Unstructured Data Management at the Edge: Five Questions VARs Must Ask Customers Now

Channel partners know that edge computing is proliferating across their customer sites. Many end users are small, require high availability and need to run applications locally, which eliminates the cloud as a viable option. Further complicating the global data explosion at the edge, unstructured data presents a variety of challenges pertaining to data management as many organisations lack the tools and infrastructure at these smaller sites.

As edge computing adoption continues to soar, channel partners are tasked with helping their end users determine how best to manage and analyse data affordably. Many customers do not have the time, energy or budget to send all of their data to the cloud for analysis. With the rise in popularity of technologies such as new forms of artificial intelligence, companies are trying to evaluate how to run analytics as close to the data source as possible.

There are five key questions that channel partners should be asking their customers in order to help them define their strategy for managing unstructured data at the edge.

Q1: Do you really need the cloud?

Despite the technology’s notoriety, the cloud is not the right solution for every organisation, for every use case. Cloud services can be significantly more expensive for compute and storage, are prone to internet outages that can cause downtime at edge sites and can’t deliver the same application performance as running onsite hardware and software. Some applications simply cannot wait for the delays associated with transferring data from edge to cloud and back. In contrast, edge solutions solve all of these problems and deliver your customers the ability to make real-time decisions locally.

Data-generating applications are typically creating unstructured data at the edge, often IoT data. For example, retailers may use applications that manage inventory and video surveillance or manufacturing plants can include thousands of sensors that generate data points every 10 seconds, 24/7. There are many more examples, depending on the industry that your customer is in.

Does all of this data need to be stored for just thirty days? 6 months? Will it be needed for analytics today? Next year? This is becoming a huge problem for your customers. The average IT team cannot manage all of this data – much less analyse it all. Channel partners should be armed with the latest options for edge versus cloud strategies.

Q2: What’s your data protection strategy?

Data protection is a key topic for any channel partner and end user. The challenge today is the technology to fully address all of the ins and outs of unstructured data doesn’t exist yet. Customers need guidance on the cost benefits of keeping everything forever versus classifying it for analysis down the road and archiving it in the interim. Variables range widely from company to company depending on retention policies, backup strategies in place and determining whether the cloud will be used for archival, or if the data can simply be tossed out after a certain time.

Hyperconverged infrastructure (HCI) solutions are an excellent way to deploy and manage edge data affordably. HCI is ideal for running edge sites while collecting unstructured data in a cost-effective, easy way. By deploying HCI at the edge, customers will have a solution that includes tools that run as an application on top of hyperconverged infrastructure while ensuring uptime.

Q3: What is your company’s uptime requirement at the edge?

As IT is moving away from the traditional data centre and to the edge, uptime at businesses with dozens, hundreds or sometimes thousands of locations is critical. In some instances, a single server solution may do the trick though there are many ways for a single server solution to have a failure that could lead to a loss of productivity and potentially huge financial impact. Channel partners should investigate exactly how detrimental a downtime event could be as they are helping customers map out their data management strategy. Most HCI solutions offer multi-node options to keep systems running 24x7.

Q4: How will you “right-size” your edge solution without sacrificing performance?

There are tons of out-of-the box IT solutions ready to plug in and go, but many are pre-built with compute, storage and features that are simply overkill for most customers’ edge environments and can lead to over-provisioning – and overspending. Channel partners should analyse the buildable offerings that can provide the customer’s exact needs – including server type and size, CPU, memory, disk configurations and SSD requirements. HCI at the edge is a customizable solution known for high availability and is a superior way to manage unstructured data, particularly for IoT and in remote/branch office sites where a smaller hardware IT footprint is a necessity.

Q5: How will you manage it all?

Now that your customer has addressed data protection, uptime requirements and narrowed down a shortlist of possible solutions, how will they manage the data? Edge infrastructure management is important but can be challenging. Managing an entire environment that may consist of dozens, hundreds or thousands of individual sites is not easy and there are no “out of the box” management solutions that have solved this problem. End users will likely wind up using tools provided by their HCI provider, so these should be evaluated carefully.

One additional complicating factor to management is the server virtualization environment that is chosen. The tools available will be different depending on virtualization approach chosen. While VMware vSphere has the lion’s share of the market today, there are other options that end users are using and considering for the future. Broadcom’s proposed acquisition of VMware is causing many users to re-think their server virtualization strategy because they fear that licensing costs will skyrocket. Microsoft’s Hyper-V and open-source Linux KVM are two options that should be considered, as they can save users significant money around software licensing.

In summary, unstructured data management and protection at the edge involves vast amounts of data that must be highly available and secure. Following this short checklist of questions, channel partners will be armed as advisors to help their end users make sound decisions on how to best deploy and manage all of this data - whether for SMB customers with a small number of sites or enterprise customers with thousands of edge sites.