under: ,
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<September 2017>