What’s in this Post:
Introduction
Data is gold. Or perhaps stated more accurately data in context is gold.
12.
If you think about a single piece of data, for example: 12. 12 is just a number. If I say to you “12” you likely wouldn’t understand what I was babbling about. If I add another piece of data and say, “The temperature is 12” that would be a bit more helpful. But it still isn’t enough data to tell you if you need to wear a jacket. If I say the temperature is 12 degrees centigrade, you’ll finally have some actionable information.
Arguably, if you consider the value that an Information Technology function brings to the modern enterprise, it is the production of actionable information. Further, this actionable information is the driver of business growth. Or more directly stated: the business that can collect, process and apply data the most effectively will be the winner in the modern marketplace.
Today’s modern alchemists include members in every role in IT including desktop application support, database architects, network engineers and data scientists. While IT teams have been transforming data into gold for decades, the quantity and quality of that gold hasn’t always been an area of focus. To turn data into gold today, you might consider improving your alchemy talents.
Keep reading for 4 tips for how to collect, process and apply data most effectively.
4 Tips for How to Turn Your Data into Gold
Tip 1: Collect all of the data
The goal to collect ‘all of the data’ is the true driver behind digital transformation and the explosion of the Internet of Things (IoT). And the tools we use to analyze, and process data presume that all of the data is known. Missing data can be critical for filling in the blanks in predictive models, dispelling confirmation bias, and ensuring your team is making the most informed decisions.
However, don’t assume that all of the data will be collected through digitalization or more IoT.
In fact, IT leaders should look beyond these phenomena to examine their own processes more deeply to uncover data sources that the business may not have considered.
Because of the nature of their work, IT personnel develop unique perspectives on business processes that others don’t readily observe. Ranging from trouble ticketing information to network traffic statistics, incremental IT data can provide the missing pieces to flesh out predictive models to help the business better anticipate customer needs or forecast an impending outage.
Tip 2: Break down network data silos
Another way that IT organizations can help turn more data into gold is by enabling data to be easily assembled. Recall the value of data is derived when it is placed in context. However, it’s not unusual for different data sources to be isolated from each other via ‘network data silos’.
Network data silos occur when data is located within different network domains such as public clouds, private clouds or isolated internal networks. Accessing data in these network data silos typically requires dealing with different security policies, overcoming technical hurdles and the configuration of complex interconnections.
This complexity is a double edge sword.
It provides us with the flexibility to deal with varying business requirements; however, it also creates the possibility for human error. One way to address this issue is to seek ways to manage complexity through software. For example, Stateless software can abstract the complexity of your different network domains by turning them all into a simple to configure endpoint in a software platform. You can maintain the control you need while all of the networking functions are simplified.
Tip 3: Move the data without breaking things
Select all, copy. Oops. You just initiated a 30 GB data transfer, consuming all the bandwidth on the connection to a remote server. The server is now so busy that you can’t even push a ‘ C’ through to kill the process.
How many times did that happen to you when you first started out as a new database admin, or as a junior data scientist? Execute this maneuver during working hours and suddenly you’re more famous than Kim Kardashian.
How could you avoid this happening to your infrastructure?
Tools such as QoS queues, MPLS class of services tags or even SD-WAN prioritization still send traffic over the same network. What if instead you could build dedicated segregated networks? One network for production traffic and a separate network for the data scientists?
It is possible to build this type of topology today using additional hardware and of course dealing with the corresponding complexity and effort of creating and managing multiple networks. It’s not surprising that organizations rarely attempt to employ this strategy and instead muddle by trying to employ QoS tagging or something similar.
However, software technology has advanced considerably in the last three years. Now you can segment traffic with software and create all of the underlying network functions to implement the network functions needed to create separate networks to keep the traffic flows completely segregated. The new type of infrastructure is called Priority Workflows.
Tip 4: Let everyone have the data
While there are certain things that we should leave to only the professionals, data analysis isn’t one of them. Practically everyone recognizes the value of data resulting in the emergence of many new advanced data manipulation and integration tools like Immuta, Meroxa, Molecula and DisruptOps. These tools provide ways for data to be securely shared, manipulated, analyzed and monetized anywhere within an organization. Couple these capabilities with Stateless’ software to create data pathways that can allow your organization’s alchemists to create 24 carat gold in quantities that The Federal Reserve Bank of the United States would be jealous of.
Summary
These four tips – collect all the data, put it in context by breaking down the barriers that keep it apart, safely move it to where it’s needed, and finally sharing it so that your entire organization can use it to create value – could certainly elevate the stature of your IT team and perhaps allow you to sleep a little easier. If nothing else, they should help your alchemists to produce enough gold to keep Smaug at bay.
